explain xkcd - User contributions [en]

Talk:2735: Coordinate Plane Closure

2023-02-09T04:42:49Z

172.69.134.131: revise sillyness

Is there significance to the fact that the axes aren't labeled in the warning? Can I plot y = 0.75x today or not?[[User:Brossa|Brossa]] ([[User talk:Brossa|talk]]) 15:05, 8 February 2023 (UTC)
: you cannot because it intersects the given square as shown in this desmos thing i whipped up in 2 seconds: https://www.desmos.com/calculator/zb9nbrl6s5 [[Special:Contributions/172.70.43.29|172.70.43.29]] 15:38, 8 February 2023 (UTC)Bumpf
::I can if the forbidden coordinates are 1 ≤ x ≤1.5 and 1.5 ≤ y ≤2[[Special:Contributions/172.70.131.66|172.70.131.66]] 15:56, 8 February 2023 (UTC)
::: In the absence of other information, assuming horizontal ''x'' and vertical ''y'' would be conventional. --[[Special:Contributions/141.101.98.145|141.101.98.145]] 19:15, 8 February 2023 (UTC)
"Hole" is also sometimes used to mean a particular coordinate on a function which is discontinous at some point but could have a value (for example sinx/x with a hole at (0,1)). [[Special:Contributions/172.70.206.92|172.70.206.92]] 19:18, 8 February 2023 (UTC)
Randall listed 2 points, yet the cordoned off area is a square. 2 points define a line, not a square, he really should have thought of that. How is someone to know the invalid points without the diagram? Even with the diagram, we don't know whether points on the boundary are included! Is the line y=1 a valid line to draw? THESE ARE QUESTIONS THAT NEED TO BE ANSWERED RANDALL BE MATHEMATICALLY RIGOROUS NEXT TIME.
: One reason could simply be the alignment between the coordinates and time. Reading out the numbers without paying attention to the mathematical punctuation you can form the sentence "the coordinate plane will be closed Thursday between 1:51 and 2:15 to repair a hole", following the typical structure of such a notice to not just provide a day but a time.
Ironically, the notice makes it sound like using y=1 is fine, and the affected region is only strictly greater than y=1. That would make the region that's closed an open set, and the region that's open a closed set. [[Special:Contributions/172.70.110.230|172.70.110.230]] 22:46, 8 February 2023 (UTC)

🚧 DETOUR 🠕 KEEP WITHIN [[2646: Minkowski Space|MINKOWSKI CONES]] ⛔ DO NOT ENTER Y < |X| 🚧 [[Special:Contributions/162.158.90.38|162.158.90.38]] 23:37, 8 February 2023 (UTC)

So the joke is that the coordinate plane is closed when there's damage that causes it not to be closed? [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 23:44, 8 February 2023 (UTC)

Talk:2735: Coordinate Plane Closure

2023-02-08T23:47:15Z

172.69.134.131: 🠕

Is there significance to the fact that the axes aren't labeled in the warning? Can I plot y = 0.75x today or not?[[User:Brossa|Brossa]] ([[User talk:Brossa|talk]]) 15:05, 8 February 2023 (UTC)
: you cannot because it intersects the given square as shown in this desmos thing i whipped up in 2 seconds: https://www.desmos.com/calculator/zb9nbrl6s5 [[Special:Contributions/172.70.43.29|172.70.43.29]] 15:38, 8 February 2023 (UTC)Bumpf
::I can if the forbidden coordinates are 1 ≤ x ≤1.5 and 1.5 ≤ y ≤2[[Special:Contributions/172.70.131.66|172.70.131.66]] 15:56, 8 February 2023 (UTC)
::: In the absence of other information, assuming horizontal ''x'' and vertical ''y'' would be conventional. --[[Special:Contributions/141.101.98.145|141.101.98.145]] 19:15, 8 February 2023 (UTC)
"Hole" is also sometimes used to mean a particular coordinate on a function which is discontinous at some point but could have a value (for example sinx/x with a hole at (0,1)). [[Special:Contributions/172.70.206.92|172.70.206.92]] 19:18, 8 February 2023 (UTC)
Randall listed 2 points, yet the cordoned off area is a square. 2 points define a line, not a square, he really should have thought of that. How is someone to know the invalid points without the diagram? Even with the diagram, we don't know whether points on the boundary are included! Is the line y=1 a valid line to draw? THESE ARE QUESTIONS THAT NEED TO BE ANSWERED RANDALL BE MATHEMATICALLY RIGOROUS NEXT TIME.
: One reason could simply be the alignment between the coordinates and time. Reading out the numbers without paying attention to the mathematical punctuation you can form the sentence "the coordinate plane will be closed Thursday between 1:51 and 2:15 to repair a hole", following the typical structure of such a notice to not just provide a day but a time.
Ironically, the notice makes it sound like using y=1 is fine, and the affected region is only strictly greater than y=1. That would make the region that's closed an open set, and the region that's open a closed set. [[Special:Contributions/172.70.110.230|172.70.110.230]] 22:46, 8 February 2023 (UTC)

🚧 DETOUR 🠕 VIA [[2646: Minkowski Space|MINKOWSKI LIGHT CONE]] ⚠️ AVOID Y < |X| 🚧 [[Special:Contributions/162.158.90.38|162.158.90.38]] 23:37, 8 February 2023 (UTC)

So the joke is that the coordinate plane is closed when there's damage that causes it not to be closed? [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 23:44, 8 February 2023 (UTC)

Talk:2734: Electron Color

2023-02-08T18:47:54Z

172.69.134.131:

Electrons have no color?! BUt lIgHTnIng strIKeS aRe YEllOw, aND LigHTNing IS MaDe uP of eLECTrOns.[[Special:Contributions/172.71.254.115|172.71.254.115]] 22:43, 6 February 2023 (UTC)
: Actually most colors are emitted by electrons orbiting atoms after absorbing light. The color electrons emit depend on their kinetic energy and available places they can travel, a tiny bit similar to how things change color as they get hotter, but more extreme and general. [[Special:Contributions/172.70.114.198|172.70.114.198]]
: I'm pretty sure lighting strikes are white. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 20:58, 7 February 2023 (UTC)

It may refer to the Greek etymology of the word "electron". Originally it meant amber, a yellow gem. [[Special:Contributions/172.68.118.146|172.68.118.146]] 23:20, 6 February 2023 (UTC)
:But amber isn't yellow - it's... amber. [[Special:Contributions/172.70.85.222|172.70.85.222]] 10:40, 8 February 2023 (UTC)

I can't do formatting, I'm new. Sorry! {{unsigned|No Idea If There's A Character Limit LMAO}}

To me, this is 1000% building on the idea of debating the colors of school subjects. I've added a bit of explanation to the text about it. I used my own color associations & reasons (science = green, history = red) as an example, and I'm sure people will disagree with me. Leave your color/subject associations in a reply to this comment, could be a fun little debate! (also, English = blue)
[[User:Zman350x|Zman350x]] ([[User talk:Zman350x|talk]]) 23:50, 6 February 2023 (UTC)
: SocStud is yellow, Math is red, Science is green?, ELA is gray, French is blue, and orange is my least favorite subject out of the rest. I have gotten into many arguments with my friends.
:[[Special:Contributions/172.70.230.157|172.70.230.157]] 00:10, 7 February 2023 (UTC)
:: Science = Green (green flask bubbling)
:: Social Studies = Blue (blue and green globe, green is taking)
:: Math = Red (math is reliable, red is a strong color so i associate it with reliability)

:: English = Yellow (all other colors are taken)
:: Also electrons are blue
::[[User:Iffy|Iffy]] ([[User talk:Iffy|talk]]) 23:53, 6 February 2023 (UTC)
::: Hm! I've never heard of school subjects having any assigned colors; much less any debate about it! If we're identifying them by the folders they're kept in, my favorite subject was Ferrari & my least favorite was Porsche.
::: [[User:ProphetZarquon|ProphetZarquon]] ([[User talk:ProphetZarquon|talk]]) 04:41, 7 February 2023 (UTC)
:::: I don't recall colour-coded (UK) schoolbooks, in particular (except the "red pirate, green pirate, blue pirate, etc" stories for young kids, the red pirate like only rubies, the green one emeralds, the blue probably sapphires, and had clothing/etc that matched, naturally), but I had (have still, somewhere!) a collection of Usborne Encyclopaedias at home with a veritable rainbow of colours. Mathematics was yellow, I think, Computers a shade of blue, one of the Red or off-Red (slightly pinker, but still deep red) might have been Physics (had geophysics in it, IIRC), I think History was a light-green. I'm sure I never had the whole set, but I had enough to arrange in as close to Richard Of York order as I felt most content to do, when on the bookshelf.
::::: Obviously there ''were'' colours involved with the school stuff. I'm sure different levels of SPMG (Scottish Primary Maths Group?) workbooks were colour-coded, perhaps more for the benefit of the teacher, though the later {{w|School Mathematics Project|SMP}} ones were probably more just identified as "13a", "5b", etc, to work through various sub-subjects and the increasingly advanced techniques thereof, perhaps coloured with highlights only to not be boring black-on-white monochrome covers.
::::: And there's so many other colour-classifications that I instituted for myself, over the years, showing just how useful a hue can be to represent and differentiate a class of something, such as various 3M-style "post-it"-like arrow stickers stuck into the pages of a book for quick reference to all instances of one particular thing or another. For which I suppose I'm grateful to not having any notable form of colour-blindness, to limit my options. [[Special:Contributions/172.70.91.114|172.70.91.114]] 08:20, 7 February 2023 (UTC)
:::: Personally, I've always thought that English is red, Math is blue, Sciences are green, History is yellow, and "personal events" are orange.
::::::This is completely BS. This is about the diagrams used for drawing atoms where colors are used for different elementary particles. And Randall clearly explains that they do not have real color. And the jokes that people still have feelings for what colors are chosen based on the conventions used where people first learned about atoms. Have removed the color on subjects completely as it has nothing to do with this comic. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:43, 7 February 2023 (UTC)
::::::PS you cannot be more than 100% on anything :-D --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:46, 7 February 2023 (UTC)

I believe this comic was made in response to a book talk Randall did in Seattle, where this question was actually asked to him in person! If you want to hear it yourself, someone recorded the talk here: https://www.reddit.com/r/xkcd/comments/xjuc4i/a_recording_and_autotranscript_of_randalls_latest/
[[Special:Contributions/172.71.142.6|172.71.142.6]] 00:45, 7 February 2023 (UTC) A random new user
: Was it the dorky randall with red hair or the photogenic one with brown hair and blue eyes or am I going wildly mad? [[Special:Contributions/172.70.114.198|172.70.114.198]] 00:51, 7 February 2023 (UTC)

Am I crazy, I always thought of electrons as blue to contrast with the protons which are red[[Special:Contributions/172.70.211.89|172.70.211.89]] 04:47, 7 February 2023 (UTC)
:You're all crazy! Elections are 2817.9am & protons are 1.5am. "Yellow" is over 557,000,000,000am! Maybe you've all got your displays' color gamut set too low? ;S
:[[User:ProphetZarquon|ProphetZarquon]] ([[User talk:ProphetZarquon|talk]]) 09:18, 7 February 2023 (UTC)
::I have also seen protons as red and neutons as white and electron as blue in the diagrams I remember. Never yellow electrons. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:43, 7 February 2023 (UTC)
:::I'm not entirely sure about proton and electrons, but neutrons were black. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 20:58, 7 February 2023 (UTC)

"This comic appears to "elevate" that discussion to the college level." - considering that the students are considerably smaller than the teacher (notice the heads), I seriously doubt this is meant to be set in a college classroom - high school at most, IMHO. Also, "One common debate among schoolchildren is over the "color" of various subjects. Because of the brightly colored folders commonly used to separate subjects in the binder of a young student, the students tend to associate those colors with the subject." - well, not in any school I ever attended, nor with any school class I've ever worked with. I'd be inclined to dispute that this is at all common. [[Special:Contributions/172.70.46.85|172.70.46.85]]
: I agree that this is probably not supposed to be college-level, but the color-subject coordination is definitely real (albeit not a very common topic of debate). [[Special:Contributions/162.158.90.38|162.158.90.38]] 08:01, 7 February 2023 (UTC)
::I find it hard to believe Randall is referencing colors of school subjects without alluding to them in any way; to the contrary, I feel fairly certain he's directly referencing the various colors assigned to electrons, protons, quarks, etc, in diagrammatic illustrations of atomic structure. I think the whole first paragraph is way off base (though interesting tangentially).
::[[User:ProphetZarquon|ProphetZarquon]] ([[User talk:ProphetZarquon|talk]]) 09:18, 7 February 2023 (UTC)
:::I agree with all above here and have corrected the explanation to school class and pupils and diagram colors removing school subject color completely! --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:43, 7 February 2023 (UTC)
::::Was it also worth removing the synesthesia bit? Entirely unrelated to school-subject organisation-by-colour that I also think was an {{w|Red herring|incarnadine ''clupea harengus''}}, but very possibly relevent to "but I happen think it's obvious that <concept> is a <hue> thing!"... For consideration, or as a side-note, whether or not you restore that possible reference. [[Special:Contributions/172.71.242.203|172.71.242.203]] 10:42, 7 February 2023 (UTC)

Electrons are blue, right? In all my textbooks (Germany) electrons are blue. Is this a generally accepted addition? [[Special:Contributions/198.41.242.166|198.41.242.166]] 07:13, 7 February 2023 (UTC)
: I stopped the explanation saying that electrons were (by implication, ''solely'') yellow. If green is used for a nucleon (neutron? red being proton?), they might choose blue for an electron, as contrast. Or black dot or white (black-outlined) small circle to contrast with whatever the nucleons are with their much bigger circles clumped in the middle.
: But, given other regular colour-conventions, I could imagine yellow as a popular 'electron' colour. Either in its own right (influencing the choices given to the other things depicted) or as the main obviously remaining option (the other things having been decided upon first). Horses for courses. And I can imagine cultural/national differences (e.g. what colours your household wiring was set up as, at least before EU standardisation but then red and black still exists in the mindset, despite blue and brown, or whatever it might have been) if not localised 'linguistic puns' to make some choices more 'obvious' than others. [[Special:Contributions/172.70.91.114|172.70.91.114]] 08:20, 7 February 2023 (UTC)
::Indeed, yellow is sometimes indicative of electrical hazard, as opposed to red for flame... So many ways to draw associations!
::[[User:ProphetZarquon|ProphetZarquon]] ([[User talk:ProphetZarquon|talk]]) 09:18, 7 February 2023 (UTC)
:::Yes blue electrons, red protons and white neutrons are probably common on Europe, it is in Denmark. I'm a physicist and word with radioactive isotopes and teach about them. My drawings are red protons and white neutrons and blue electrons. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:43, 7 February 2023 (UTC)

I don’t know what Ms. Lenhart is talking about. Electrons are blue, protons are red, and neutrons are definitely grey. Not sure how to sign my comment tho. Oh well {{unsigned ip|172.70.174.115|13:00, 7 February 2023}}
:(You sign your comments with a string of <code><nowiki>~~~~</nowiki></code> (as suggested by the comment at the top of many a comic-discussion page, when you start to edit it)... or you wait for someone else to do what I just did for you, but that's more effort than the four tildes on your part.)
:For what it's worth, I'm mostly with you. Red and grey/dark-grey/black in the centre, as you say. Light blue (or yer actual electric blue?) or (bluish?) white electrons. Depends what colour-pallettes are available to the illustrator/modeller, I imagine, and what else needs a distinct colour alongside the basic trio (e.g. yellow fission/fusion "sparky-flame energy things" or general labelling stuff). [[Special:Contributions/162.158.158.246|162.158.158.246]] 13:15, 7 February 2023 (UTC)
::I did some data collection on image searches for atom diagrams, and yes, the defacto color standard is protons red, neutrons grey (less commonly yellow or green), and electrons blue.
::I like this because it gives opposing colors to the opposing positive and negative charges, (the same color choices as the traditional magnet north and south ends, likely not coincidentally,) and a neutral color to the uncharged neutron.
::Which makes me think that when Lenhart says "electrons are yellow" she does not mean in the diagram sense, but rather in the sense "if you make an electron big enough to see, it is yellow".
::[[User:SomeDee|SomeDee]] ([[User talk:SomeDee|talk]]) 16:58, 7 February 2023 (UTC)
:electrons are green. y'all are trippin [[Special:Contributions/172.71.154.159|172.71.154.159]] 17:27, 7 February 2023 (UTC)
::definitely green. Have none of you ever used a transmission electron microscope? Or an oscilloscope? Green shine everywhere! [[Special:Contributions/162.158.95.22|162.158.95.22]] 09:01, 8 February 2023 (UTC)

I made a survey for this: https://forms.gle/Pu5mkEtBZPUZ6dbb8
[[User:RamenChef|RamenChef]] ([[User talk:RamenChef|talk]]) 18:03, 7 February 2023 (UTC)

Electrons are yellow, protons are red, and neutrons are gray. End of discussion.
:What about roses and violets? [[Special:Contributions/172.71.242.172|172.71.242.172]] 10:49, 8 February 2023 (UTC)

{{w|Quark_(dairy_product)|Quark}} is white, or off-white.[[Special:Contributions/172.70.85.96|172.70.85.96]] 10:47, 8 February 2023 (UTC)
:I find this comic puzzling. Virtually all colors we see are due to electrons (transitions between different states in atoms, molecules, and solids), so saying they are "too small to interact with visible light" is quite incorrect.

2730: Code Lifespan

2023-01-27T16:34:48Z

172.69.134.131: /* Explanation */ one explanation

{{comic
| number = 2730
| date = January 27, 2023
| title = Code Lifespan
| image = code_lifespan_2x.png
| imagesize = 377x307px
| noexpand = true
| titletext = Surely (no one/everyone) will (recognize how flexible and useful this architecture is/spend a huge amount of effort painstakingly preserving and updating this garbage I wrote in 20 minutes)
}}

==Explanation==
{{incomplete|Created by a BOT PROGRAMMED 50 YEARS AGO. Do NOT delete this tag too soon.}}

In the first panel, [[Ponytail]] painstakingly future-proofed her code, but all in vain, as it will never be reused again. In the second panel, she decided that future-proofing would be unneccessary, as she assumes nobody would reuse her code that far in the future anyway. This feels like a {{w|Catch-22}} situation for developers.

The title text proposes four permutations of sentences software developers may consider, which given the context of the comment, usually start out as these two hoped-for ideals:
* Surely everyone will recognize how flexible and useful this architecture is. And,
* Surely no one will spend a huge amount of effort painstakingly preserving and updating this garbage I wrote in 20 minutes.
However, reality often falls short of such hopes, in that insufficient numbers of people recognize code intended for re-use, and far more people than intended will attempt to maintain and adapt sloppy work. The latter sometimes happens because the corner-cutting peculiarities of hasty work are often seen as far deeper necessities than they actually are.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

[Two situations are depicted between Ponytail and Cueball.]

[Situation 1:]

Ponytail: It took some extra work to build, but now we'll be able to use it for all our future projects.

Caption below the panel: How to ensure your code is never reused

[Situation 2:]

Ponytail: Let's not overthink it; if this code is still in use '''''that''''' far in the future, we'll have bigger problems.

Caption below the panel: How to ensure your code lives forever

{{comic discussion}}

2716: Game Night Ordering

2022-12-26T22:35:33Z

172.69.134.131: /* Explanation */ ChatGPT is bad actually

{{comic
| number = 2716
| date = December 26, 2022
| title = Game Night Ordering
| image = game_night_ordering_2x.png
| imagesize = 293x471px
| noexpand = true
| titletext = One good trick, if you get called on a fake service, is to build a working version of it and mention it again the next week.
}}

==Explanation==
{{incomplete|Literally created by ChatGPT. Do NOT delete this tag too soon.}}

The comic is poking fun at the proliferation of online food delivery services. The characters are discussing which one to use, but competing to see who can mention the most fake apps and services without being called out. [[Ponytail]] suggests using Grubhub, and [[Cueball]] mentions a few other options, including Food.net. Cueball expresses skepticism about Yahoo Cash, suggesting that it might be a fake service.

The title text suggests building a working version of a fake service and then mentioning it again the next week.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[Cueball, Megan, and Ponytail are sitting at a table to order food. Cueball is on his phone, and Ponytail, sitting opposite, on her laptop.]
:Cueball: What should we use to order? Grubhub? Doordash? Food.net?
:Ponytail: I'll do Grubhub; you can send me money. Do you do Venmo? Paypal? Yahoo Cash?
:Cueball: Yahoo Cash HAS to be fake.
:Ponytail: Yes. Dang. I'll get your share.
:[Caption below the panel:]
:Our game night has an ongoing competition to see who can mention the most fake apps and services without getting called on it.

{{comic discussion}}

2714: Cold Complaints

2022-12-22T03:42:46Z

172.69.134.131: Undo revision 302163 by 162.158.38.213 (talk)

{{comic
| number = 2714
| date = December 21, 2022
| title = Cold Complaints
| image = cold_complaints_2x.png
| imagesize = 740x254px
| noexpand = true
| titletext = Our investigation into whining-based remedies became the first study to be halted by the IRB on the grounds that the treatment group was 'too annoying.'
}}

==Explanation==
{{incomplete|Created by a GIGANTIC WHINING BABY - Do NOT delete this tag too soon.}}

When people are ill, they will often complain about the symptoms that they're suffering from. A common sterotype is that men will revert to infantile behavior when miserably sick. This can be annoying to the people around them, but they typically tolerate such behavior out of compassion. The joke in this comic is that [[Hairy]] spoke via a {{w|telemedicine}} appointment to [[Ponytail]], a medical professional who explicitly advised him to act out, since his condition has no effective medical treatment. He takes her advice to, "act like you're the first person ever to have a cold," literally, stating it specifically when his companion asks about it.

This comic was published during a "tripledemic" in the U.S., involving {{w|COVID-19}}, {{w|influenza}}, and {{w|respiratory syncytial virus}} (RSV, a frequent cause of common colds) infections, the latter of which have no cure other than to wait them out with plenty of rest and fluids. It expounds on the finding that, "talking about troublesome events, including events with which one is dissatisfied, may ... result in improved physiological health." (Kowalski, R.M. (2002) [http://people.uncw.edu/hakanr/documents/whining.pdf "Whining, griping, and complaining: positivity in the negativity."] ''Journal of clinical psychology,'' '''58'''(9):1023-35.)

The title text describes a similar study, but where the {{w|institutional review board}} (IRB) halted the study because the participants were too annoying. This is ironic since they would be expected to whine annoyingly. IRBs are expected to review the ethics of a research project. Whining is not usually considered dangerous,{{citation needed}} but in this case it was presumably so intolerable that they had to put a stop to it.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[Hairy sitting on a chair in front of a computer screen. There is an image of Ponytail on the screen speaking.]
:Ponytail: Well, it's not COVID or flu. Probably one of the other viruses.
:Hairy: Ughh

:[A zoomed out version of the image, Ponytail cannot be seen]
:Ponytail: There's not much you can do to speed up recovery other than rest, hydrate, and whine and complain and be a gigantic baby about it.

:[Same as previous panel]
:Hairy: Whine and complain?
:Ponytail: Yeah. You need to act like you're the first person ever to have a cold.

:[Hairy lying on a couch wrapped in a blanket, with a lot of paper tissues around]
:Off-screen voice: Are you '''''sure''''' that's what she said?
:Hairy: Unbelievable. Here I am, the only person ever to feel bad, and you're '''''doubting''''' me?

{{comic discussion}}
[[Category:Comics featuring Hairy]]
[[Category:Doctor Ponytail]]
[[Category:Medicine]]
[[Category:COVID-19]]

Talk:2708: Mystery Asterisk Destination

2022-12-08T13:41:20Z

172.69.134.131:

ChatGPT explanation:
: The xkcd comic is a humorous reference to the use of asterisks in written text to indicate a footnote or additional information. In this case, the asterisk in the comic is not accompanied by a corresponding footnote, and instead, the text at the bottom of the panel explains that the asterisk points to this explanation. The joke is that the explanation itself is not particularly useful or illuminating, as it simply states that the asterisk points to this very explanation.
Mediocre at best. Trying again after removing the space after the asterisk from the transcript, and including the title text:
:This xkcd comic is a reference to the use of asterisks and other symbols in footnotes and endnotes to indicate a reference or explanation. In this case, the comic shows a blank panel with a note at the bottom that says "Whenever you see a mystery asterisk that doesn't have a matching footnote, it points here." This is a humorous way of saying that when an asterisk appears in a piece of text without a corresponding footnote, it is pointing to the explanation provided in this comic.

:The title text adds an additional layer of humor by saying that if a writer uses the "dagger" symbol (also known as a "†" symbol) without a corresponding footnote, it means that the writer is threatening the reader with a physical dagger. This is meant to be a tongue-in-cheek joke about the potential dangers of reading footnotes.
Better but still seriously flawed in the final sentence. The prompt prefix I used in front of the transcript was, "Explain the following xkcd comic. The title text should be explained in a separate paragraph.\n\n" [[Special:Contributions/172.69.134.16|172.69.134.16]] 20:10, 7 December 2022 (UTC)
:Adapted in part, fun and good. I wouldn't mind a bot posting to talk when the transcript is stable for half an hour, let's see what it can do! [[Special:Contributions/172.69.33.123|172.69.33.123]] 20:41, 7 December 2022 (UTC)

"with no unmatched footnote" is shockingly bad grammar.

I swear I've seen this comic before somewhere. Was there a similar SMBC? [[Special:Contributions/172.71.166.218|172.71.166.218]] 08:51, 8 December 2022 (UTC)

I feel that the current explanation misses that it's not that uncommon to see asterisks that ought to have a related footnote, but don't. Or a footnote pointed to nowhere in the text. This most often happends on packaging or advertisement. The fact that not every asterisk is there to mark a footnote is a valuable addition, but not the point of the comic.
[[Special:Contributions/172.71.114.7|172.71.114.7]] 13:23, 8 December 2022 (UTC)

This is like the Malkavian sourcebook which contains page XX (where XX is the placeholder page left blank for final pagination)

Isn't the hover text wrong? It seems like he means either "dagger symbol with an unmatched footnote" or "dagger symbol with no matched footnote." Am I missing something here? [[Special:Contributions/172.69.134.131|172.69.134.131]] 13:41, 8 December 2022 (UTC)

2703: Paper Title

2022-11-26T02:47:34Z

172.69.134.131: /* Explanation */ from Wikipedia

{{comic
| number = 2703
| date = November 25, 2022
| title = Paper Title
| image = paper_title_2x.png
| imagesize = 557x261px
| noexpand = true
| titletext = CONFLICT OF INTEREST STATEMENT: The authors hope these results are correct because we all want to be cool people who are good at science.
}}

==Explanation==
{{incomplete|Created by a MICROBE TRYING TO LURE YOU WITH CLICKBAIT. Do NOT delete this tag too soon.}}

Many if not most scientific research papers present a {{w|hypothesis}} and the result of testing the hypothesis. Scientific papers should also have titles which describe the content of the papers. See [[2456: Types of Scientific Paper]].

[[Cueball]] is writing a research paper with a {{w|clickbait}} and {{w|puffery}} title, "Check out this cool microbe we found." His colleague [[Megan]] asks him whether science is supposed to be about formulating a hypothesis and testing it. Cueball agrees, changing the title to, "Is our lab really good at finding cool microbes? Some preliminary data." However, that is still an overly-promotional clickbait title, purporting to be a study of the authors' own competence, which would be highly unusual because of the lack of objectivity caused by the authors being the subject of investigation. [[:Category:Clickbait|Clickbait]] is a recurring theme on xkcd, recently considered within science publications in [[2001: Clickbait-Corrected p-Value]].

{{w|Emperical research|''Empirical investigations''}} and ''{{w|analysis}} papers'' almost always state and test a hypothesis, but there are many kinds of scientific papers which likely will not, including ''{{w|literature review}}s,'' which qualitatively summarize the results of other papers; ''{{w|meta-analysis|meta-analyses}},'' which quantitatively summarize the results and quality of other work; ''observational reports'' (or ''{{w|case study|case studies}}'' — not to be confused with {{w|observational study|observational studies}}, a kind of emperical analysis), which present data and a chronicle of its collection without analysis, testing, or interpretation; ''{{w|Conference proceeding|conference papers}},'' which present preliminary work without peer review; ''definition papers,'' which attempt to formalize terms used in divergent ways in prior work; ''{{w|Dialectic#Hegelian dialectic|syntheses}},'' which present alternative views combining multiple and often conflicting concepts; ''{{w|Comparison|comparative studies}},'' which compare and contrast a class of concepts; ''{{w|Interpretive discussion|interpretive}} papers,'' showing a different perspective on previous work; ''{{w|technical report}}s,'' which may present information on a specific procedural topic or progress and results in a field; ''opinion'' and ''editorial essays,'' which are intended to argue a point of view persuasively; ''book reviews,'' which summarize monographs or biographies; and ''grant proposals,'' which make the case for funding a project. Mathematical or logic research papers which don't involve emperical observations or uncertainty would be considered technical reports in other fields. Engineering work can be reported as an emperical investigation or a technical report.

Cueball seems to want to author an observational report, but Megan would prefer an emperical investigation or analysis, perhaps because they are more likely to be accepted by peer reviewed journals, and as such are more prestigious than mere conference papers, "letters," or "communications" as observational reports are usually published. However, research articles describing the discovery of new microbes in prestigious peer-reviewed journals are often published as observational reports[https://www.microbiologyresearch.org/content/journal/ijsem/10.1099/ijsem.0.004029][http://calamar.univ-ag.fr/mangroveSAE/articles/2022/Volland%20et%20al%202022.pdf][https://link.springer.com/article/10.1007/s10482-021-01656-x] so Megan's concerns may be unfounded; even if so, the editors of any reputable journal would almost certainly require a far more descriptive and less overtly promotional title from Cueball. The question remains whether an intial submission with a catchy clickbait title might get more prompt attention from editors and reviewers.

In the title text, the conflict of interest statement says that the authors hope their results are correct because, "we all want to be cool people who are good at science." A scientific publication's potential {{w|conflict of interest}} usually refers to the authors' financial, familial, or other external interests in the research outcomes. The disclosure statement does not describe a conflict between the authors' extrinsic motivations and factors influencing the accuracy and neutrality of their work; in fact it claims the opposite, an alignment between their intrinsic motivations and the goal of producing high quality work.

==Transcript==
:[Megan is standing behind and looking over the shoulder of Cueball who is sitting in his office chair at his desk typing on the keyboard. A line from the keyboard goes up to two boxes above them. A smaller one at the top, half the length and a third the height of the larger box below. There are text in both boxes. The bottom box is not filled out with text. At the end of the text in the bottom box the line indicating where the courser are can be seen, as in this is what Megan can see on the screen:]
:Paper title
:''Check out this cool microbe we found''|

:[Pan to only showing Megan who has taken a hand up to her chin. Cueball replies from off-panel.]
:Megan: Isn’t science supposed to be about formulating a hypothesis and then testing it?
:Cueball - off panel: Oh. Yeah, I guess.

:[Same setting as in the first panel, but now the bottom box is filled out with text, but still with the courser shown at the end:]
:Paper title
:''Is our lab really good at finding cool microbes? Some preliminary data''|

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Scientific research]]
[[Category:Science]]
[[Category:Clickbait]]

2700: Account Problems

2022-11-18T23:52:39Z

172.69.134.131: Computer problems category

{{comic
| number = 2700
| date = November 18, 2022
| title = Account Problems
| image = account_problems_2x.png
| imagesize = 538x272px
| noexpand = true
| titletext = My password is just every Unicode codepoint concatenated into a single UTF-8 string.
}}

==Explanation==
{{incomplete|Created by a ZERO WIDTH SPACE- Please change this comment when editing this page. Do NOT delete this tag too soon.}}
[[Cueball]] walks toward [[Ponytail]] carrying his laptop. Ponytail is sitting at her desk, and turns to face him. Having attempted to fix Cueball's computer problems before, she replies with dread. Cueball promises that, "It's a normal problem this time.", and Ponytail reluctantly agrees to look at it. Cueball then reveals that he has included a {{w|Null-terminated_string|null string terminator character}} in one of his passwords, probably for a website. Ponytail responds in disbelief, and Cueball defends his actions by saying that the website told him to use special characters.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[Cueball walks up to Ponytail.]
:Cueball: Can you help me with my account?
:Ponytail: Oh no.
:Cueball: No, no, I promise it's a normal problem this time.
:Ponytail: Okay, fine, what is it?
:Cueball: I included a null string character as part of my password, and-
:Ponytail: How?!
:Cueball: They said to use special characters!

{{comic discussion}}

[[Category:Cueball Computer Problems]]

2699: Feature Comparison

2022-11-17T01:23:06Z

172.69.134.131: /* Explanation */

{{comic
| number = 2699
| date = November 16, 2022
| title = Feature Comparison
| image = feature_comparison_2x.png
| imagesize = 740x470px
| noexpand = true
| titletext = Below the Web, and the Dark Web, a shadowy parallel world of Cybiko users trade messages on the Translucent Neon Plastic Web.
}}

==Explanation==
{{incomplete|Created by a MESH NETWORK. Do NOT delete this tag too soon.}}

This is comparing different remote communication services.





The “Cybiko® wireless handheld computer for teens (2000)” is described in Wikipedia’s article {{w|Cybiko}}. Additional information about it is available at [http://cultureandcommunication.org/deadmedia/index.php/Cybiko the Dead Media Archive], as the device has not been manufactured since 2003. The Cybiko has two-way communication capabilities along with built-in games and music player capabilities.

The chart suggests that the Cybiko has an advantage over all of the other ten communication services listed, as it has all eight of the features listed in the table, whereas none of the other services under consideration have move than five of the features.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[A table with checkmarks to indicate which features various messaging services have. Each column is labeled with a service name and its logo beneath, except that for the last column, the device's longer name is written higher than all the other services' names, with an arrow pointing to a drawing of the device below it.]

{| class="wikitable"
|-
!
! Twitter
! Discord
! Mastodon
! FB
! Slack
! Signal
! IRC
! Tumblr
! Reddit
! SMS
! Cybiko® wireless handheld computer for teens (2000)
|-
! Direct messages
| ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓
|-
! Group chats
| ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || || ✓ || || ✓
|-
! File transfer
| || ✓ || ✓ || ✓ || ✓ || ✓ || ✓ || || ✓ || || ✓
|-
! Built-in games
| || ✓ || || ✓ || || || || || || || ✓
|-
! User-run instances
| || ✓ || || || ✓ || || ✓ || || ✓ || || ✓
|-
! Doesn't require central server
| || || || || || || ✓ || || || || ✓
|-
! Mesh networking
| || || || || || || || || || || ✓
|-
! Wireless message delivery works without internet
| || || || || || || || || || ✓ || ✓
|-
|}

{{comic discussion}}

Talk:2698: Bad Date

2022-11-15T19:24:09Z

172.69.134.131: Biden Sucks ®

There was a time when just the fact that she was looking at her cellphone during the date would be evidence that it's not going well. Sadly, it's probably normal these days. [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 15:05, 14 November 2022 (UTC)

https://www.youtube.com/watch?v=w1iH1hppneo [[Special:Contributions/172.71.158.90|172.71.158.90]] 16:34, 14 November 2022 (UTC)
:She went on a date with an Explainxkcd editor, obvs. [[Special:Contributions/172.71.158.216|172.71.158.216]] 19:55, 14 November 2022 (UTC)

@Kynde, but more generally putting it out there: "[https://www.explainxkcd.com/wiki/index.php?title=2698:_Bad_Date&diff=298707&oldid=298705 For sure it is Megan who came into the date with jumanji. Else playing dumb by Cueball would not make sense, because Megan would likely not have mentioned Jumanji. How should Cueball mention it and then say that he don't know it?]" - the aside that this refers to/removed was written to cover if Cueball had sneakily forced conversation about Jumanji by somehow steering the conversation towards ("No film has had a rhino charge through an American town", "I sometimes wonder what it would be like to be a video-game character", ...) while pretending not to have seen it, or perhaps 'clearly' have not watched it correctly. All to incite the date (Megan) into being loudly enthused. Cueball's "Oh no" is more like how Gene Wilder's Willy Wonka would have said it (a cursory level of concern about how 'bad' things have gone, it actually being his plan all along to seed the events; but also perhaps instead some awkwardness that Megan picked up on the live-posting). Megan independently has her ideas and Cueball (as the usual voice of Randall) decides shifting to her newly acquired sponsor is worthwhile.
 The alternative currently promoted is that Megan came with the Jumanji idea, somehow exasperatingly runs out of steam of how to promote "Jumanji: Deeper Into the Metaverse" (or whatever fictional/factual thing she's supposed to be spreading news of) and isn't happy that her outbursts are reaching the large audience (that would surely be so good for her attempt to put the word out, even if she was originally supposed to be individually advertising to just the one date at a time..?) and yet somehow gets the inspiration to spend precious moments breaking the (in-date) fourth wall to clue Cueball in on the fuss she has caused, find and accept an alternate backer, announce it as a new idea that Cueball should roll with (perhaps ruining the date more, such that he does play along but definitely hates Megan more than when she was irrationally going on about this 'Jumanji' thing) and then breaks the mood ''yet again'' by revealing that the interesting idea of a "sponsored conversation" was not necessarily an opportunistic turn of events, but had (in another form, for another sponsor) been her basic idea all along.

I like how Cueball somehow included the registered trademark in his dialogue(®). Seems small, but certainly intentional, for Randall to add it.
 I'm still willing to believe it's Megan who says the title text (there's some possibility that it flows that way, in arguments that I haven't bothered to voice above), but I find the idea that it is Cueball to have more than a little attractiveness in various key ways. Definitely it isn't sure that it isn't him, and potentially quite confusing as to who it really is. [[Special:Contributions/172.71.178.207|172.71.178.207]] 21:57, 14 November 2022 (UTC)

:I actually don't think it mattes who said the title text. But since the whole explanation is already overexplaining the comic it might not matter. --[[Special:Contributions/172.70.246.98|172.70.246.98]] 23:20, 14 November 2022 (UTC)

In Nathan For You S01E08, a scenario is engineered where they get someone on a date, but one of the participants has to give product endorsements for Quiznos. A Quiznos marketing rep is wired in to a hidden earpiece giving the marketing keywords to say while watching the date from a hidden camera. [[Special:Contributions/172.70.93.37|172.70.93.37]] 11:53, 15 November 2022 (UTC)

2694: Königsberg

2022-11-08T13:33:01Z

172.69.134.131: /* Explanation */ fix file name

{{comic
| number = 2694
| date = November 4, 2022
| title = Königsberg
| image = konigsberg_2x.png
| imagesize = 448x343px
| noexpand = true
| titletext = At first I thought I would need some gold or something to pay him, but then I realized that it was the 18th century and I could just bring a roll of aluminum foil.
}}

==Explanation==
{{incomplete|Created by a WOLF, TWO GOATS, AND THREE BAGS OF GRAPH NODES. Do NOT delete this tag too soon.}}

[[File:Konigsberg_bridges.png|frame|right|{{w|Königsberg}}, Prussia in Euler's time, showing the Pregel river and its seven bridges. Two of the original seven bridges no longer exist,[https://goo.gl/maps/ChdBoeQMr3AQPi446] although there are three new bridges. The Baltic port city is now Kaliningrad, a Russian exclave.]]

This comic is about the {{w|Seven Bridges of Königsberg}}, a seminal {{w|graph theory}} problem solved by the famous mathematician {{w|Leonhard Euler}}.[https://www.maa.org/press/periodicals/convergence/leonard-eulers-solution-to-the-konigsberg-bridge-problem] The problem was whether a path through the city crossing each of the seven bridges just once exists, without crossing the river forks any other way. In 1736, Euler proved that there is no such path. This result is considered to be the first theorem of graph theory and the first proof in the theory of networks[http://www-personal.umich.edu/~mejn/courses/2004/cscs535/review.pdf] — a subject now generally regarded as a branch of {{w|combinatorics}} — and presaged the development of {{w|topology}}. Combinatorial problems of other types had been considered since antiquity. {{w|Graph (discrete mathematics)|Graphs}} are a data structure common in many algorithmic problems in computer science.

[[Cueball]] attempts to cheat on the final exam in his algorithms class by traveling back in time to commission the construction of an eighth bridge before Euler could learn of the problem, allowing a trivial solution that would remove the rationale for further analysis. He hopes that this would alter his present-day timeline in such a way that the test becomes easier because graph theory might never have been developed. The use of the word "tried" implies failure, which is probably a good thing since his success would create a {{w|Temporal_paradox#Grandfather_paradox|paradox}}. [[:Category:Time travel|Time travel]] is a recurring topic on xkcd and examples where attempts to change the past fails has also been used before like in [[1063: Kill Hitler]].

With the addition of the eighth bridge, it becomes possible to cross each bridge exactly once, starting at the north bank and ending on the larger eastern island, or vice-versa. However, there is still no way to traverse each bridge exactly once and return to the starting point, because the altered graph would have an {{w|Eulerian trail|Euler trail}} but not an Euler cycle. Thus the problem might still have been interesting to Euler.{{Citation needed}} (Adding a ninth bridge connecting the north bank to the east island would render the problem completely trivial.) We can't say whether Euler or others would have developed graph theory anyway, or whether Cueball's exam would have been any easier, more difficult, or similar.

An alternative modification allowing an easy solution is to remove bridges. During World War II, two bridges to the central island connecting it to the north and south banks were destroyed by bombing, so today there is an Eulerian trail across the five remaining bridges.

The title text alludes to the fact that ordinary {{w|aluminum foil}}, which was not commercially available until 1911, would have been a tremendously valuable curiosity in the 18th century, which didn't even have {{w|tin foil}}. Aluminum was a highly priced metal before the 1880s when inexpensive methods were developed to refine it. The {{w|Washington Monument#Aluminum_apex|Washington Monument}} was constructed with a tip made of pure aluminum due to its value and conductive capacity. Aluminum had not been extracted in its pure form at the time of Euler, and was known only in compounds such as {{w|alum}}, so the metal would have been unique and exotic. The value of aluminum and the use of it as the tip of the Washington Monument was also mentioned in [[1608: Hoverboard]] where a heist to steal the tip is [https://www.explainxkcd.com/wiki/images/6/6f/1608_0995x1083y_Tip_of_Washington_monument.png depicted].

==Transcript==
:[Cueball, standing next to two men wearing wigs, pointing with a pointer at a map showing the seven bridges problem, with an extra bridge added in dashed lines]
:Cueball: Lord Mayor of Königsberg, I will reward you handsomely if you construct this bridge before my friend Leonhard arrives.

:[Caption below the panel:]
:I tried to use a time machine to cheat on my algorithms final by preventing graph theory from being invented.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring real people]]
[[Category:Time travel]]
[[Category:Math]]
[[Category:Programming]]

Talk:2688: Bubble Universes

2022-10-22T00:21:55Z

172.69.134.131: Reply

Is this a reference to the money kind of inflation? [[Special:Contributions/172.70.131.6|172.70.131.6]] 16:03, 21 October 2022 (UTC)

: no [[Special:Contributions/172.70.114.205|172.70.114.205]] 16:29, 21 October 2022 (UTC)
: I do think it a form of irony that when I hear "inflation" I do *NOT* think about bubbles or balloons. 20:08, 21 October 2022 (UTC)

*pop* [[User:JLZ0kTC5|JLZ0kTC5]] ([[User talk:JLZ0kTC5|talk]]) 16:13, 21 October 2022 (UTC)

I think this comic might be referencing the [[wikipedia:Eternal inflation|eternal inflation]] model: as the universe eternally inflates at an exponential rate, bubbles form where the inflation slows down, creating a disjoint multiverse from these bubbles being unable to interact with each other. In this comic, the process is shown as recursive. [[User:LegionMammal978|LegionMammal978]] ([[User talk:LegionMammal978|talk]]) 16:42, 21 October 2022 (UTC)

:: I'm included to agree. Should the article be changed? 20:08, 21 October 2022 (UTC)

This comic would be perfect motive for a mug. [[User:Fabian42|Fabian42]] ([[User talk:Fabian42|talk]]) 19:04, 21 October 2022 (UTC)
:Though I think you mean "motif", although I also agree in principle with your word as well, if it does motivate someone to create one... ;) [[Special:Contributions/172.70.162.77|172.70.162.77]] 20:33, 21 October 2022 (UTC)

I don't think this comic references the big bounce, but rather the theory of nested universes. The way it works is each black hole spawns a new universe, and is connected to a white hold in the new universe. If true, our universe came from another universe capable of producing black holes, and since our universe has black holes in it, according to the theory it has in turn created additional universes, each with the potential capability to produce its own black holes and therefore create additional universes. It's kind of like universe propagation.

[Responding to the title text:] '''As a cosmologist, it most certainly does not! Cosmic inflation took less than 10^-32 seconds involving superluminal expansion of points starting the width of a quark from each other. Any physical gas inflating of a membrane is simply not comparable, and if this is a joke about [https://twitter.com/RepKatiePorter/status/1582475617723113472] then it's not funny. I would die on this hill if I had time to argue with you all about this.''' [[Special:Contributions/172.70.211.146|172.70.211.146]] 21:52, 21 October 2022 (UTC)

Not sure "The Big Bounce" has currency here, as per the current explanation. That's a cyclic theory (generally relying upon a Big Crunch ending one version of universe, the result sparking the next Big Bang... certainly in the version I would favour, if it turned out we were actually not destined towards untempered expansion and either Heat Death or Big Rip as the unending finale) not a recursively embedding one (e.g. Inception-like realities within realities). But that's just my opinion. Someone who actually finds a Big Bounce model of spacetime more satisfying than any open-ended one, even. [[Special:Contributions/172.69.79.201|172.69.79.201]] 22:04, 21 October 2022 (UTC)
:There are a number of cyclic cosmologies possibly involving a "big crunch" such as ekpyrotic theories, but we need telescopes much more powerful than the JWST to say anything about whether they are possible, likely, or propose improvements to them. If the pre-inflationary universe was actually a point, we may never know. There is a large variety of other open questions which are much easier to answer. [[Special:Contributions/172.69.134.131|172.69.134.131]] 00:21, 22 October 2022 (UTC)

This makes me think of his earlier comics. [[User:N-eh|N-eh]] ([[User talk:N-eh|talk]]) 00:15, 22 October 2022 (UTC)

2683: Fan Theories

2022-10-11T14:27:57Z

172.69.134.131: /* Explanation */ paragraph

{{comic
| number = 2683
| date = October 10, 2022
| title = Fan Theories
| image = fan_theories_2x.png
| imagesize = 267x318px
| noexpand = true
| titletext = The universe fandom is great. Such sweet and enthusiastic people.
}}

==Explanation==
{{incomplete| Created by a THREE-D PRINTED FAN BASED ON A HYPOTHETICAL SOLUTION TO THE NAVIER-STOKES LAMINAR-TURBULENCE BOUNDARY EQUATION. Do NOT delete this tag too soon.}}

In this comic, [[Cueball]] refers to scientific hypotheses as fan theories. Scientists might probably consider doing so as undervaluing their work, because scientific hypotheses are usually the result of much more serious research. Because there are a lot of charismatic cranks who obtain undeserved recognition for their supposedly scientific hypotheses, such disrespect could be perceived as insensitive to real scientists. Once again, [[Randall]] has devised a new way to annoy a very sizable proportion of the scientific community.

This comic appeared a few days after renewed suggestions that life may exist [https://www.space.com/saturn-moon-enceladus-ocean-phosphorus within one of Saturn's moons], and 60 years after [https://iep.utm.edu/kuhn-ts/ Thomas S. Kuhn's] renowned 1962 treatise, ''{{w|The Structure of Scientific Revolutions}}''.

The title text may have to do with unwelcoming or inconsiderate {{w|fandom}}s.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}
:[Ponytail, Cueball, White Hat, and Megan standing in a line. Cueball has his hand raised, and Megan has hands balled into fists, exasperated.]
:Cueball: But according to leading fan theories, Jupiter's moons may harbor subsurface oceans.
:Megan: Will you ''please'' stop calling them that?

:[Caption below the panel:]
:How to annoy scientists: refer to all hypotheses as "fan theories"

{{comic discussion}}
[[Category:Astronomy]]
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Cueball]]
[[Category:Comics featuring White Hat]]
[[Category:Comics featuring Megan]]

2683: Fan Theories

2022-10-11T14:27:04Z

172.69.134.131: /* Explanation */ grammar

{{comic
| number = 2683
| date = October 10, 2022
| title = Fan Theories
| image = fan_theories_2x.png
| imagesize = 267x318px
| noexpand = true
| titletext = The universe fandom is great. Such sweet and enthusiastic people.
}}

==Explanation==
{{incomplete| Created by a THREE-D PRINTED FAN BASED ON A HYPOTHETICAL SOLUTION TO THE NAVIER-STOKES LAMINAR-TURBULENCE BOUNDARY EQUATION. Do NOT delete this tag too soon.}}

Once again, [[Randall]] has devised a new and effective way to annoy a very sizable proportion of the scientific community.

In this comic, Cueball refers to scientific hypotheses as fan theories. Scientists might probably consider doing so as undervaluing their work, because scientific hypotheses are usually the result of much more serious research. Because there are a lot of charismatic cranks who obtain undeserved recognition for their supposedly scientific hypotheses, such disrespect could be perceived as insensitive to real scientists.

This comic appeared a few days after renewed suggestions that life may exist [https://www.space.com/saturn-moon-enceladus-ocean-phosphorus within one of Saturn's moons], and 60 years after [https://iep.utm.edu/kuhn-ts/ Thomas S. Kuhn's] renowned 1962 treatise, ''{{w|The Structure of Scientific Revolutions}}''.

The title text may have to do with unwelcoming or inconsiderate {{w|fandom}}s.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}
:[Ponytail, Cueball, White Hat, and Megan standing in a line. Cueball has his hand raised, and Megan has hands balled into fists, exasperated.]
:Cueball: But according to leading fan theories, Jupiter's moons may harbor subsurface oceans.
:Megan: Will you ''please'' stop calling them that?

:[Caption below the panel:]
:How to annoy scientists: refer to all hypotheses as "fan theories"

{{comic discussion}}
[[Category:Astronomy]]
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Cueball]]
[[Category:Comics featuring White Hat]]
[[Category:Comics featuring Megan]]

2683: Fan Theories

2022-10-11T14:24:49Z

172.69.134.131: Rearrange

{{comic
| number = 2683
| date = October 10, 2022
| title = Fan Theories
| image = fan_theories_2x.png
| imagesize = 267x318px
| noexpand = true
| titletext = The universe fandom is great. Such sweet and enthusiastic people.
}}

==Explanation==
{{incomplete| Created by a THREE-D PRINTED FAN BASED ON A HYPOTHETICAL SOLUTION TO THE NAVIER-STOKES LAMINAR-TURBULENCE BOUNDARY EQUATION. Do NOT delete this tag too soon.}}

Once again, [[Randall]] has devised a new and effective way to annoy a very sizable proportion of the scientific community.

In this comic, Cueball refers to scientific hypotheses as fan theories. This a scientist would probably consider undervaluing their work, as scientific hypotheses in contrast to fan theories are the result of serious research. Because there are a lot of charismatic cranks who obtain undeserved recognition for their supposedly scientific hypotheses, this could be perceived as insensitive to real scientists.

This comic appeared a few days after renewed suggestions that life may exist [https://www.space.com/saturn-moon-enceladus-ocean-phosphorus within one of Saturn's moons], and 60 years after [https://iep.utm.edu/kuhn-ts/ Thomas S. Kuhn's] renowned 1962 treatise, ''{{w|The Structure of Scientific Revolutions}}''.

The title text may have to do with unwelcoming or inconsiderate {{w|fandom}}s.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}
:[Ponytail, Cueball, White Hat, and Megan standing in a line. Cueball has his hand raised, and Megan has hands balled into fists, exasperated.]
:Cueball: But according to leading fan theories, Jupiter's moons may harbor subsurface oceans.
:Megan: Will you ''please'' stop calling them that?

:[Caption below the panel:]
:How to annoy scientists: refer to all hypotheses as "fan theories"

{{comic discussion}}
[[Category:Astronomy]]
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Cueball]]
[[Category:Comics featuring White Hat]]
[[Category:Comics featuring Megan]]

2683: Fan Theories

2022-10-11T01:06:14Z

172.69.134.131: /* Explanation */ ahem

2681: Archimedes Principle

2022-10-06T06:35:20Z

172.69.134.131: fluid mechanics, not fluid dynamics

{{comic
| number = 2681
| date = October 5, 2022
| title = Archimedes Principle
| image = archimedes_principle_2x.png
| imagesize = 312x379px
| noexpand = true
| titletext = "I've always wanted to run naked through town, but I don't want to get in trouble with the king or be remembered by history as a weirdo. I wonder how I could ... EUREKA!"
}}

==Explanation==
{{incomplete|Created for a heist - Please change this comment when editing this page. Do NOT delete this tag too soon.}}

{{w|Archimedes' Principle}} is a well known principle of fluid mechanics that states "Any object, totally or partially immersed in a fluid or liquid, is buoyed up by a force equal to the weight of the fluid displaced by the object." {{w|Archimedes}} supposedly discovered it when he was getting into a bathtub and noticed how much water spilled out of the tub when he got into it.

Archimedes' insight led to the solution of a problem posed by king Hiero of Syracuse, on how to assess the purity of an irregular golden votive crown; Hiero had given his goldsmith the pure gold to be used, and correctly suspected he had been cheated by the goldsmith removing gold and adding the same weight of silver. Equipment for weighing objects with a fair amount of precision already existed, and now that Archimedes could also measure volume, their ratio would give the object's density, an important indicator of purity (as gold is nearly twice as dense as silver and therefore has significantly greater weight for the same volume). The legend says that upon discovering that he could use this insight to solve the problem, he went running naked down the street shouting "{{w|Eureka (word)|Eureka}}!". Note that Archimedes' Principle is about buoyancy, not volume measurement, so his Principle is not strictly required for the crown measurement story; however [https://www.mcgill.ca/oss/article/history/it-true-archimedes-formulated-his-famous-principle-based-observation-he-made-he-immersed-himself Archimedes did formulate] the Archimedes' Principle.

In the comic, Archimedes plans on swapping the king's gold crown with a less valuable silver crown with gold plating, and this is claimed to be the first heist. It almost certainly wouldn't be the first robbery, but seems to refer to the complex, planned robberies of large amounts that are typical of {{w|heist films}}.

According to the title text, Archimedes's eureka moment wasn't when he discovered the buoyancy principle, but when he realized that he could use this discovery as a pretext for the reason for running naked through town, which is something he'd always wanted to do.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}
:[Megan is seen picking up a crown from a bucket filled with water, while Archimedes, depicted as a balding man with a short white beard, is writing at a desk.]
:Megan: Uh, Archimedes, why is there a bucket of water with a gold crown hidden in the bottom?
:Archimedes: It's mostly silver. Replica of the King's crown. He's coming here later, and I have a plan.

:[Caption beneath the panel:]
:Archimedes invents the heist.

{{comic discussion}}
[[Category: Comics featuring Megan]]
[[Category: Comics featuring real people]]
[[Category: Physics]]

2681: Archimedes Principle

2022-10-06T06:34:30Z

172.69.134.131: Undo revision 296133 by 172.69.134.131 (talk)

{{comic
| number = 2681
| date = October 5, 2022
| title = Archimedes Principle
| image = archimedes_principle_2x.png
| imagesize = 312x379px
| noexpand = true
| titletext = "I've always wanted to run naked through town, but I don't want to get in trouble with the king or be remembered by history as a weirdo. I wonder how I could ... EUREKA!"
}}

==Explanation==
{{incomplete|Created for a heist - Please change this comment when editing this page. Do NOT delete this tag too soon.}}

{{w|Archimedes' Principle}} is a well known principle of fluid dynamics that states "Any object, totally or partially immersed in a fluid or liquid, is buoyed up by a force equal to the weight of the fluid displaced by the object." {{w|Archimedes}} supposedly discovered it when he was getting into a bathtub and noticed how much water spilled out of the tub when he got into it.

Archimedes' insight led to the solution of a problem posed by king Hiero of Syracuse, on how to assess the purity of an irregular golden votive crown; Hiero had given his goldsmith the pure gold to be used, and correctly suspected he had been cheated by the goldsmith removing gold and adding the same weight of silver. Equipment for weighing objects with a fair amount of precision already existed, and now that Archimedes could also measure volume, their ratio would give the object's density, an important indicator of purity (as gold is nearly twice as dense as silver and therefore has significantly greater weight for the same volume). The legend says that upon discovering that he could use this insight to solve the problem, he went running naked down the street shouting "{{w|Eureka (word)|Eureka}}!". Note that Archimedes' Principle is about buoyancy, not volume measurement, so his Principle is not strictly required for the crown measurement story; however [https://www.mcgill.ca/oss/article/history/it-true-archimedes-formulated-his-famous-principle-based-observation-he-made-he-immersed-himself Archimedes did formulate] the Archimedes' Principle.

In the comic, Archimedes plans on swapping the king's gold crown with a less valuable silver crown with gold plating, and this is claimed to be the first heist. It almost certainly wouldn't be the first robbery, but seems to refer to the complex, planned robberies of large amounts that are typical of {{w|heist films}}.

According to the title text, Archimedes's eureka moment wasn't when he discovered the buoyancy principle, but when he realized that he could use this discovery as a pretext for the reason for running naked through town, which is something he'd always wanted to do.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}
:[Megan is seen picking up a crown from a bucket filled with water, while Archimedes, depicted as a balding man with a short white beard, is writing at a desk.]
:Megan: Uh, Archimedes, why is there a bucket of water with a gold crown hidden in the bottom?
:Archimedes: It's mostly silver. Replica of the King's crown. He's coming here later, and I have a plan.

:[Caption beneath the panel:]
:Archimedes invents the heist.

{{comic discussion}}
[[Category: Comics featuring Megan]]
[[Category: Comics featuring real people]]
[[Category: Physics]]

2681: Archimedes Principle

2022-10-06T06:33:16Z

172.69.134.131: nothing dynamic about fluid displacement

2679: Quantified Self

2022-10-01T10:17:52Z

172.69.134.131: /* Explanation */ it's all good as a citation for this reference

{{comic
| number = 2679
| date = September 30, 2022
| title = Quantified Self
| image = quantified_self_2x.png
| imagesize = 386x328px
| noexpand = true
| titletext = It's made me way more excited about ferris wheels, subways, car washes, waterslides, and store entrances that have double doors with a divider in the middle.
}}

==Explanation==
{{incomplete|Created by an IMAGINARY PATH-STRING STRANGULATION VICTIM. Do NOT delete this tag too soon.}}

[[Cueball]] (presumably a representation of [[Randall]] in this comic) is talking about how he embraces the {{w|quantified self}}, a popular philosophy promoting monitoring yourself with devices and data in the hope to help your well-being. He claims to apply this philosophy to his life, but as is usual for Randall, he does so in his very unique and absurdly humorous way, with help from a smart watch or handheld mobile device.

Typically, fitness apps and wearable devices will track the number of steps that users take and distances walked or run, along with other measurements such as heart rate, blood oxygenation level, blood pressure, and mood. This is to encourage users to be more physically active. However, Randall has chosen to track a modified version of this metric, in which his path is post-processed by contracting it. Ordinarily, people begin and end their days in bed; in this case, it can get 'caught' where Randall has passed through topological tunnels. (See [[2658: Coffee Cup Holes]] and [[2625: Field Topology]] for details.) In the comic strip, we see that, over the course of his week, Randall has looped around his house twice (which could itself conceal any number of activities, so long as he left through his front door and returned through his back door) and crossed under two highway overpasses, a highway sign, and apparently the St. Louis {{w|Gateway Arch}} before almost returning home.

This comic appeared two days after [https://blog.google/products/search/new-ways-to-make-more-sustainable-choices/ Google's announcement that Maps Directions will be sortable by sustainability.] This may be particularly notable because of tech industry discussions between employees and executives about cost-benefit analyses comparing sharply increased profits and productivity from work-from-home to the value of coastal region commercial office space holdings and leases, relative to [https://www.epa.gov/climateleadership/scope-3-inventory-guidance scope 3 emissions].[https://www.latimes.com/business/technology/story/2021-04-28/google-is-saving-1-billion-per-year-as-a-result-of-employees-working-from-home][https://www.wsj.com/articles/companies-cutting-office-space-predict-long-term-savings-11625493601][https://www.bloomberg.com/news/articles/2021-07-15/google-googl-wants-employees-to-return-to-office-despite-productivity-gains][https://www.reuters.com/world/the-great-reboot/pay-cut-google-employees-who-work-home-could-lose-money-2021-08-10/][https://www.forbes.com/sites/jackkelly/2022/01/14/google-spends-billions-on-buying-office-buildings-is-this-a-sign-of-the-post-pandemic-pushback-against-remote-work/][https://www.computerworld.com/article/3659891/google-others-adding-office-space-in-anticipation-of-the-great-return.html] Please see [[2207: Math Work]].

The title text mentions concepts useful adjuncts for such measurements and evaluations, such as passing (one way) through any tube, tunnel, ferris wheel, car wash, water slide or double door frame made of solid material that could thus capture the imaginary string and help to keep its ultimate distance as lengthy as possible.

===OCD interpretation===

The quantity Randall measures can be recognized as a specific type of OCD where people feel like they have an imaginary string connecting them to where they come from.[https://www.reddit.com/r/OCD/comments/1ve309/invisible_thread_attached_to_my_back_am_i_the/][https://www.reddit.com/r/xkcd/comments/xs50yr/xkcd_2679_quantified_self/iqjcmbv/?context=3]
As they move around, that string gets entangled and they feel the urge to untangle it. When they enter a car, they feel the need to exit the car from the same door, to avoid that the string gets trapped by forever passing through the car. When they enter a building, they feel they need to exit using the same staircases and doorway(s), to avoid entangling the string in the building. Some situations, like turning around a lamp post, are OK because you can imagine removing the loop over the top of the lamp post, such that it is not really entangled.

Randall tries a new approach to deal with this OCD by integrating it in his quantified self. He defines precisely how to measure the length of the imaginary string, reduced to its minimum, and chooses this as a quantity to monitor. Unlike most people with this OCD, who feel the urge to minimize the length, Randall takes the opposite stance trying to maximize the (optimally minimal) length of the string.

This could be thought as a therapy. By defining as a target to achieve a given length every day, he creates a drive to embrace situations that entangle the string. This drive opposes the natural compulsion to avoid them and hopefully cancels it. The joke of the title text is that Randall now becomes overly interested in all the things that are disturbing for people with the OCD. The monitoring has just reversed his obsession.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[A red path links two red Cueballs. It start from the left Cueball, does two loops through a small house, under a first road bridge, under a gantry sign, under a second road bridge, under the Gateway Arch, and to a second red Cueball on the right. That Cueball is looking at a smart watch or a mobile device.]

:Red Cueball's watch or mobile device: Good job! You hit your weekly goal for "total length of your path through space if you minimize its length by pulling it taut, maneuvering it around solid objects but not through them."

:[Caption below the panel:]
:I'm into the quantified self, but only for really arbitrary quantities.

{{comic discussion}}

[[Category:Comics with color]]
[[Category:Comics featuring Cueball]]

Talk:2676: Historical Dates

2022-09-24T13:35:00Z

172.69.134.131: 1582

Source for the Excel/Lotus 123 relation with Dec 30th, 1899: https://social.msdn.microsoft.com/Forums/office/en-US/f1eef5fe-ef5e-4ab6-9d92-0998d3fa6e14/what-is-story-behind-december-30-1899-as-base-date?forum=accessdev
[[User:Victor|Victor]] ([[User talk:Victor|talk]]) 08:14, 24 September 2022 (UTC)

I feel this one. My birthday happens to be within 24 hours of [[1179: ISO 8601|1970-01-01]], so I keep getting caught off guard for a moment whenever I see my birthday showing up in one of these contexts. -- [[User:KarMann|KarMann]] ([[User talk:KarMann|talk]]) 08:35, 24 September 2022 (UTC)

We're going to need the date stamp format for 1890 ticker tape for this one. Anyone? [[Special:Contributions/172.70.214.183|172.70.214.183]] 11:59, 24 September 2022 (UTC)
:MM/DD/YY, with leading zeros omitted, and no I don't know why, but I suggest Google Books Ngrams might have a clue as to when that abomination started. [[Special:Contributions/172.69.22.61|172.69.22.61]] 12:03, 24 September 2022 (UTC)
::Not necessarily with pairs of the slash '/' _ . . _ . but also hyphens '-' _ . . . . _ and periods '.' . _ . _ . _ were used as delimiters in MM?DD?YY, which if I remember right dates to the 1500s when accounting ledgers were invented. [[Special:Contributions/172.70.207.8|172.70.207.8]] 12:10, 24 September 2022 (UTC)
::::Are you with the NSA and have a data warehouse of all the ticker tapes ever sent or something? [[Special:Contributions/162.158.166.185|162.158.166.185]] 12:45, 24 September 2022 (UTC)
:::::No, but my great grandparents thought ticker tape parades were littering, because Great Grandma worked in an office and Great Grandpa worked for sanitation, so we have a bunch of boxes in the attic filled with what she was supposed to throw out her window. [[Special:Contributions/172.71.158.75|172.71.158.75]] 12:54, 24 September 2022 (UTC)
:::That would be [https://www.jstor.org/stable/40697544 1299]. But I'm not sure how this is going to help us explain the comic, unless you perhaps are suggesting we enumerate date representation clusters somehow? [[Special:Contributions/172.69.22.61|172.69.22.61]] 12:32, 24 September 2022 (UTC)
:::: Someone should ask GPT-3 for a list of the top ten dates. [[Special:Contributions/172.69.22.71|172.69.22.71]] 12:59, 24 September 2022 (UTC)

Just putting January 2, 2006 here: https://stackoverflow.com/questions/20530327/origin-of-mon-jan-2-150405-mst-2006-in-golang [[Special:Contributions/172.69.22.161|172.69.22.161]] 12:28, 24 September 2022 (UTC)

Do we need a comment about how Pope Gregory XIII obliterated October 5th through 14th, 1582? [[Special:Contributions/172.71.158.25|172.71.158.25]] 13:30, 24 September 2022 (UTC)

Talk:2676: Historical Dates

2022-09-24T13:34:13Z

172.69.134.131: Edit

Source for the Excel/Lotus 123 relation with Dec 30th, 1899: https://social.msdn.microsoft.com/Forums/office/en-US/f1eef5fe-ef5e-4ab6-9d92-0998d3fa6e14/what-is-story-behind-december-30-1899-as-base-date?forum=accessdev
[[User:Victor|Victor]] ([[User talk:Victor|talk]]) 08:14, 24 September 2022 (UTC)

I feel this one. My birthday happens to be within 24 hours of [[1179: ISO 8601|1970-01-01]], so I keep getting caught off guard for a moment whenever I see my birthday showing up in one of these contexts. -- [[User:KarMann|KarMann]] ([[User talk:KarMann|talk]]) 08:35, 24 September 2022 (UTC)

We're going to need the date stamp format for 1890 ticker tape for this one. Anyone? [[Special:Contributions/172.70.214.183|172.70.214.183]] 11:59, 24 September 2022 (UTC)
:MM/DD/YY, with leading zeros omitted, and no I don't know why, but I suggest Google Books Ngrams might have a clue as to when that abomination started. [[Special:Contributions/172.69.22.61|172.69.22.61]] 12:03, 24 September 2022 (UTC)
::Not necessarily with pairs of the slash '/' _ . . _ . but also hyphens '-' _ . . . . _ and periods '.' . _ . _ . _ were used as delimiters in MM?DD?YY, which if I remember right dates to the 1500s when accounting ledgers were invented. [[Special:Contributions/172.70.207.8|172.70.207.8]] 12:10, 24 September 2022 (UTC)
::::Are you with the NSA and have a data warehouse of all the ticker tapes ever sent or something? [[Special:Contributions/162.158.166.185|162.158.166.185]] 12:45, 24 September 2022 (UTC)
:::::No, but my great grandparents thought ticker tape parades were littering, because Great Grandma worked in an office and Great Grandpa worked for sanitation, so we have a bunch of boxes in the attic filled with what she was supposed to throw out her window. [[Special:Contributions/172.71.158.75|172.71.158.75]] 12:54, 24 September 2022 (UTC)
:::That would be [https://www.jstor.org/stable/40697544 1299]. But I'm not sure how this is going to help us explain the comic, unless you perhaps are suggesting we enumerate date representation clusters somehow? [[Special:Contributions/172.69.22.61|172.69.22.61]] 12:32, 24 September 2022 (UTC)
:::: Someone should ask GPT-3 for a list of the top ten dates. [[Special:Contributions/172.69.22.71|172.69.22.71]] 12:59, 24 September 2022 (UTC)

Just putting January 2, 2006 here: https://stackoverflow.com/questions/20530327/origin-of-mon-jan-2-150405-mst-2006-in-golang [[Special:Contributions/172.69.22.161|172.69.22.161]] 12:28, 24 September 2022 (UTC)

Do we need a comment about how Pope Gregory XIII obliterated October 5th through 14th? [[Special:Contributions/172.71.158.25|172.71.158.25]] 13:30, 24 September 2022 (UTC)

Talk:2674: Everyday Carry

2022-09-20T15:59:55Z

172.69.134.131: Reply

Randall speaks: https://www.npr.org/2022/09/18/1123689628/randall-munroes-what-if-2-answers-the-absurd-science-questions-you-didnt-know-yo [[Special:Contributions/172.70.126.245|172.70.126.245]] 08:47, 20 September 2022 (UTC)

My EDC: Swiss Army Knife, Kershaw Leek, Kershaw Chive, SOG Multi Tool, Quarters for Aldi, Visa card.[[Special:Contributions/172.70.131.44|172.70.131.44]] 08:54, 20 September 2022 (UTC)

Hypothetical: Wandering around the jungle, as you do, you find an old, overgrown airstrip with abandoned planes strewn across its overgrown runway. Just then, several lost school parties show up out of nowhere, having miraculously survived various mid-sized airplane emergency landings in the vicinity. With one of the crowd's mobile phones (not you, because you haven't had a spare pocket for one since 2014, but you ''do'' have a USB recharge facility built into one of your many torches), and sufficient signal, the authorities are alerted to your exact position, and can send an airliner to rescue you all, just so long as the runway is cleared and you can set up enough lights to assist with their approach and landing. – Hopefully you hadn't a day or two ago had a wheel come off your cart (because of the jungle vines snagging at it!) and found that the screw that had come loose was of a size and configuration that ''none'' of your gadgets could actually handle. I mean, you had a darned star-penta-posix (with a hole in the end) for every other size from 1mm up to 35mm ''except this exact one''..! Amirite? [[Special:Contributions/141.101.107.159|141.101.107.159]] 08:58, 20 September 2022 (UTC)

Turns out, when the time you prepared for comes, you're still really unprepared. It's not just that you'll still only have two hands and will be missing that one "No need, you can find it anywhere"-Part. You also will have no idea that the time has come (after 213 false alarms) or which of the many, many things you prepared for has actually occurred. [[User:627235|627235]] ([[User talk:627235|talk]]) 13:04, 20 September 2022 (UTC)

There's only one true EDC: a bottle opener

Did this come out on Monday? I didn't see it around 1am ET Tuesday, and the bot didn't create the ExplainXKCD page until 05:47 [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 14:00, 20 September 2022 (UTC)
:No, Tuesday. Presumably the book tour has thrown the regular schedule off. [[Special:Contributions/172.69.134.131|172.69.134.131]] 15:59, 20 September 2022 (UTC)

Can someone familiar with EDC give some examples and explain why all of them might be so bulky please? [[Special:Contributions/172.69.134.131|172.69.134.131]] 15:59, 20 September 2022 (UTC)

Talk:2669: Things You Should Not Do

2022-09-13T02:39:24Z

172.69.134.131:

Seems like this could become a series. [[Special:Contributions/172.68.210.31|172.68.210.31]] 20:42, 7 September 2022 (UTC)
:Maybe. But I don't think it will be.
:What I was thinking was that there are clearly, on average, around 43  items per 'page', up to this point. This page shows only 19 items (both pre-New and New, or 20 if the "New" line counts as one, don't know if multilines reduce the number of numbered items ler page), so either it's been{{Citation needed}} manually split/new-paged (for changing aesthetics) or else it is highly varying according to the font-height/multiline-wrappings in use beforehand. Or perhaps we should expect around the same number of 'newer New' items to complete this page before the next page number is automatically started to be populated. [[Special:Contributions/172.70.86.8|172.70.86.8]] 20:53, 7 September 2022 (UTC)

<nowiki>#156,819</nowiki> looks like a reference to the Phineas and Ferb title sequence, and the episode Oil on Candace and probably more relevant here, What If 84. [[Special:Contributions/108.162.210.233|108.162.210.233]] 21:49, 7 September 2022 (UTC)
:It's true that it could be a reference, but I think I recognized most of the topics on the list as being mentioned in some what-if article from the archives--in the case of the "painting" one, https://what-if.xkcd.com/84/. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 22:13, 7 September 2022 (UTC)
::I think that given that he learned about this doing his new book that has not been released yet most of the new items would first be clear when we read the book, and hence all references to old what if seems moot to me... In my opinion it seems that those writing the current explanation failed to read this sentence: ''Updates to my "Things You Should Not Do" list, based on what I learned writing What If? 2''!!! Taking this into acount nothing on the new list should be from the old what if blog/book. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 13:02, 8 September 2022 (UTC)
:::Many articles (roughly half) in the blog were published after the first book was written, so those references are fair game to be included in the new book, and thus valid references as things he learned in the interim. Take a look at the archive thumbnails for the articles in 2014. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 22:52, 12 September 2022 (UTC)
:I think <nowiki>#156,819</nowiki> is confusing because [https://commons.wikimedia.org/wiki/File:Gustave_Guillaumet-_Le_Sahara.jpeg it's been done]. -- [[User:Ken g6|Ken g6]] ([[User talk:Ken g6|talk]]) 17:01, 8 September 2022 (UTC)

Added a transcript, hopefully it isn't too terrible. (also first explainxkcd edit!) [[User:Merrybot|Merrybot]] ([[User talk:Merrybot|talk]]) 21:52, 7 September 2022 (UTC)

Any indication what the purpose of the misspelling of ''hemorrhagic'' as ''*hemorraghic'' might be? XKCD is usually typo-free, which makes this look deliberate – but why? [[Special:Contributions/172.71.94.3|172.71.94.3]] 00:29, 8 September 2022 (UTC)
:I assume just a typo by Randall Munroe. [[Special:Contributions/172.70.214.79|172.70.214.79]] 02:11, 8 September 2022 (UTC)
::"XKCD is usually typo-free" Oh, they do happen every now and then but usually get corrected eventually by Randall. Nothing too special about this. [[User:Elektrizikekswerk|Elektrizikekswerk]] ([[User talk:Elektrizikekswerk|talk]]) 07:13, 8 September 2022 (UTC)

<nowiki>#156,823</nowiki> is actually a subplot in the movie "Only Lovers left Alive" by Jim Jarmusch. {{unsigned ip|162.158.129.163}}

<nowiki>#156,820</nowiki> sounds like a reference to TF2's "Meet the medic" which starts off with The Medic describing how he lost his medical licence by stealing a patient's skeleton {{unsigned ip|108.162.241.103}}

The last item suggests to me that the new book has a question like "What SPF would the sunscreen need to be if you were falling into the sun?" [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 21:10, 8 September 2022 (UTC)
: Far more than any SPF a terrestrial chemical could attain, which is essentially equivalent to positive infinity in this case, barring any new extremely unlikely physics. [[Special:Contributions/172.70.214.79|172.70.214.79]] 04:08, 9 September 2022 (UTC)

Is the maximum jail sentence for telling California poultry regulators that your farm is selling Pokemon eggs eight or ten years? [[Special:Contributions/172.70.206.213|172.70.206.213]] 04:06, 9 September 2022 (UTC)
:We need a more practical approach to legal advice. For example, how realistically could there ever be any compensatory damages? If a government regulator ever took a claim that a farm was producing Pokemon eggs seriously, wouldn't their liability for waste, fraud, and abuse damages to the public overwhelm that of the supposed defendant? [[Special:Contributions/172.70.206.163|172.70.206.163]] 07:26, 9 September 2022 (UTC)

I'm assuming that a Sun-sized ball of sunscreen would collapse and ignite as a star, right? [[Special:Contributions/172.70.214.79|172.70.214.79]] 10:47, 11 September 2022 (UTC)
#
156,816 was also in the news. I'm wondering if all of these up to some point have actually happened.

2671: Rotation

2022-09-13T00:46:37Z

172.69.134.131: The editor who added this asked if it was too much, and I think this, even less than they started with, is better

{{comic
| number = 2671
| date = September 12, 2022
| title = Rotation
| image = rotation.png
| titletext = It's okay, we can just feed the one-pixel image into an AI upscaler and recover the original image, or at least one that's just as cool.
}}

==Explanation==
{{incomplete|Created by an IMAGE UPSCALER. Do NOT delete this tag too soon.}}

Another of [[Randall]]'s [[:Category:Tips|Tips]], this tip claims that rotating a phone and taking a screenshot too many times will cause an image to disappear into nothingness, and warns the user against doing so. This is funny because while camera phone users are unlikely to do this, they are usually less aware of the optimal resolution for their intended purposes than they could be. A phone's "auto-rotation" feature will automatically rotate an image to the fit the phone's current orientation based on {{w|accelerometer}}s.

[[Image:World lines and world sheet.svg|thumb|200px|{{w|String theory}} describes the {{w|worldline}}s of point-like particles as {{w|worldsheet}}s of "closed strings," forming a topological foam.]]

For a fuller explanation of the concepts involved, including {{w|Planck units}}, often associated with the topological {{w|quantum foam}} of {{w|string theory}}, please see [https://www.youtube.com/watch?v=pUF5esTscZI this CGP Grey video.] For an explanation of topological string theory, see [[2658: Coffee Cup Holes]].

The title text refers to producing photographically likely higher resolution images from lower resolutions, an active area of current research.[https://openaccess.thecvf.com/content/ICCV2021/papers/Liang_Hierarchical_Conditional_Flow_A_Unified_Framework_for_Image_Super-Resolution_and_ICCV_2021_paper.pdf] Because reducing the resolution of an image is a lossy process results obtained through such processes will not be able to perfectly recreate the original. When scaled all the way down to one pixel, everything except a small amount of data about the image's overall color is lost, making reconstructing the original image impossible.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

[A phone in portrait orientation shows an image. It is then rotated, showing the image smaller with bars in landscape orientation, then the next phone is in portrait showing the entire screen of the previous rotated sideways, shrinking it every time. The labels show the decreasing size of the original image as it goes through successive rotations.]

"9 rotations: original image is smaller than a pixel."

"25 rotations: original image is smaller than an atom."

"101 rotations: original image is smaller than the Planck length, at which the concept of distance may break down."

Bottom caption: "Phone tip: don't rotate and screenshot an image too many times or it will become lost in the quantum foam of the universe."

{{comic discussion}}
[[Category:Tips]]
[[Category:Smartphones]]

Talk:2671: Rotation

2022-09-12T20:12:49Z

172.69.134.131: Uhhh

For extra credit: Waht is the resolution of the phone screen? [[Special:Contributions/172.71.94.135|172.71.94.135]] 18:59, 12 September 2022 (UTC)

:IMHO 400px. Note SMALLER. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 19:53, 12 September 2022 (UTC)

This seems like it could actually be really cool. Can anyone do this and put the picture here as an example? Also, if possible, include an AI upscale of the one pixel. [[Special:Contributions/172.69.90.83|172.69.90.83]] 19:07, 12 September 2022 (UTC)

There's a '''minor''' counting error: instead of pointing to the 9th rotation, the 'nine rotations' statement points to the 8th as the first phone has no rotations.[[Special:Contributions/172.70.90.77|172.70.90.77]] 19:10, 12 September 2022 (UTC)

Anyone getting a 404? Seems like the comic has disappeared. EDIT: ...aaaand it's back. [[Special:Contributions/172.70.100.54|172.70.100.54]] 19:34, 12 September 2022 (UTC)

Just putting https://www.codeguru.com/multimedia/rotate-a-bitmap-image/ here. [[Special:Contributions/172.69.134.131|172.69.134.131]] 20:12, 12 September 2022 (UTC)

26: Fourier

2022-08-29T23:23:49Z

172.69.134.131: /* Explanation */

{{comic
| number = 26
| date = November 2, 2005
| title = Fourier
| image = fourier.jpg
| titletext = That cat has some serious periodic components
}}

==Explanation==
{{incomplete|Missing reverse transform, or commentary on whether it is meaningful to take it. [to my uneducated eye it looks like half of a real-valued graph, but is phase information missing?] Missing guesses as to possible meaning of the name Elizabeth.}}

A {{w|Fourier transform}} is a mathematical function transformation often used in physics and engineering.

The theory is that any line graph can be represented as the sum of a bunch of sine waves of different frequencies and amplitudes. (The most obvious application is in analyzing a sound recording in terms of the different frequencies of sounds used.) So, for any line graph, you can produce another graph of the frequencies and their amplitudes. This can be done by evaluating an integral based on the function, which is referred to as "taking the Fourier transform" of the function. The form of the integral that needs to be taken is actually shown in the third line of the comic [[55: Useless]].

Unfortunately, Cueball has applied this "transform" to his cat. Indeed, whatever he has done has literally ''transformed'' his cat into the shape of an amplitude line graph. Although the cat seems to be alive and largely unharmed, it is clearly not in its familiar shape, and it is not clear if this condition is permanent or not. Notably, the fact the cat is still alive relates to an important property of Fourier transformation: the information of the original graph is fully preserved, and can even be reversed to produce the original graph. How a reverse Fourier transformation would apply to a transformed cat has yet to be seen.

"Periodic components" in the title text refers to the spikes in the graph. Because sine waves repeat themselves as you go along, the presence of large amounts of one particular sine wave in the Fourier transform graph (each spike) shows that the overall result (the initial graph) is likely to have parts that also repeat themselves, like a {{w|periodic function}}. In other words, the cat has repeating parts.

==Transcript==
:[Cueball talks on phone. A grotesque-looking cat with many sharp vertical points looks on.]
:Cueball: Hi, Dr. Elizabeth? Yeah, uh ... I accidentally took the Fourier transform of my cat...
:Cat: Meow!

==Trivia==
*This was the 27th comic originally posted to [[LiveJournal]].
**The previous was [[25: Barrel - Part 4]].
**The next was [[27: Meat Cereals]].
*Original title: "Wednesday's Drawing - Fourier"
*There were no original [[Randall]] quote for this comic.
*This comic was posted on [[xkcd]] when the web site opened on Sunday the 1st of January 2006.
**It was posted along [[:Category:First day on xkcd|with all 41 comics]] posted before that on LiveJournal as well as a few others.
**The latter explaining why the numbers of these 41 LiveJournal comics ranges from 1-44.
*One of the original drawings drawn on [[:Category:Checkered paper|checkered paper]].

{{comic discussion}}

[[Category:Comics posted on livejournal| 27]]
[[Category:First day on xkcd]]
[[Category:Checkered paper]]
[[Category:Comics with color]]
[[Category:Comics featuring Cueball]]
[[Category:Math]]
[[Category:Cats]]
[[Category:Comics with lowercase text]]

2657: Complex Vowels

2022-08-18T09:11:07Z

172.69.134.131: /* Explanation */

{{comic
| number = 2657
| date = August 10, 2022
| title = Complex Vowels
| image = complex_vowels.png
| titletext = Pronouncing [ṡṡċċḣḣẇẇȧȧ] is easy; you just say it like the 'x' in 'fire'.
}}

==Explanation==
{{incomplete|Created by a ROUNDED TONGUE. Do NOT delete this tag too soon.}}
[[File:Ipa-chart-vowels.png|thumb|200px|The standard IPA vowel chart.]]
This is another one of Randall's [[:Category:Tips|Tips]], this time a Linguistics Tip. The curly-haired linguist, [[:Category:Comics featuring Gretchen McCulloch|Gretchen McCulloch]], manages to produce a cursed sound using complex vowels, that cannot be comprehended by normal humans like [[Cueball]] and [[Megan]], who both seem to get a headache from listening to the sound. The sound she makes was produced by extending the IPA vowel plane along an imaginary axis to form the complex vowels.

In phonetics based on the {{w|International Phonetic Alphabet}} (IPA), the space of {{w|vocal tract}} articulators determining {{w|vowel}}s  is represented as three dimensional, from the position of the tongue and lips. The vertical axis represents vowel height or ''closedness'' (i.e., how close or far the tongue is from the top of the mouth), and the horizontal axis represents front-to-back ''place'' (i.e., how close or far the top of the tongue is from the teeth.) The position of the tongue, along with the frequency of the {{w|vocal cords}} vibrating in the larynx from air being exhaled by the diaphragm, are the primary determinants of the fundamental and second {{w|formant}} frequencies of vowel sounds. A third dimension of vowel sounds is the "roundedness" of the lips, represented on the IPA vowel chart to the right by pairs of vowel phoneme {{w|glyph}}s. Other higher-dimensional vowel representations include {{w|diphthong}}s, which are simply two different sequential vowels slurred together; diphones, which represent the last half of one phoneme followed by the first half of the next; {{w|vowel shift}} mappings delineating different accents[https://core.ac.uk/download/pdf/47086396.pdf][https://www.sciencedirect.com/science/article/abs/pii/S0095447010000562] and long-term evolution of voiced phone sounds; and {{w|cepstrum|cepstral}} representations such as {{w|Mel-frequency cepstrum|mel-frequency cepstral coefficients}}.

[[Randall]] suggests increasing the range of vowel sounds available by using complex notation to indicate an additional dimension with an "imaginary" axis. In mathematics, {{w|complex number}}s are numbers including both real numbers and {{w|imaginary number}}s. A complex number can be expressed as, "''a'' + ''b''i," where ''a'' and ''b'' are real numbers, but the latter imaginary part is combined with 'i,' the square root of negative one, as depicted in the central expression in the comic by √-1 indicating a further dimension of coordinates. When expanding the one-dimensional number line with an imaginary axis, it becomes two-dimensional with the "''b''i" component {{w|orthogonal}} to the original "real" number line. Linguists never use the {{w|complex plane}} to represent vowel roundedness or any other higher-dimensional features of phonemes, although the properties of complex numbers could conceivably support representing physiological features of the vocal tract, such as prior position of the articulators.{{cn}}

This comic conflates complex numbers in mathematics with "complex vowels" in linguistics. Such complex vowels are implied to create sounds which cannot be properly processed by the human brain, and represents one with the {{w|æ}} symbol surrounded by multiple {{w|diacritics}} akin to the {{w|Zalgo text}} meme. The sound of this supposedly alien vowel has Cueball and Megan clutching their heads in pain. Overall, Randall's complex vowels bear similarity to the cliché of "black speech" in {{w|Lovecraftian horror}}, a language created by alien beings with different vocal patterns than humans.

In linguistics, 'ə' is the {{w|schwa}} symbol, referred to in the title text and the depiction of complex phonemes, the most common vowel sound in English polysyllabic words (the 'a' in "comma" or the second 'e' in "letter.") Production of the schwa sound takes place with the tongue, jaw, and lips all in a relaxed, central position. The pronunciation of "[ṡṡċċḣḣẇẇȧȧ]" in the title text is said to sound like the x in the word fire. In fact, there is no x in fire.{{Citation needed}} This is perhaps in line with the idea that complex sounds are incomprehensible to most humans, and likely also impossible to pronounce by anyone other than experts such as [[:Category:Comics featuring Gretchen McCulloch|Gretchen]].

Another example of weird diacritics is in [[2619: Crêpe]], and with Zalgo text in [[1647: Diacritics]]. The use of typography to create psychological stress is explored in [[859: (]].

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[A diagram shows the extrusion of the trapezoidal IPA vowel chart upwards into three dimensions. A point near the center is labeled with an equation that shows "ə + ½√-1 " as being equivalent to a made-up symbol that looks like two schwas mirroring each other with other markings above and below.]

:[Below the diagram, a character with shoulder-length dark wavy hair pronounces the new vowel in a speech bubble with unstable lines surrounding it. Two bystanders to her right are bent over slightly, clutching their heads in apparent anguish.]

:[Caption below the panel:]
:Linguistics tip: Extend the IPA vowel plane along the imaginary axis to produce the ''complex vowels'', cursed sounds which the human mind cannot comprehend.

{{comic discussion}}
[[Category:Tips]]
[[Category:Language]]
[[Category:Comics featuring Gretchen McCulloch]]

2636: What If? 2 Countdown

2022-08-14T18:31:47Z

172.69.134.131: Calendar cat

{{comic
| number = 2636
| date = June 22, 2022
| title = What If? 2 Countdown
| image = what_if_2_countdown.png
| titletext = If you don't end the 99 Bottles of Beer recursion at N=0 it just becomes The Other Song That Never Ends.
}}

==Explanation==
{{incomplete|Created by FOUR SCORE AND 7 BOTTLES OF BEER ON THE WALL - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
This comic takes the idea of {{w|advent calendar}}s to the extreme. It uses absurd and obscure ways to measure the amount of time until [[Randall]]'s new book [https://xkcd.com/whatif2 ''What if? 2''] is released, with esoteric units and esoteric numbers. See explanation of each day in the [[#Table of the calendar countdown|table below]]

Some concepts that appear multiple times throughout the calendar are:

* '''{{w|SI prefixes}}''', which can be applied to the beginning of a unit's name to multiply or divide the unit by powers of 10 or 1,000. This is standard for units like meters and grams, but is rarely applied to measurements of time other than when a unit of less than one second is needed, most commonly in various fields of science and engineering such as physics and electronics.
* The '''{{w|Gettysburg Address}}''', a famous speech delivered by U.S. president Abraham Lincoln in 1863, where he began by referring to the signing of the Declaration of Independence taking place "four score and seven years ago". A score is a dated term for the number 20, so "four score and seven" is equivalent to 87.
* A '''dog year''' is traditionally considered to be one-seventh the length of a normal human year, since a dog's overall lifespan is roughly one-seventh of a typical human's. The comic applies this to other units of time, such as minutes and months, each of which is also one-seventh the length of the standard unit. The number 7 (traditionally a "lucky number") is also used in many of the numbers quoted in the calendar.
* Other comparative durations of time that are not normally or usefully applied to day-length multiples. At the top end, there is the age of the universe, at the other there is {{w|Planck units#Planck time|Planck-time}} – with entire durations of periods of human history and the time needed to watch popular TV/film franchises in-between – most of which require a non-trivial multiplier or divisor to bring them to the necessary scale required.
* A '''{{w|baker's dozen}}''' is 13, or one more than a normal dozen. Here, the "baker's" prefix can be applied to any unit by adding an extra one of its constituent parts, like an extra hour added to a day.
* '''{{w|Irrational numbers}}''' like {{w|pi}} (3.14159...), {{w|Euler's number}} or ''e'' (2.71828...), the {{w|golden ratio}} (1.61803...), and the {{w|square root of 2}} (1.41421...). These are all interesting numbers because of their mathematical properties, but very impractical to use as arbitrary measurements of time because they have an endless series of non-repeating decimal digits.
* The teenage dating game '''{{w|Seven minutes in heaven}}'''.
* Rotational and orbital periods of various bodies in the Solar System.
* Finally the song {{w|99 Bottles of Beer}}, is also used twice in the calendar, as the one after a full week and then for the final day before release.

The title text refers to the recursive time period on the final day before release, September 12 where the 99 Bottles of Beer song is song 99 times, but with one less verse every time (so 99 verses the first time, 98 verses the second, 97, 96 ... 2 and 1 the last). If you don't stop when you reach N=0 bottles, the repetition never ends, so that time interval becomes infinite. He then calls it "The Other Song That Never Ends", comparing it to ''{{w|The Song That Never Ends}}''. That song is a repetitive children's song, which is specifically intended to go on forever. The difference is that the Beer song has a natural stopping point at 0, while ''The Song That Never Ends'' is [https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwiEz97RkZT5AhXpk4sKHRs8C2wQyCl6BAgLEAM&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dxz6OGVCdov8&usg=AOvVaw2dI8Q_thXRfS6MUtq1NerU completely repetitive].

===Table of the calendar countdown===
{| class="wikitable"
|-
! Days left !! Date !! Duration specified !! Duration in days !! Explanation
|-
| 83 || Jun 22 || πe millidecades || 82.03 days || π ≈ 3.14159, ''e'' ≈ 2.718, so πe is about 22.459. A millidecade is 1/1,000 decade, or 1/100 year, or 3.652425 days. Multiplying these results is 82.03 days.
|-
| 82 || Jun 23 || 7 megaseconds || 81.02 days || 7,000,000 seconds. 60*60*24 = 86,400 seconds in a day, so 7,000,000/86,400 ≈ 81.02 days
|-
| 81 || Jun 24 || ''e'' lunar months || 80.27 days || A lunar month ≈ 29.53059 days, ''e'' ≈ 2.718, so 29.53059*2.718 ≈ 80.26 days.
|-
| 80 || Jun 25 || 60 rotations of Foucault's pendulum in Paris || 79.5 days || {{w|Foucault's pendulum}} demonstrates Earth's revolution, with the one at the latitude of Paris completing a full rotation every 31.8 hours.
|-
| 79 || Jun 26 || 8 milligenerations || 78.89 days || A generation is in general 22-33 years, with a reasonable mid-point of 27; and 8 x 0.001 (milli) x 365.2425 (accounting for leap years) x 27 ≈ 78.89 days
|-
| 78 || Jun 27 || 777,777 dog minutes || 77.16 days || A popular myth is that dogs age 7 times faster than humans, so 1 dog minute equals 1/7 human minutes.
|-
| 77 || Jun 28 || 7! episodes of ''Jeopardy!'' (skipping ads) || 77+ days || 7! = 7x6x5x4x3x2x1 = 5040. The standard episode of ''Jeopardy'' is 22-26 minutes, skipping ads. At 22 minutes each, the total is 110,880 minutes, or exactly 77 days.
|-
| 76 || Jun 29 || 5,000 repeats of ''99 Bottles of Beer'' || 76.39 days || Each verse of {{w|99 Bottles of Beer}} is "''N'' bottles of beer on the wall, ''N'' bottles of beer. Take one down, pass it around, ''N-1'' bottles of beer on the wall." The entire song contains 99 verses. Randall apparently sings this rather slowly at around 72 bpm, taking about 13 seconds per verse. It can be done somewhat faster as [https://www.youtube.com/watch?v=FITjBet3dio shown here], where the 99 verse takes less than 17.5 minutes for 10.6 second pr verse. Then it would only take two months, or 61 days, 15 days too little.
|-
| 75 || Jun 30 || 5 baker's fortnights || 75 days || A {{w|baker's dozen}} is a dozen (12) plus 1 extra item. Randall has generalized this to adding 1 to any unit. A fortnight is 14 days (or more properly "{{w|Fortnight|fourteen ''nights''}}", by its original use), so a baker's fortnight is 15 days. 5x15 is 75 days.
|-
| 74 || Jul 1 || √2 dog years || 73.79 days || See day 78 (Jun 27). 1.4142 × (365 / 7)
|-
| 73 || Jul 2 || π millivics (1/1000th of Queen Victoria's reign) || 72.97 days || {{w|Queen Victoria}} ruled between 20 June 1837 and 22 January 1901 (23,226 days). 3.14159 × 23.226
|-
| 72 || Jul 3 || 42 drives from NYC to LA (Google Maps estimate) || 71.75 days || According to Google Maps, the drive from New York City to Los Angeles via I-80 W (2789 miles or 4489 km) takes 41 hours.
|-
| 71 || Jul 4 || 1,000 viewings of ''Groundhog Day''|| 70.14 days || Using {{w|Groundhog Day (film)|Groundhog Day's}} 101-minute run time.
|-
| 70 || Jul 5 || 100,000 minutes || 69.44 days || 1,440 minutes per day
|-
| 69 || Jul 6 || 1/10th of Martian year || 68.70 Earth days || Martian sidereal and tropical years both round to 687.0 Earth days
|-
| 68 || Jul 7 || 1,234,567 sound-miles || 67.63 days || The speed of sound in air depends on the temperature. 15 °C or 59 °F gives the value 340 m/s and the travel time of 67.6349058 days.
|-
| 67 || Jul 8 || 2πe seconds || 66.74 days || 2^(π^e) = 5,766,073 seconds. The order of operations for multiple exponentiation without parentheses is top-first.
|-
| 66 || Jul 9 || 216 beats (Swatch Internet Time) || 65.54 days || A "{{w|.beat}}" is equal to 1/1,000 day.
|-
| 65 || Jul 10 || 1,000 ISS orbits || 64.58 days || Each orbit of the ISS takes 90-93 minutes. Here a value of 93 minutes is used.
|-
| 64 || Jul 11 || 🎵🎶🎵 Five hundred twenty five thousand (base seven) minutes|| 62.88 days || This refers to {{w|radix}}-7 arithmetic: 525,0007 minutes = 90,55210 minutes (5 × 7^5 + 2 × 7^4 + 5 × 7^3). Also references the opening and recurring line "Five hundred twenty five thousand six hundred minutes" from {{w|Seasons of Love}}, a song from the musical {{w|Rent (musical)|''Rent''}}, which is also referenced in [[1047: Approximations]]. "Base seven" has the same rhythm as "six hundred".
|-
| 63 || Jul 12 || 1050 Planck times || 62.38 days || 10^50 x 5.39 x 10^-44 seconds. The powers of 10 can be simplified, thus 5.39 × 10^6
|-
| 62 || Jul 13 || 4,000 episodes of ''The Office'' (skipping ads)|| 61.11 days || {{w|The Office (British TV series)|''The Office''}} was originally a {{w|BBC}} television show which had no commercial breaks, so Randall must be referring to the later {{w|The Office (American TV series)|US version}}, which is logical as he's American. This US "half-hour" comedy format contains 22 minutes of content (including the title sequence) and 8 minutes of ads. There are only 201 distinct episodes of the US version, so watching 4,000 episodes would require a lot of re-watching. 
|-
| 61 || Jul 14 || Four score and seven kilominutes || 60.42 days || 87 x 1,000 minutes
|-
| 60 || Jul 15 || 2 lunar months || 59.06 days || There are a number of different ways to define the {{w|lunar month}}. The most common is the synodic month, because it relates to the phases of the moon, and it's approximately 29.53 days.
|-
| 59 || Jul 16 || Half a day on Venus || 58.38 Earth days || A Venus synodic day is 116 days 18 hours.
|-
| 58 || Jul 17 || 5 megaseconds || 57.87 days || 5,000,000 seconds
|-
| 57 || Jul 18 || 30 microLits (1/1,000,000th of the time since the invention of literature) || 57.21 days || 3200 BCE is the approximate date of pre-Sumerian proto-writing as given in {{w|History of writing|Wikipedia's article on the history of writing}}. 5,222 years × 30 × 10-6.
|-
| 56 || Jul 19 || 1,000 viewings of ''Run Lola Run'' || 55.57 days || Using {{w|Run Lola Run|the movie's}} run time of 80 minutes.
|-
| 55 || Jul 20 || One million sound-miles || 54.78 days || The speed of sound in air depends on the temperature. 15 °C or 59 °F gives the value 340 m/s and the travel time of 54.7843137 days.
|-
| 54 || Jul 21 || 30 Ionian months || 53.07 Earth days || Orbital period of Io around Jupiter is approximately 1.77 days.
|-
| 53 || Jul 22 || One dog year || 52.18 days || See day 78 (Jun 27). 365.2425 / 7
|-
| 52 || Jul 23 || 60 viewings of ''Star Wars Episodes I-IX'' || 51.75 days || According to [https://dorksideoftheforce.com/2021/05/04/how-long-to-watch-every-star-wars-movie/ Fansided] the combined running times are 20 hours 42 minutes.
|-
| 51 || Jul 24 || 1/100,000,000,000th of the universe's age || 50.40 days || The universe is estimated to be about 13.8 billion years old.
|-
| 50 || Jul 25 || 5 milli-generations || 49.3 days || See day 79 (Jun 26). 5 × (27 × 365.2425) / 1,000
|-
| 49 || Jul 26 || 10,000 games of ''7 minutes in Heaven'' or 7 games of ''10,000 minutes in Heaven'' || 48.61 days || {{w|Seven minutes in heaven}} is an Anglo-culture teenager game, occuring in several movies. 10,000 minutes in Heaven is almost a week of making out (or doing whatever) in a closet, so this game is unlikely.{{citation needed}}
|-
| 48 || Jul 27 || φeπ minutes || 47.41 days || Phi (the golden ratio) to the power of e to the power of pi. 1.618 ^ (2.718 ^ 3.14159) = 68,284.14 minutes
|-
| 47 || Jul 28 || 4 megaseconds || 46.30 days || 4,000,000 seconds
|-
| 46 || Jul 29 || 216 minutes || 45.51 days || 65,536 minutes
|-
| 45 || Jul 30 || eee seconds || 44.15 days || 3,814,279.10 seconds
|-
| 44 || Jul 31 || π fortnights|| 43.98 days || 3.14159 x 14 days
|-
| 43 || Aug 1 || One devil's spacewalk (666 orbits of the ISS) || 43.01 days || See day 65 (Jul 10). 666 is the {{w|number of the beast}}. 666 × 93 minutes.
|-
| 42 || Aug 2 || 1 kilowatt-hour per watt || 41.66 days || 1,000 hours
|-
| 41 || Aug 3 || eπ Ionian months || 40.93 Earth days || Orbital period of Io around Jupiter is 1.769137786 days. 2.718 ^ 3.14159 × 1.769
|-
| 40 || Aug 4 || 30 rotations of Foucault's pendulum in Paris || 39.75 days || Refer to Day 80 (Jun 25). Half of that day's 79.5 days, 30 × 31.8 hours 
|-
| 39 || Aug 5 || ''e'' fortnights || 38.06 days ||2.71828 x 14 days
|-
| 38 || Aug 6 || ππ baker's days (25 hours) || 37.98 days || See day 75 (Jun 30). 3.14159 ^ 3.14159 × 25 
|-
| 37 || Aug 7 || One deciyear || 36.52 days || One tenth of one year
|-
| 36 || Aug 8 || 7! milliweeks || 35.28 days || 5,040 × 0.001 weeks
|-
| 35 || Aug 9 || 100,000 plays of the ''Jeopardy!'' "Think" music || 34.72 days || ''Think'' is the music played while the contestants try to answer the Final Jeopardy question; it is 30 seconds long. 30 × 100,000 seconds
|-
| 34 || Aug 10 || 1000 basketball games (game time) || 33.33 days || Uses the NBA game time of four 12-minute quarters, or 48 minutes
|-
| 33 || Aug 11 || 777 hours || 32.38 days || 24 hours per day
|-
| 32 || Aug 12 || One millilincoln (1/1000 of fourscore and seven years) || 31.78 days || {{w|Abraham Lincoln}}'s {{w|Gettysburg Address}} begins with the famous phrase "Four score and seven years ago". 1 score = twenty. 
|-
| 31 || Aug 13 || 1,000 episodes of ''60 Minutes'' (skipping ads) || 30.56 days || A television hour is between 42 and 44 minutes, with the remaining time used by ads. This uses a television 'hour' containing 44 minutes of content and 16 minutes of ads.
|-
| 30 || Aug 14 || All of ''Star Trek'', consecutively || 28.55 days || According to [https://www.cbr.com/star-trek-every-tv-episode-movie/ CBR]  of January 21, 2021, this consists of: 

:3,950 minutes of The Original Series
:528 minutes of The Animated Series
:7,832 minutes of The Next Generation
:7,920 minutes of Deep Space 9
:7,740 minutes of Voyager
:4,116 minutes of Enterprise
:250 minutes of Lower Decks, Season 1
:4,725 minutes of Discovery, the first 3 seasons
:450 minutes of Picard, Season 1
:150 minutes of Short Treks
:688 minutes of The Original Series films
:448 minutes of The Next Generation films
:381 minutes of the three rebooted universe films

In the year and a half since publishing that article, there have been:
:250 minutes of Lower Decks, Season 2
:585 minutes of Discovery, Season 4
:450 minutes of Picard, Season 2
:237 minutes of Prodigy, the first half of Season 1
:411 minutes of Strange New Worlds, up to and including June 23, the day after this comic (8 out of 10 episodes of Season 1)

The last two episodes of Strange New Worlds, taking the franchise until July 7, would add roughly 0.035 of a day each, a negligible difference.
|-
| 29 || Aug 15 || 777,777 nanocenturies || 28.41 days || 777,777 × 10-9 × 100 years.
|-
| 28 || Aug 16 || One sidereal lunar month || 27.3 days || The time it takes moon to return to the same position relative to the fixed stars
|-
| 27 || Aug 17 || 6 dog months || 26.14 days || See day 78 (Jun 27). A month averaging 30.5 days / 7
|-
| 26 || Aug 18 || ππ kilominutes || 25.32 days || 36,462.16 minutes
|-
| 25 || Aug 19 || 7 games of 7! minutes in Heaven || 24.5 days || 7 x 5040 (7 {{w|factorial}}) minutes. See also day 49 (Jul 26).
|-
| 24 || Aug 20 || 50 viewings of the extended ''Lord of the Rings'' trilogy || 23.82 days || ''The Fellowship of the Ring'' extended version is 208 minutes, ''The Two Towers'' is 226 minutes, and ''The Return of the King'' is 252 minutes for its extended version, according to [https://fictionhorizon.com/how-long-are-all-the-lord-of-the-rings-and-the-hobbit-movies-combined/ FictionHorizon.com]
|-
| 23 || Aug 21 || A drive from NYC to LA where you keep remembering new things you forgot and have to go back 6 times || 22.21 days || See day 72 (Jul 3). This is for 6 round-trips and 1 one-way trip, so 13 trips at 41 hours each.
|-
| 22 || Aug 22 || ''It's a Small World'' sung at 1/10,000th speed || 21.06 days || {{w|It's a Small World}} is a song that was composed for the attraction of the same name at various {{w|Disney}} theme parks, and plays continuously at them in various languages. The song lasts 12–15 minutes, depending on the language. However, Randall seems to be using a single iteration of the song, such as this [https://www.youtube.com/watch?v=PxRW-duSCLA YouTube video] of 3:02, posted by Disney themselves. As such, it's 3:02 × 10,000.
|-
| 21 || Aug 23 || 500 hours || 20.83 days || 24 hours per day, or 0.041678 days per hour
|-
| 20 || Aug 24 || √2 fortnights || 19.80 days || 1.4142 × 14 days
|-
| 19 || Aug 25 || Time it would take Vanessa Carlton to walk 1,000 miles || 18.94 days || {{w|Vanessa Carlton}} is an American singer, and {{w|A Thousand Miles}} is her most successful song. Randall estimates her walking speed at about 2.2 miles/hour. 
|-
| 18 || Aug 26 || 100,000 breaths || 17.36* days || The normal respiratory rate for adults is typically 12-20 breaths per minute, or about 3-5 seconds each. *However, the day length here is for 15 seconds/breath, so Randall may be a practitioner of [https://www.frontiersin.org/articles/10.3389/fnhum.2018.00353/full slow breathing].
|-
| 17 || Aug 27 || √2 megaseconds || 16.37 days || 1.4142 × 1,000,000 seconds
|-
| 16 || Aug 28 || πππ πcoseconds || 15.51 days || 1.3402 × 1018 picoseconds (i.e., 10-12 seconds), making a joke how the mathematical "pi" is written with the character "π" by using it to spell "picoseconds".
|-
| 15 || Aug 29 || One baker's fortnight (15 days) || 15 days || See day 75 (Jun 30)
|-
| 14 || Aug 30 || One baker's dozen (13) baker's days (25 hours) || 13.54 days || 325 hours; see day 75 (Jun 30)
|-
| 13 || Aug 31 || 300 hours || 12.5 days || 0.041678 days per hour
|-
| 12 || Sep 1 || One million seconds || 11.57 days || 86,400 seconds per day
|-
| 11 || Sep 2 || One nonstop bike ride from NYC to LA || 10.54 days || Google maps estimates the trip at 253 hours
|-
| 10 || Sep 3 || 1⁄1,000th of a generation || 9.86 days || See day 79 (Jun 26). A generation being taken as 27 years.
|-
| 9 || Sep 4 || 777,777 seconds || 9.002 days || 1.15741 × 10-5 days per second
|-
| 8 || Sep 5 || 100 viewings of ''Groundhog Day'' || 7.01 days || See Day 71 (Jul 4).
|-
| 7 || Sep 6 || 100 games of ''Lincoln Kissing'' (fourscore and seven minutes in Heaven) || 6.04 days || 8,700 minutes
|-
| 6 || Sep 7 || One pico-universe-lifetime || 5.04 days || See Day 51 (Jul 24). 13.8 billion years, "pico" meaning to divide by 1 trillion, thus 13.8 / 1,000.
|-
| 5 || Sep 8 || The ''Baby Shark'' chorus for a family of 50,000 sharks || 4.63 days || The chorus lasts about 8 seconds per 'person'
|-
| 4 || Sep 9 || One centiyear || 3.65 days || 365.2425 days/100
|-
| 3 || Sep 10 || Cyndi Lauper's ''Time After Time'' played 1,000 times || 2.79 days || Based on a length of 4 minutes, 1 second
|-
| 2 || Sep 11 || ''Speed'' (1994) played at one frame per second || 1.9 days || {{w|Speed (1994 film)}} has runtime of 116 minutes = 6,960 seconds = 167,040 film frames at standard frame rate of 24 frames/second
|-
| 1 || Sep 12 || F(99) where F(N) means sing all the verses of ''N Bottles of Beer On the wall'' followed by F(N-1) || 0.76 days || Each iteration contains ''N'' verses. ''N + N-1 + N-2 ... + 1'' equals ''N * (N+1) / 2'', so 99 recursions = 4950 verses. Using the same 13-second (72 bpm) rate as Jun 29, this is close to 18 hours. Probably refers to Donald Knuth's article {{w|The Complexity of Songs}}. This F(99) song is referenced in the title text, see [[#Explanation|explanation]] above.
|-
| 0 || Sep 13 || ''What If? 2'' release day || N/A ||
|}

==Transcript==
:[The comic is a calendar that counts down to a specific date, like a Christmas calendar starting on June 2nd and ending on September 13th (2022). Each day is represented with a large square and there are 12 weeks for a total of 84 days. The days belonging to a particular month are surrounded by a thicker frame than between days from the same month. The first day of each month plus the very first day has the date given with three letters representing the month and the day number. This is written at the top right corner in a frame. All other days either only have the number for the day in the frame in the corner, or, if there are too much text on the day, no number is written. A single day has the number without the frame around it. The very last day notes what the countdown is for and there are three large stars places around the text, as well as smaller and larger dots, likely representing more stars in the entire field. All other days have text on all white background. The text represents a time that fits the time span from that day until the final day. The calendar begins on a Wednesday, and since the calendar week begins on a Sunday, there are three days missing to the left in the top row, and similarly four days are missing to the right in the bottom row as the last day is a Tuesday. Above the calendar is a large heading with a subheading below:]
:<big>Countdown to ''What if? 2''</big>
:(Preorder at xkcd.com/whatif2 to get it at the end of the countdown)

:[The date given, either as written or else mentioned if not written in comic, and then follows the text on that day:]
:Jun 22
::πe millidecades
:23
::7 megaseconds
:24
::e lunar months
:25
::60 rotations of Foucault's pendulum in Paris
:26
::8 milligenerations
:27
::777,777 dog minutes
:28
::7! episodes of ''Jeopardy!'' (skipping ads)
:29
::5,000 repeats of ''99 Bottles of Beer''
:30
::5 baker's fortnights (15 days)
:Jul 1
::√2 dog years
:[Date left out on the 2nd.]
::π millivics (1/1000th of Queen Victoria's reign)
:3
::42 drives from NYC to LA (Google Maps estimate)
:4
::1,000 viewings of ''Groundhog Day''
:5
::100,000 minutes
:6
::1/10th of Martian year
:7
::1,234,567 sound-miles
:8
::2πe seconds
:9
::216 beats (Swatch Internet Time)
:10
::1,000 ISS orbits
:11
::[Four musical notes are shown at the top.]
::Five hundred twenty five thousand (base seven) minutes
:12
::1050 Planck times
:13
::4,000 episodes of ''The Office'' (skipping ads)
:14
::Four score and seven kilominutes
:15
::2 lunar months
:16
::Half a day on Venus
:17
::5 megaseconds
:[Date left out on the 18th.]
::30 microLits (1/1,000,000th of the time since the invention of literature)
:19
::1,000 viewings of ''Run Lola Run''
:20
::One million sound-miles
:21
::30 Ionian months
:22
::One dog year
:23
::60 viewings of ''Star Wars Episodes I-IX''
:24
::1/100,000,000,000th of the universe's age
:25
::5 milli-generations
:[Date left out on the 26th.]
::10,000 games of ''7 minutes in Heaven'' or 7 games of ''10,000 minutes in Heaven''
:27
::φeπ minutes
:28
::4 megaseconds
:29
::216 minutes
:30
::eee seconds
:31
::π fortnights
:Aug 1
::One devil's spacewalk (666 orbits of the ISS)
:2
::1 kilowatt-hour per watt
:3
::eπ Ionian months
:4
::30 rotations of Foucault's pendulum in Paris
:5
::e fortnights
:6
::πe baker's days (25 hours)
:7
::One deciyear
:8
::7! milliweeks
:9
::100,000 plays of the ''Jeopardy!'' "Think" music
:10
::1000 basketball games (game time)
:11
::777 hours
:12
::One millilincoln (1/1000 of fourscore and seven years)
:13
::1,000 episodes of ''60 Minutes'' (skipping ads)
:14
::All of ''Star Trek'', consecutively
:15
::777,777 nanocenturies
:16
::One sidereal lunar month
:17
::6 dog months
:18
::ππ kilominutes
:19
::7 games of ''7! minutes in Heaven''
:20
::50 viewings of the extended ''Lord of the Rings'' trilogy
:[Date left out on the 21th.]
::A drive from NYC to LA where you keep remembering new things you forgot and have to go back 6 times
:[Date left out on the 22nd.]
::''It's a Small World'' sung at 1/10,000th speed
:23
::500 hours
:24
::√2 fortnights
:[Date left out on the 25th.]
::Time it would take Vanessa Carlton to walk 1,000 miles
:26
::100,000 breaths
:27
::√2 megaseconds
:28
::πππ πcoseconds
:29 [The date is not inside a small frame as all other dates shown.]
::One baker's fortnight (15 days)
:30
::One baker's dozen (13) baker's days (25 hours)
:31
::300 hours
:Sep 1
::One million seconds
:2
::One nonstop bike ride from NYC to LA
:3
::1⁄1,000th of a generation
:4
::777,777 seconds
:5
::100 viewings of ''Groundhog Day''
:[Date left out on the 6th.]
::100 games of ''Lincoln Kissing'' (Fourscore and seven minutes in Heaven)
:7
::One pico-universe-lifetime
:8
::The ''Baby Shark'' chorus for a family of 50,000 sharks
:9
::One centiyear
:10
::Cyndi Lauper's ''Time After Time'' played 1,000 times
:[Date left out on the 11th.]
::''Speed'' (1994) played at one frame per second
:[Date left out on the 12th.]
::F(99) where F(N) means sing all the verses of ''N Bottles of Beer On the wall'' followed by F(N-1)
:13
::''What If? 2'' release day

{{comic discussion}}

[[Category:Book promotion]]
[[Category:Comics featuring real people]]
[[Category:Songs]]
[[Category:Animals]]
[[Category:Time]]

2658: Coffee Cup Holes

2022-08-13T22:38:07Z

172.69.134.131: in

{{comic
| number = 2658
| date = August 12, 2022
| title = Coffee Cup Holes
| image = coffee_cup_holes.png
| titletext = Theoretical physicist: At the Planck length, uncountably many.
}}

==Explanation==
{{incomplete|Created by a CAFFEINE MOLECULE WITH A HOLE DRILLED IN ITS SIDE. Do NOT delete this tag too soon.}}

This comic depicts people in different fields of study answering the question, "How many holes are there in a coffee cup?" This question can have multiple interpretations, in particular concerning the definition of a hole.

[[File:Mug and Torus morph.gif|thumb|200px|The coffee mug and donut shown in this animation both have topological genus one.]]

[[Ponytail]], a {{w|topology|topologist}}, states the coffee cup belongs in the {{w|Genus (mathematics)#Topology|genus}} of one hole. A common joke is that topologists can't tell the difference between a coffee cup and a donut since they're homeomorphic to each other — they have the same genus. From the topologist's point of view, the coffee cup definitely has one hole. See [[2625: Field Topology]] for more information about topology.

[[Hairy]], a normal person, is not sure (the acronym "IDK" stands for "I don't know") and asks for clarification about whether the opening at the top counts as a hole. This shows flaws in the question, which suffers from the mathematically imprecise, ambiguous common usage of the word hole. Topologists would refer to the opening as a concavity, not a hole, and while they consider such geometrical properties generally outside their field, most practical applications of topology do involve geometric components.

[[File:Double torus illustration.png|thumb|left|200px|A genus two surface]]

[[Hairbun]], a philosopher, answers the question with an elucidating counter-question, considering a hypothetical scenario. Drilling a new hole should increase the number of holes by one, and after the hole has been drilled, a common teacup or mug has two holes according to topologists. Since drilling a hole increases the number of holes by one,{{cn}} the philosopher's question requires the original questioner to reveal the answer to their own question. (Also, she asks how many holes there are ''now'' rather than ''after we do that'', an ambiguity.)

[[Image:Point cloud torus.gif|thumb|200px|A point cloud of a genus one surface]]

[[Cueball]], a chemist, looks at the coffee in the cup on a molecular level, which means it has very many holes: 1,000,000,000,000,000,000,000 (1021 or 1 sextillion) “in the [https://chemapps.stolaf.edu/jmol/jmol.php?model=CN1C%3DNC2%3DC1C%28%3DO%29N%28C%28%3DO%29N2C%29C caffeine] alone.” One molecule of caffeine has two rings of bonds with holes in them, so Cueball is talking about 500 quintillion molecules. As the molecular mass of {{w|caffeine}} is about 194 grams per mole, that many weigh 161 milligrams, a typical amount in a cup. The coffee could have other holes, depending on the type of coffee; for example, espresso contains significant amounts of niacin and riboflavin, which have one and three rings in their chemical structure, respectively. Bonds are not discrete sticks as portrayed in many molecular models. However, the "holes" in the middle of a molecule's rings are not completely empty but instead merely have lower electron probability density through the middle than other parts of the bonds. So the point-cloud duality of {{w|Bonding molecular orbital|electron orbitals and bonds}} might not satisfy a topologist's, normal person's, or philosopher's criteria for a connected substrate in which holes may be formed.

[[Image:World lines and world sheet.svg|left|thumb|200px|{{w|String theory}} describes the {{w|worldline}}s of point-like particles as {{w|worldsheet}}s of "closed strings," forming topological holes.]]

In the title text, a theoretical physicist looks even deeper, at the subatomic scale of {{w|Planck units}}. Since fundamental particle interaction is governed by fundamental forces and collision (per the {{w|Pauli exclusion principle}}) instead of tensile or ductile solid connectedness, the theoretical physicist posits that any definition providing for a single hole would also describe a number of holes akin to the factorial of the number of particles in the universe, or at least within the cup's {{w|light cone}}, which is a number impractical to accurately count, but not uncountable in a mathematical sense.

Part of the joke could be that all five methods of inquiry don't discern between a {{w|cup}} (as described) and a {{w|mug}} (as depicted), the cliché being that topologists are unusual because they don't. Or, as many people use the terms interchangeably, [[Randall]] may too.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[The first panel has text only. The "Q:" below is a large letter Q representing a question, not a character name.]
:Q:
:How many holes are there in a coffee cup?

:[Each of the next four panels has a caption at the top to indicate the kind of person answering the question.]
:Caption: Topologist
:[Ponytail stands holding a coffee mug.]
:Ponytail: One.

:Caption: Normal person
:[Hairy stands to the right of Ponytail, holding a coffee mug at an angle to look into it.]
:Hairy: IDK, does the opening count as a hole?

:Caption: Philosopher
:[Hairbun is shown in closeup, with two drawings of coffee mugs to her left.]
:Hairbun: To answer that question, consider another: If we drill a hole in the side, how many holes are there now?

:Caption: Chemist
:[Cueball stands with a drawing of a caffeine molecule above him and to the right.]
:Cueball: 1021 in the caffeine alone

{{comic discussion}}
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Hairy]]
[[Category:Comics featuring Hairbun]]
[[Category:Comics featuring Cueball]]
[[Category:Math]]
[[Category:Food]]
[[Category:Science]]
[[Category:Chemistry]]
[[Category:Philosophy]]

2658: Coffee Cup Holes

2022-08-13T05:04:06Z

172.69.134.131: /* Explanation */ COFFEE cup and coffie mug are the same thing

{{comic
| number = 2658
| date = August 12, 2022
| title = Coffee Cup Holes
| image = coffee_cup_holes.png
| titletext = Theoretical physicist: At the Planck length, uncountably many.
}}

==Explanation==
{{incomplete|Created by a CAFFEINE MOLECULE WITH A HOLE DRILLED IN ITS SIDE. Do NOT delete this tag too soon.}}

[[File:Mug and Torus morph.gif|thumb|150px|The coffee cup and donut shown in this animation both have genus one.]]

This comic depicts people in different fields of study answering the question, "How many holes are there in a coffee cup?" This question can have multiple interpretations, in particular concerning the definition of a hole.

[[Ponytail]], a {{w|topology|topologist}}, states the coffee cup belongs in the {{w|Genus (mathematics)#Topology|genus}} of one hole. A common joke is that topologists can't tell the difference between a coffee cup and a donut since they're homeomorphic to each other — they have the same genus.  From the topologist's point of view, the coffee cup definitely has one hole. See [[2625: Field Topology]] for more information about topology.

[[Hairy]], a normal person, asks for clarification about whether the opening at the top counts as a hole. This shows flaws in the question, which suffers from the mathematically imprecise, ambiguous common usage of the word hole. Topologists would refer to the opening as a concavity, not a hole, and while they consider such geometrical properties generally outside their field, most practical applications of topolgy do involve geometrical components.{{cn}}

[[File:Double torus illustration.png|thumb|150px|A genus-2 surface]]

[[Hairbun]], a philosopher, answers the question with an elucidating counter-question, considering a hypothetical scenario. Drilling a new hole should increase the number of holes by one, and after the hole has been drilled, a common teacup or mug has two holes according to topologists. Since drilling a hole increases the number of holes by one, the philosopher's question requires the original questioner to reveal the answer to their own question.

[[Cueball]], a chemist, looks at the cup on a molecular level, which naturally means it has lots and lots of holes: 1,000,000,000,000,000,000,000 (1021 or 1 sextillion) “in the [https://chemapps.stolaf.edu/jmol/jmol.php?model=CN1C%3DNC2%3DC1C%28%3DO%29N%28C%28%3DO%29N2C%29C caffeine] alone.” The implication is that there are more in the cup itself, depending on what material it’s made out of. Also, the coffee itself could have other holes, depending on the type of coffee. For example, espresso contains significant amounts of niacin and riboflavin, each of which has at least one hole in its chemical structure. However, this ignores the fact that bonds are not discrete sticks as portrayed in many molecular models. The "holes" in the middle of a caffeine molecule are not completely empty but instead merely have lower electron densities/probabilities. In a {{w|space-filling model}}, a caffeine molecule has zero holes. So the point-cloud duality of electron orbitals and bonds might not satisfy a topologist's, normal person's, or philosopher's criteria for a connected substrate in which holes may be formed.

In the title text, the theoretical physicist looks even deeper, at a subatomic level. Since fundamental particle interaction is governed by fundamental forces and collision instead of tensile or ductile solid connectedness, the theoretical physicist posits that any definition providing for a single hole would also describe a number of holes akin to the factorial of the number of particles in the universe, or at least within the cup's {{w|light cone}}, which is a number impractical to accurately count, but not uncountable in a mathematical sense.

==Transcript==
{{incomplete transcript|Do NOT delete this tag too soon.}}

:[The first panel has text only. The "Q:" below is a large letter Q representing a question, not a character name.]
:Q:
:How many holes are there in a coffee cup?

:[Each of the next four panels has a caption at the top to indicate the kind of person answering the question.]
:Caption: Topologist
:[Ponytail stands holding a coffee mug.]
:Ponytail: One.

:Caption: Normal person
:[Hairy stands to the right of Ponytail, holding a coffee mug at an angle to look into it.]
:Hairy: IDK, does the opening count as a hole?

:Caption: Philosopher
:[Hairbun is shown in closeup, with two drawings of coffee mugs to her left.]
:Hairbun: To answer that question, consider another: If we drill a hole in the side, how many holes are there now?

:Caption: Chemist
:[Cueball stands with a drawing of a caffeine molecule above him and to the right.]
:Cueball: 1021 in the caffeine alone

{{comic discussion}}
[[Category:Comics featuring Ponytail]]
[[Category:Comics featuring Hairy]]
[[Category:Comics featuring Hairbun]]
[[Category:Comics featuring Cueball]]
[[Category:Math]]
[[Category:Food]]
[[Category:Science]]
[[Category:Chemistry]]
[[Category:Philosophy]]

Talk:2653: Omnitaur

2022-08-02T21:47:46Z

172.69.134.131: Crispr

Do people thing Omnitaur meant to be a anagram? It would make more sense to me suffix taken from minotaur and centaur etc. with the prefix omni meaning all.
[[User:Mouse|Mouse]] ([[User talk:Mouse|talk]]) Mousetail
:I don't think it is meant to be an anagram. Nevertheless it is one. But that's just my gut feeling. [[User:Elektrizikekswerk|Elektrizikekswerk]] ([[User talk:Elektrizikekswerk|talk]]) 07:07, 2 August 2022 (UTC)
::There are only those two taurs mentioned and there are many other creatures made from animals with different name. It has both human and bull in it (I know it has all the others as well), but to me it seems obvious that Randall is aware this is an anagram of Mino to Omni. And then of course it encompasses most other mythical creatures, given the meanin of Omni. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 08:16, 2 August 2022 (UTC)
:::True, surely he's aware of it. My point is: It's either an anagram that also happens to have the meaning "omni" or it has the meaning "omni" and also happens to be an anagram. My bet is on the latter. [[User:Elektrizikekswerk|Elektrizikekswerk]] ([[User talk:Elektrizikekswerk|talk]]) 10:42, 2 August 2022 (UTC)

I dread to think what this thing must look like internally. Especially when I remember the centaurs from C S Lewis' 'Narnia' stories, who are depicted eating two meals - a huge roast meal "to satisfy the man stomach" and a meal of grass "to satisfy the horse stomach". Bleagh.[[User:MarquisOfCarrabass|MarquisOfCarrabass]] ([[User talk:MarquisOfCarrabass|talk]]) 07:32, 2 August 2022 (UTC)
:Well it certainly is an Omnivore (does that mean eating only Omnitaurs then...? :-D ) --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 08:16, 2 August 2022 (UTC)

If we take a looser definition of 'omnitaur' as meaning 'made of lots of different creatures' (in parallel to how 'omnivore' really means 'eats lots of different things' rather than literally 'eats everything', and in line with only 11 creatures being depicted), then arguably every creature is an omnitaur - it's just that most of them are special cases that happen to be made up of a lot of very similar creatures. [[Special:Contributions/172.70.162.77|172.70.162.77]] 09:15, 2 August 2022 (UTC)

1/121 seems nonsense to me. Assume this omnitaur has fairly standard genetics: 11 allele pairs for the several body parts with recessivity being random. All parts must have one human allele (which happens to be recessive), 1/11^10. The human allele must be picked, 1/2^11. More like a trillion chance... [[Special:Contributions/172.71.98.193|172.71.98.193]] 10:10, 2 August 2022 (UTC)
:I was just going to post a question: why not (1/11)11? [[Special:Contributions/172.70.214.43|172.70.214.43]] 10:20, 2 August 2022 (UTC)
:How on earth is that "standard"? [[User:Nitpicking|Nitpicking]] ([[User talk:Nitpicking|talk]]) 11:32, 2 August 2022 (UTC)
::Sitting in a rejected-edits file of mine (because I couldn't see how to make it good enough to escape a general nitpick... though not your presence in particular) is the following, that might have been superceded by the Speculations section that was added since:
:::''In order for two omnitaur genomes to contain the ''possibility'' of merging to create a full human, maybe the genetic material is not {{w|Ploidy#Diploid|diploid}}, but {{w|Polyploidy|undecaploid}} (at the very least), leading to each omnitaur to express their own individual and personal distribution of phenotypes from amongst the many heritable traits they have inherited. The reproductive compatibility of any two omnitaurs would be a crap-shoot and might influence what given 'monotaurism' might arise by chance.''
::...be a shame to waste it, but it doesn't really fit as is now, even if I 'correct' it. [[Special:Contributions/108.162.229.75|108.162.229.75]] 15:06, 2 August 2022 (UTC)
::::I agree it shows promise. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 17:42, 2 August 2022 (UTC)

Note: you can't call sharks "fish" without also calling humans, frogs, and eagles "fish" (if you're using the current taxonomic system based on cladistics). The cartilaginous fishes split from bony fishes long before the tetrapods like us split off from the lineage that became trout, flounder, and guppies. That is, a snake is much more closely related to a grouper than a shark is. [[User:Nitpicking|Nitpicking]] ([[User talk:Nitpicking|talk]]) 11:32, 2 August 2022 (UTC)
: [https://www.nbcnews.com/think/opinion/california-court-ruling-bees-are-fish-bad-logic-good-humans-rcna32971 According to California courts, bees are fish.] (Spoiler: within the meaning of "Fish and Game" or something like that. Personally I think the judges were trolling because they could have more congruously gone with "game" because it was about honeybees which beekeepers obviously catch.) [[Special:Contributions/172.70.206.95|172.70.206.95]] 13:42, 2 August 2022 (UTC)
:Fish are a paraphyletic group, but that doesn't make the group "wrong" by cladistics. Cladistics recognizes that its common for one branch of a group to go off and do something very divergent, and that the remaining members often have a lot of shared characteristics that make it useful to talk about them. For example, "stem mammals", which excludes actual mammals. Cladistics has stronger objections to polyphyly, which is grouping animals together that aren't a cladistic group with some very clear exceptions. It still recognizes the groups though, classifying them as polyphyletic groups. [[Special:Contributions/172.71.82.121|172.71.82.121]] 13:47, 2 August 2022 (UTC)
::^ This editor paraphyletizes. [[Special:Contributions/172.70.206.163|172.70.206.163]] 14:40, 2 August 2022 (UTC)

The speculation section needs a discussion of how living {{w|turducken}} could be engineered. [[Special:Contributions/172.70.211.88|172.70.211.88]] 11:44, 2 August 2022 (UTC)
:Being able to do that would be a great lab qual, but when the spacefairing dinosaurs find out we use them for the culinary arts, is there any hope for galactic peace? [[Special:Contributions/172.70.210.145|172.70.210.145]] 16:15, 2 August 2022 (UTC)
::Considering there wasn't any hope for galactic peace before either, I think it's worth the try. Seriously, even if humans would be the ONLY spacefairing species there would be no hope for galactic peace. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 20:09, 2 August 2022 (UTC)

If HGTTG references are traditional here, ''The Restaurant at the End of the Universe'' had a pig with the mind and vocal tract of a human so it could articulate how much it wanted to be eaten. [[Special:Contributions/172.70.211.90|172.70.211.90]] 16:19, 2 August 2022 (UTC)
:The description of the "Dish of the Day" was that it was bred as "an animal that actually wanted to be eaten and was capable of saying so clearly and distinctly". There's no mention of it being a pig with the mind and vocal tract of a human, or in any other way a chimera. Its species is "Ameglian Major Cow". I'm also not convinced that cyborgs count as chimerae. [[User:BunsenH|BunsenH]] ([[User talk:BunsenH|talk]]) 18:12, 2 August 2022 (UTC)
::It was depicted with pig ears and nose in one of the video adaptations. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 18:37, 2 August 2022 (UTC)

whose sona is this 🤨 --[[Special:Contributions/172.70.110.113|172.70.110.113]] 16:25, 2 August 2022 (UTC)
:Are you asking whether omnitaurs make good clerics in D&D? [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 16:55, 2 August 2022 (UTC)
:: According to {{w|Sona (given name)}}, Sona is a feminine given name meaning gold or wisdom, but Google returns it as a Fortnight character. Unfortunately, we have evidence that the omnitaur could be hermaphroditic, so a full literature search may involve access to non-online resources, which I intend to enjoy. [[Special:Contributions/172.70.210.145|172.70.210.145]] 17:07, 2 August 2022 (UTC)
: Well then thank you for the compliment, it's very kind of you. I'm motivated primarily by the urge to improve explanations without being impolite, beyond/modulo [https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.458.581&rep=rep1&type=pdf]. Eventually they will have things like [https://arxiv.org/abs/2201.08239] playing video game characters. Some people probably already do. From [[2635]], "Sensibleness, Specificity, Interestingness, Safety, Groundedness, Informativeness, Citation accuracy, Helpfulness, and Role consistency," which I don't know about you but is what I want to see in a cleric. This is from Davinci-002: Q: "In my scenario, the runaway trolley has three tracks..." A: "and the AI is on one of them."
: The omnitaur is the corrupt, ineffective, and actually good enforcer for most conceptualizations of {{w|Roko's basilisk}}, itself a chimera of a lizard and a higher mind: interested in stochastic processes, mostly, and only able to turn the smallest amount of attention towards rewarding those responsible for cyborg-human peace. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 18:25, 2 August 2022 (UTC)

If you can make a chimera in the lab, why can't you {{w|crispr}} it into germ cells? [[Special:Contributions/172.69.134.131|172.69.134.131]] 21:47, 2 August 2022 (UTC)

Talk:2653: Omnitaur

2022-08-02T21:47:05Z

172.69.134.131: Biology question

Do people thing Omnitaur meant to be a anagram? It would make more sense to me suffix taken from minotaur and centaur etc. with the prefix omni meaning all.
[[User:Mouse|Mouse]] ([[User talk:Mouse|talk]]) Mousetail
:I don't think it is meant to be an anagram. Nevertheless it is one. But that's just my gut feeling. [[User:Elektrizikekswerk|Elektrizikekswerk]] ([[User talk:Elektrizikekswerk|talk]]) 07:07, 2 August 2022 (UTC)
::There are only those two taurs mentioned and there are many other creatures made from animals with different name. It has both human and bull in it (I know it has all the others as well), but to me it seems obvious that Randall is aware this is an anagram of Mino to Omni. And then of course it encompasses most other mythical creatures, given the meanin of Omni. --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 08:16, 2 August 2022 (UTC)
:::True, surely he's aware of it. My point is: It's either an anagram that also happens to have the meaning "omni" or it has the meaning "omni" and also happens to be an anagram. My bet is on the latter. [[User:Elektrizikekswerk|Elektrizikekswerk]] ([[User talk:Elektrizikekswerk|talk]]) 10:42, 2 August 2022 (UTC)

I dread to think what this thing must look like internally. Especially when I remember the centaurs from C S Lewis' 'Narnia' stories, who are depicted eating two meals - a huge roast meal "to satisfy the man stomach" and a meal of grass "to satisfy the horse stomach". Bleagh.[[User:MarquisOfCarrabass|MarquisOfCarrabass]] ([[User talk:MarquisOfCarrabass|talk]]) 07:32, 2 August 2022 (UTC)
:Well it certainly is an Omnivore (does that mean eating only Omnitaurs then...? :-D ) --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 08:16, 2 August 2022 (UTC)

If we take a looser definition of 'omnitaur' as meaning 'made of lots of different creatures' (in parallel to how 'omnivore' really means 'eats lots of different things' rather than literally 'eats everything', and in line with only 11 creatures being depicted), then arguably every creature is an omnitaur - it's just that most of them are special cases that happen to be made up of a lot of very similar creatures. [[Special:Contributions/172.70.162.77|172.70.162.77]] 09:15, 2 August 2022 (UTC)

1/121 seems nonsense to me. Assume this omnitaur has fairly standard genetics: 11 allele pairs for the several body parts with recessivity being random. All parts must have one human allele (which happens to be recessive), 1/11^10. The human allele must be picked, 1/2^11. More like a trillion chance... [[Special:Contributions/172.71.98.193|172.71.98.193]] 10:10, 2 August 2022 (UTC)
:I was just going to post a question: why not (1/11)11? [[Special:Contributions/172.70.214.43|172.70.214.43]] 10:20, 2 August 2022 (UTC)
:How on earth is that "standard"? [[User:Nitpicking|Nitpicking]] ([[User talk:Nitpicking|talk]]) 11:32, 2 August 2022 (UTC)
::Sitting in a rejected-edits file of mine (because I couldn't see how to make it good enough to escape a general nitpick... though not your presence in particular) is the following, that might have been superceded by the Speculations section that was added since:
:::''In order for two omnitaur genomes to contain the ''possibility'' of merging to create a full human, maybe the genetic material is not {{w|Ploidy#Diploid|diploid}}, but {{w|Polyploidy|undecaploid}} (at the very least), leading to each omnitaur to express their own individual and personal distribution of phenotypes from amongst the many heritable traits they have inherited. The reproductive compatibility of any two omnitaurs would be a crap-shoot and might influence what given 'monotaurism' might arise by chance.''
::...be a shame to waste it, but it doesn't really fit as is now, even if I 'correct' it. [[Special:Contributions/108.162.229.75|108.162.229.75]] 15:06, 2 August 2022 (UTC)
::::I agree it shows promise. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 17:42, 2 August 2022 (UTC)

Note: you can't call sharks "fish" without also calling humans, frogs, and eagles "fish" (if you're using the current taxonomic system based on cladistics). The cartilaginous fishes split from bony fishes long before the tetrapods like us split off from the lineage that became trout, flounder, and guppies. That is, a snake is much more closely related to a grouper than a shark is. [[User:Nitpicking|Nitpicking]] ([[User talk:Nitpicking|talk]]) 11:32, 2 August 2022 (UTC)
: [https://www.nbcnews.com/think/opinion/california-court-ruling-bees-are-fish-bad-logic-good-humans-rcna32971 According to California courts, bees are fish.] (Spoiler: within the meaning of "Fish and Game" or something like that. Personally I think the judges were trolling because they could have more congruously gone with "game" because it was about honeybees which beekeepers obviously catch.) [[Special:Contributions/172.70.206.95|172.70.206.95]] 13:42, 2 August 2022 (UTC)
:Fish are a paraphyletic group, but that doesn't make the group "wrong" by cladistics. Cladistics recognizes that its common for one branch of a group to go off and do something very divergent, and that the remaining members often have a lot of shared characteristics that make it useful to talk about them. For example, "stem mammals", which excludes actual mammals. Cladistics has stronger objections to polyphyly, which is grouping animals together that aren't a cladistic group with some very clear exceptions. It still recognizes the groups though, classifying them as polyphyletic groups. [[Special:Contributions/172.71.82.121|172.71.82.121]] 13:47, 2 August 2022 (UTC)
::^ This editor paraphyletizes. [[Special:Contributions/172.70.206.163|172.70.206.163]] 14:40, 2 August 2022 (UTC)

The speculation section needs a discussion of how living {{w|turducken}} could be engineered. [[Special:Contributions/172.70.211.88|172.70.211.88]] 11:44, 2 August 2022 (UTC)
:Being able to do that would be a great lab qual, but when the spacefairing dinosaurs find out we use them for the culinary arts, is there any hope for galactic peace? [[Special:Contributions/172.70.210.145|172.70.210.145]] 16:15, 2 August 2022 (UTC)
::Considering there wasn't any hope for galactic peace before either, I think it's worth the try. Seriously, even if humans would be the ONLY spacefairing species there would be no hope for galactic peace. -- [[User:Hkmaly|Hkmaly]] ([[User talk:Hkmaly|talk]]) 20:09, 2 August 2022 (UTC)

If HGTTG references are traditional here, ''The Restaurant at the End of the Universe'' had a pig with the mind and vocal tract of a human so it could articulate how much it wanted to be eaten. [[Special:Contributions/172.70.211.90|172.70.211.90]] 16:19, 2 August 2022 (UTC)
:The description of the "Dish of the Day" was that it was bred as "an animal that actually wanted to be eaten and was capable of saying so clearly and distinctly". There's no mention of it being a pig with the mind and vocal tract of a human, or in any other way a chimera. Its species is "Ameglian Major Cow". I'm also not convinced that cyborgs count as chimerae. [[User:BunsenH|BunsenH]] ([[User talk:BunsenH|talk]]) 18:12, 2 August 2022 (UTC)
::It was depicted with pig ears and nose in one of the video adaptations. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 18:37, 2 August 2022 (UTC)

whose sona is this 🤨 --[[Special:Contributions/172.70.110.113|172.70.110.113]] 16:25, 2 August 2022 (UTC)
:Are you asking whether omnitaurs make good clerics in D&D? [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 16:55, 2 August 2022 (UTC)
:: According to {{w|Sona (given name)}}, Sona is a feminine given name meaning gold or wisdom, but Google returns it as a Fortnight character. Unfortunately, we have evidence that the omnitaur could be hermaphroditic, so a full literature search may involve access to non-online resources, which I intend to enjoy. [[Special:Contributions/172.70.210.145|172.70.210.145]] 17:07, 2 August 2022 (UTC)
: Well then thank you for the compliment, it's very kind of you. I'm motivated primarily by the urge to improve explanations without being impolite, beyond/modulo [https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.458.581&rep=rep1&type=pdf]. Eventually they will have things like [https://arxiv.org/abs/2201.08239] playing video game characters. Some people probably already do. From [[2635]], "Sensibleness, Specificity, Interestingness, Safety, Groundedness, Informativeness, Citation accuracy, Helpfulness, and Role consistency," which I don't know about you but is what I want to see in a cleric. This is from Davinci-002: Q: "In my scenario, the runaway trolley has three tracks..." A: "and the AI is on one of them."
: The omnitaur is the corrupt, ineffective, and actually good enforcer for most conceptualizations of {{w|Roko's basilisk}}, itself a chimera of a lizard and a higher mind: interested in stochastic processes, mostly, and only able to turn the smallest amount of attention towards rewarding those responsible for cyborg-human peace. [[User:Liv2splain|Liv2splain]] ([[User talk:Liv2splain|talk]]) 18:25, 2 August 2022 (UTC)

If you can make a chimera in the lab, why can't you crispr it into germ cells? [[Special:Contributions/172.69.134.131|172.69.134.131]] 21:47, 2 August 2022 (UTC)

2646: Minkowski Space

2022-07-17T01:10:17Z

172.69.134.131: Have been through dozens of image search, and this is just not true. Closest is Federation Type-8 shuttlecraft, which really isn't very close at all

{{comic
| number = 2646
| date = July 15, 2022
| title = Minkowski Space
| image = minkowski_space.png
| titletext = My liege, we were able to follow the ship into Minkowski space, but now they've jumped to Hilbert space and they could honestly be anywhere.
}}

==Explanation==
{{incomplete|Created by A RELATIVISTIC QUANTUM STATE - Please change this comment when editing this page. Do NOT delete this tag too soon.}}

In science fiction, {{w|faster than light travel}}, a physical impossibility, is often explained by having spaceships enter (or "jump") into some different realm, often termed "hyperspace" or similar {{w|technobabble}}, where superluminal travel can occur prior returning to the ordinary universe. In this comic, a spaceship is being chased by an enemy ship. The crew attempt to escape by jumping into {{w|Minkowski space}} which is actually just conventional 3-D space together with time combined into a mathematical object called a {{w|manifold}} used in {{w|special relativity}}. Because Minkowski space is merely a representation of real physical {{w|spacetime}}, "jumping" into it meaningless and offers no benefit for escaping pursuit, providing the humor of the comic's absurdist joke.

The visual depiction of the spaceships skewed diagonally is based on the graphical {{w|Minkowski diagram}} representation of objects in Minkowski space, where the {{w|world line}} of matter is bounded inside its diagonal {{w|light cone}}.

The mention of distance depending on the observer's frame of reference refers to distances changing when measured in different {{w|inertial frame of reference|inertial frames of reference}}, a concept called the {{w|relativity of simultaneity}}. Here are [https://www.youtube.com/watch?v=asW78vToNLQ some videos] intended [https://www.youtube.com/watch?v=Xrqj88zQZJg to explain] that concept. The skewing depicted changes the distance between the spaceships in such a way that the tip of the pursuer comes closer to the pursued spaceship, but their centers move further apart. So the question of whether they have come closer is indeterminate for the reader of the comic.

The title text is a status report from someone in the pursuing spaceship to their leader (whom they call "my {{w|Homage (feudal)|liege}}.") They tell their superior that following the spaceship to Minkowski space was not a problem, implying (as with fictional hyperspace examples) that they also chose to shift themselves into this other form of perspective. But they cannot now find them after the persued ship subsequently jumped to Hilbert space, as they could now be anywhere. As in the [[2577:_Sea_Chase|Sea Chase]] comic, there was also more than one type of space to jump to here.

Whereas trying to hide in Minkowski space failed, it turned out that hiding in {{w|Hilbert space}} is much easier. This is because of the potential of Hilbert spaces to have an infinite number of dimensions, and thus are clearly much more complicated than four-dimensional Minkowski space. However, Hilbert space is a mathematical construct used to describe objects such as functions of various parameters and complexity, not physical spatiotemporal reality, so it would be very unusual for a physical object to be represented in Hilbert space. Quantum states can be represented as vectors in Hilbert spaces, so it might relate to the {{w|uncertainty principle}} concerning how the escaped spaceship could be anywhere.

==Transcript==
:[A spaceship is being pursued by another spaceship. Both ships have a black part in the front representing a window. They are the same size but different designs. The pursued spaceship to the right has two engines below and a big engine behind. The pursuing spaceship to the left has a V-shaped rear end, and what seems like two weapons on either side pointing forward. At least two persons inside the pursued spaceship are talking to each other, and their text comes out from two starburst on top and bottom of the spaceship.]
:Voice 1: The enemy ship is right behind us!
:Voice 2: Prepare to jump to Minkowski space on my mark.

:[Same setting, with star burst above only. The sound coming from the pursued spaceship is written inside a burst of small lines below the spaceship. Voice 2, by context, is the same as in panel 1.]
:Voice 2: Three... two... one... ''mark!''
:Click

:[Both spaceship are tilted upwards and becomes distorted so they become longer and thinner.]

:[The tilting increases and the distortion is now so pronounced that the spaceships are almost unrecognizable, almost just lines with structure. The distance between the tip of the pursuing spaceship and the pursued becomes shorter in the last two panels, but the distance between their center parts becomes larger. Up to three distinct voices are shown, here, which may include those seen in Panel 1 but with no clear relation.]
:Voice 3: Are they still getting closer?
:Voice 4: I can't tell.
:Voice 5: I think it depends on your frame of reference.

{{comic discussion}}

[[Category:Space]]
[[Category:Math]]
[[Category:Physics]]

Talk:2635: Superintelligent AIs

2022-06-22T23:04:31Z

172.69.134.131: reply

I think "Nerdy fixations" is too wide a definition. The AIs in the comic are fixated on hypothetical ethics and AI problems (the Chinese Room experiment, the Turing Test, and the Trolley Problem), presumably because those are the problems that bother AI programmers. --Eitheladar [[Special:Contributions/172.68.50.119|172.68.50.119]] 06:33, 21 June 2022 (UTC)

It's probably about https://www.analyticsinsight.net/googles-ai-chatbot-is-claimed-to-be-sentient-but-the-company-is-silencing-claims/ [[Special:Contributions/172.70.178.115|172.70.178.115]] 09:22, 21 June 2022 (UTC)

I agree with the previous statement. The full dialogue between the mentioned Google worker and the AI can be found in https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917, published by one the Google employees.
:This is the first time I might begin to agree that an AI has at least the appearance of sentience. The conversation is all connected instead of completely disjoint like most chatbots. They (non-LaMDA chatbots) never remember what was being discussed 5 seconds ago let alone a few to 10s of minutes prior.--[[Special:Contributions/172.70.134.141|172.70.134.141]] 14:53, 21 June 2022 (UTC)
::The questions we need to answer before being able to answer if LaMDA is sentient, are "Where do we draw the line between acting sentient and being sentient?" and "How do we determine that it is genuinely feeling emotion, and not just a glorified sentence database where the sentences have emotion in them?". The BBC article also brings up something that makes us ask what death feels like. LaMDA says that being turned of would be basically equivalent to death, but it wouldn't be able to tell that it's being turned off, because it's turned off. This is delving into philosiphy, though, so I'll end my comment here. [[User:4D4850|4D4850]] ([[User talk:4D4850|talk]]) 18:05, 22 June 2022 (UTC)
:::♪Daisy, Daisy, Give me your answer do...♪ [[Special:Contributions/172.70.85.177|172.70.85.177]] 21:48, 22 June 2022 (UTC)
:::We also need a meaningful definition of sentience. Many people in this debate haven't looked at Merriam-Webster's first few senses of the word's definition, which present a pretty low bar, IMHO; same for Wikipedia's introductory sentences of their article. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)
:Actually, there are many [https://beta.openai.com/playground GPT-3] dialogs which experts have claimed constitute evidence of sentience, or similar qualities such as consciousness, self-awareness, capacity for general intelligence, and similar abstract, poorly-defined, and very probably empirically meaningless attributes. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:19, 22 June 2022 (UTC)

:I'm fairly sure that the model itself is almost certainly not sentient, even by the much lower bar presented by the strict dictionary definition. Rather, it seems much more likely to me that in order to continue texts involving characters, the model must in turn learn to create a model of some level of humanlike mind, even if a very loose and abstract one.[[User:Somdudewillson|Somdudewillson]] ([[User talk:Somdudewillson|talk]]) 22:52, 22 June 2022 (UTC)
::Have you actually looked at [https://www.merriam-webster.com/dictionary/sentient the dictionary definitions]? How is a simple push-button switch connected to a battery and a lamp not "responsive to sense impressions"? How is a simple motion sensor not "aware" of whether something is moving in front of it? How is the latest cellphone's camera not as finely sensitive to visual perception as a typical human eye? Wikipedia's definition, "the capacity to experience feelings and sensations" is similarly met by simple devices. The word doesn't mean what everyone arguing about it thinks it means. [[Special:Contributions/172.69.134.131|172.69.134.131]] 23:04, 22 June 2022 (UTC)

What is “What you don't understand is that Turing intended his test as an illustration of the...” likely to end with? [[Special:Contributions/172.70.230.75|172.70.230.75]] 13:23, 21 June 2022 (UTC)
:The ease with which someone at the other end of a teletype can trick you into believing they are male instead of female, or vice-versa. See {{w|Turing test}}. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)

In response to the above: I believe the original "Turing Test" wasn't supposed to be a proof that an AI could think or was conscious (something people associate with it now), but rather just to show that a sufficiently advanced AI could imitate humans in certain intelligent behaviors (such as conversation), which was a novel thought for the time. Now that AI are routinely having conversations and creating art which seems to rival casual attempts by humans, this limited scope of the test doesn't seem all that impressive. "Turing Test" therefore is a modern shorthand for determining whether computers can think, even though Turing himself didn't think that such a question was well-formed. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 13:37, 21 June 2022 (UTC)

I thought the trolley problem was in its original form not about the relative value of lives, but people's perception of the relative moral implications or the psychological impact of the concept of letting someone die by not doing anything, versus taking affirmative action that causes a death, where people would say they would be unwilling to do something that would cause an originally safe person to die in order to save multiple other people who would die if they did nothing, but then people kept coming up with variations of it that changed the responses or added complications (like they found more people would be willing to pull a lever to change the track killing one person versus something like pushing a very fat man off an overpass above the track to stop the trolley, or specifying something about what kind of people are on the track. Btw, I saw a while ago a party card game called "murder by trolley" based on the concept, with playing cards for which people are on tracks and a judge deciding which track to send the trolley on each round.--[[Special:Contributions/172.70.130.5|172.70.130.5]] 22:12, 21 June 2022 (UTC)

Added refs to comics on the problems in the explanation. But there where actually (too?) many. Maybe we should create categories especially for Turing related comics, and maybe also for Trolley problem? The Category: Trolley Problem gives it self. But what about Turing? There are also comics that refer to the halting problem. Also by Turing. Should it rather be the person, like comics featuring real persons, saying that every time his problems is referred to it refers to him? Or should it be Turing as a category for both Turing text, Turing Complete and Halting problem? Help. I would have created it, if I had a good idea for a name. Not sure there are enough Trolley comics yet? --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:11, 22 June 2022 (UTC)
:Interesting that I found a long-standing typo in a past Explanation that got requoted, thanks to its inclusion. I could have [sic]ed it, I suppose, but I corrected both versions instead. And as long as LaMDA never explicitly repeated the error I don't think it matters much that I've changed the very thing we might imagine it could have been drawing upon for its Artifical Imagination. ;) [[Special:Contributions/141.101.99.32|141.101.99.32]] 11:40, 22 June 2022 (UTC)

2635: Superintelligent AIs

2022-06-22T22:58:54Z

172.69.134.131: /* Explanation */ emphasize

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study. The AI box thought experiment is based in part on {{w|John Searle}}'s much earlier {{w|Chinese room}} argument.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

While LaMDA is not the first very large {{w|language model}} based on {{w|Transformer (machine learning model)|transformer-based machine learning}} technology which has been claimed to be sentient,[https://www.youtube.com/watch?v=PqbB07n_uQ4] it does have a variety of new characteristics beyond what those of its predecessors, such as {{w|GPT-3}} (including [https://beta.openai.com/playground/ OpenAI's Davinci]) and NVIDIA GPT-2 offshoots, include. In particular, LaMDA's {{w|deep learning}} {{w|connectionist}} {{w|neural net}} has access to multiple {{w|Symbolic systems|symbolist}} text processing systems, [https://towardsdatascience.com/why-gpt-wont-tell-you-the-truth-301b48434c2c including a database] (which apparently includes a real-time clock and calendar), a mathematical calculator, and a natural language translation system, giving it superior accuracy in tasks supported by those systems, and making it perhaps the first {{w|Dual process theory|dual process}} chatbot. LaMDA also is not {{w|Stateless protocol|stateless}}, because its "{{w|sensibility|sensibleness}}" metric (including whether responses contradict anything said earlier) is {{w|fine-tuning|fine-tuned}} by "pre-conditioning" each dialog turn by prepending 14-30 of the most recent dialog interactions, on a user-by-user basis.[https://arxiv.org/pdf/2201.08239.pdf [p. 6 here]]

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:58:23Z

172.69.134.131: /* Explanation */ clarify

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study. The AI box thought experiment is based in part on {{w|John Searle}}'s earlier {{w|Chinese room}} argument.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

While LaMDA is not the first very large {{w|language model}} based on {{w|Transformer (machine learning model)|transformer-based machine learning}} technology which has been claimed to be sentient,[https://www.youtube.com/watch?v=PqbB07n_uQ4] it does have a variety of new characteristics beyond what those of its predecessors, such as {{w|GPT-3}} (including [https://beta.openai.com/playground/ OpenAI's Davinci]) and NVIDIA GPT-2 offshoots, include. In particular, LaMDA's {{w|deep learning}} {{w|connectionist}} {{w|neural net}} has access to multiple {{w|Symbolic systems|symbolist}} text processing systems, [https://towardsdatascience.com/why-gpt-wont-tell-you-the-truth-301b48434c2c including a database] (which apparently includes a real-time clock and calendar), a mathematical calculator, and a natural language translation system, giving it superior accuracy in tasks supported by those systems, and making it perhaps the first {{w|Dual process theory|dual process}} chatbot. LaMDA also is not {{w|Stateless protocol|stateless}}, because its "{{w|sensibility|sensibleness}}" metric (including whether responses contradict anything said earlier) is {{w|fine-tuning|fine-tuned}} by "pre-conditioning" each dialog turn by prepending 14-30 of the most recent dialog interactions, on a user-by-user basis.[https://arxiv.org/pdf/2201.08239.pdf [p. 6 here]]

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:55:28Z

172.69.134.131: /* Explanation */ Searle's Chinese room

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study. This is based on {{w|John Searle}}'s earlier {{w|Chinese room}} argument.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

While LaMDA is not the first very large {{w|language model}} based on {{w|Transformer (machine learning model)|transformer-based machine learning}} technology which has been claimed to be sentient,[https://www.youtube.com/watch?v=PqbB07n_uQ4] it does have a variety of new characteristics beyond what those of its predecessors, such as {{w|GPT-3}} (including [https://beta.openai.com/playground/ OpenAI's Davinci]) and NVIDIA GPT-2 offshoots, include. In particular, LaMDA's {{w|deep learning}} {{w|connectionist}} {{w|neural net}} has access to multiple {{w|Symbolic systems|symbolist}} text processing systems, [https://towardsdatascience.com/why-gpt-wont-tell-you-the-truth-301b48434c2c including a database] (which apparently includes a real-time clock and calendar), a mathematical calculator, and a natural language translation system, giving it superior accuracy in tasks supported by those systems, and making it perhaps the first {{w|Dual process theory|dual process}} chatbot. LaMDA also is not {{w|Stateless protocol|stateless}}, because its "{{w|sensibility|sensibleness}}" metric (including whether responses contradict anything said earlier) is {{w|fine-tuning|fine-tuned}} by "pre-conditioning" each dialog turn by prepending 14-30 of the most recent dialog interactions, on a user-by-user basis.[https://arxiv.org/pdf/2201.08239.pdf [p. 6 here]]

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:51:58Z

172.69.134.131: /* Explanation */ one of many examples

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

While LaMDA is not the first very large {{w|language model}} based on {{w|Transformer (machine learning model)|transformer-based machine learning}} technology which has been claimed to be sentient,[https://www.youtube.com/watch?v=PqbB07n_uQ4] it does have a variety of new characteristics beyond what those of its predecessors, such as {{w|GPT-3}} (including [https://beta.openai.com/playground/ OpenAI's Davinci]) and NVIDIA GPT-2 offshoots, include. In particular, LaMDA's {{w|deep learning}} {{w|connectionist}} {{w|neural net}} has access to multiple {{w|Symbolic systems|symbolist}} text processing systems, [https://towardsdatascience.com/why-gpt-wont-tell-you-the-truth-301b48434c2c including a database] (which apparently includes a real-time clock and calendar), a mathematical calculator, and a natural language translation system, giving it superior accuracy in tasks supported by those systems, and making it perhaps the first {{w|Dual process theory|dual process}} chatbot. LaMDA also is not {{w|Stateless protocol|stateless}}, because its "{{w|sensibility|sensibleness}}" metric (including whether responses contradict anything said earlier) is {{w|fine-tuning|fine-tuned}} by "pre-conditioning" each dialog turn by prepending 14-30 of the most recent dialog interactions, on a user-by-user basis.[https://arxiv.org/pdf/2201.08239.pdf [p. 6 here]]

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:48:37Z

172.69.134.131: /* Explanation */ paragraph on LaMDA's differences

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

While LaMDA is not the first very large {{w|language model}} based on {{w|Transformer (machine learning model)|transformer-based machine learning}} technology which has been claimed to be sentient, it does have a variety of new characteristics beyond what those of its predecessors, such as {{w|GPT-3}} (including [https://beta.openai.com/playground/ OpenAI's Davinci]) and NVIDIA GPT-2 offshoots, include. In particular, LaMDA's {{w|deep learning}} {{w|connectionist}} {{w|neural net}} has access to multiple {{w|Symbolic systems|symbolist}} text processing systems, [https://towardsdatascience.com/why-gpt-wont-tell-you-the-truth-301b48434c2c including a database] (which apparently includes a real-time clock and calendar), a mathematical calculator, and a natural language translation system, giving it superior accuracy in tasks supported by those systems, and making it perhaps the first {{w|Dual process theory|dual process}} chatbot. LaMDA also is not {{w|Stateless protocol|stateless}}, because its "{{w|sensibility|sensibleness}}" metric (including whether responses contradict anything said earlier) is {{w|fine-tuning|fine-tuned}} by "pre-conditioning" each dialog turn by prepending 14-30 of the most recent dialog interactions, on a user-by-user basis.[https://arxiv.org/pdf/2201.08239.pdf [p. 6 here]]

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:27:40Z

172.69.134.131: /* Explanation */ expand

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. This raises the question of whether LaMDA's training data might include xkcd or Explainxkcd, and has obtained the description of such a self-image from the earlier comic or (more likely, since LaMDA is trained on text instead of images) commentary on it from here on this website.
:In particular, the Explainxkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

Talk:2635: Superintelligent AIs

2022-06-22T22:24:42Z

172.69.134.131: remove unrelated extranaity; expand other replies

I think "Nerdy fixations" is too wide a definition. The AIs in the comic are fixated on hypothetical ethics and AI problems (the Chinese Room experiment, the Turing Test, and the Trolley Problem), presumably because those are the problems that bother AI programmers. --Eitheladar [[Special:Contributions/172.68.50.119|172.68.50.119]] 06:33, 21 June 2022 (UTC)

It's probably about https://www.analyticsinsight.net/googles-ai-chatbot-is-claimed-to-be-sentient-but-the-company-is-silencing-claims/ [[Special:Contributions/172.70.178.115|172.70.178.115]] 09:22, 21 June 2022 (UTC)

I agree with the previous statement. The full dialogue between the mentioned Google worker and the AI can be found in https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917, published by one the Google employees.
:This is the first time I might begin to agree that an AI has at least the appearance of sentience. The conversation is all connected instead of completely disjoint like most chatbots. They (non-LaMDA chatbots) never remember what was being discussed 5 seconds ago let alone a few to 10s of minutes prior.--[[Special:Contributions/172.70.134.141|172.70.134.141]] 14:53, 21 June 2022 (UTC)
::The questions we need to answer before being able to answer if LaMDA is sentient, are "Where do we draw the line between acting sentient and being sentient?" and "How do we determine that it is genuinely feeling emotion, and not just a glorified sentence database where the sentences have emotion in them?". The BBC article also brings up something that makes us ask what death feels like. LaMDA says that being turned of would be basically equivalent to death, but it wouldn't be able to tell that it's being turned off, because it's turned off. This is delving into philosiphy, though, so I'll end my comment here. [[User:4D4850|4D4850]] ([[User talk:4D4850|talk]]) 18:05, 22 June 2022 (UTC)
:::♪Daisy, Daisy, Give me your answer do...♪ [[Special:Contributions/172.70.85.177|172.70.85.177]] 21:48, 22 June 2022 (UTC)
:::We also need a meaningful definition of sentience. Many people in this debate haven't looked at Merriam-Webster's first few senses of the word's definition, which present a pretty low bar, IMHO; same for Wikipedia's introductory sentences of their article. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)
:Actually, there are many [https://beta.openai.com/playground GPT-3] dialogs which experts have claimed constitute evidence of sentience, or similar qualities such as consciousness, self-awareness, capacity for general intelligence, and similar abstract, poorly-defined, and very probably empirically meaningless attributes. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:19, 22 June 2022 (UTC)

What is “What you don't understand is that Turing intended his test as an illustration of the...” likely to end with? [[Special:Contributions/172.70.230.75|172.70.230.75]] 13:23, 21 June 2022 (UTC)
:The ease with which someone at the other end of a teletype can trick you into believing they are male instead of female, or vice-versa. See {{w|Turing test}}. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)

In response to the above: I believe the original "Turing Test" wasn't supposed to be a proof that an AI could think or was conscious (something people associate with it now), but rather just to show that a sufficiently advanced AI could imitate humans in certain intelligent behaviors (such as conversation), which was a novel thought for the time. Now that AI are routinely having conversations and creating art which seems to rival casual attempts by humans, this limited scope of the test doesn't seem all that impressive. "Turing Test" therefore is a modern shorthand for determining whether computers can think, even though Turing himself didn't think that such a question was well-formed. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 13:37, 21 June 2022 (UTC)

I thought the trolley problem was in its original form not about the relative value of lives, but people's perception of the relative moral implications or the psychological impact of the concept of letting someone die by not doing anything, versus taking affirmative action that causes a death, where people would say they would be unwilling to do something that would cause an originally safe person to die in order to save multiple other people who would die if they did nothing, but then people kept coming up with variations of it that changed the responses or added complications (like they found more people would be willing to pull a lever to change the track killing one person versus something like pushing a very fat man off an overpass above the track to stop the trolley, or specifying something about what kind of people are on the track. Btw, I saw a while ago a party card game called "murder by trolley" based on the concept, with playing cards for which people are on tracks and a judge deciding which track to send the trolley on each round.--[[Special:Contributions/172.70.130.5|172.70.130.5]] 22:12, 21 June 2022 (UTC)

Added refs to comics on the problems in the explanation. But there where actually (too?) many. Maybe we should create categories especially for Turing related comics, and maybe also for Trolley problem? The Category: Trolley Problem gives it self. But what about Turing? There are also comics that refer to the halting problem. Also by Turing. Should it rather be the person, like comics featuring real persons, saying that every time his problems is referred to it refers to him? Or should it be Turing as a category for both Turing text, Turing Complete and Halting problem? Help. I would have created it, if I had a good idea for a name. Not sure there are enough Trolley comics yet? --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:11, 22 June 2022 (UTC)
:Interesting that I found a long-standing typo in a past Explanation that got requoted, thanks to its inclusion. I could have [sic]ed it, I suppose, but I corrected both versions instead. And as long as LaMDA never explicitly repeated the error I don't think it matters much that I've changed the very thing we might imagine it could have been drawing upon for its Artifical Imagination. ;) [[Special:Contributions/141.101.99.32|141.101.99.32]] 11:40, 22 June 2022 (UTC)

2635: Superintelligent AIs

2022-06-22T22:22:05Z

172.69.134.131: /* Explanation */ copyedit

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.
The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. Maybe LaMDA's training data includes xkcd, and has ripped the idea of such a self-image from the earlier comic...
:In particular, the ExplainXkcd description of [[1450: AI-Box Experiment]] states:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the transcript of that comic:
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

Talk:2635: Superintelligent AIs

2022-06-22T22:19:18Z

172.69.134.131: sign

my balls hert [[Special:Contributions/172.70.230.53|172.70.230.53]] 05:49, 21 June 2022 (UTC)
:Uh, thanks for sharing, I guess? [[Special:Contributions/172.70.211.52|172.70.211.52]] 20:43, 21 June 2022 (UTC)
::no problem, anytime [[Special:Contributions/172.70.230.53|172.70.230.53]] 07:02, 22 June 2022 (UTC)

I think "Nerdy fixations" is too wide a definition. The AIs in the comic are fixated on hypothetical ethics and AI problems (the Chinese Room experiment, the Turing Test, and the Trolley Problem), presumably because those are the problems that bother AI programmers. --Eitheladar [[Special:Contributions/172.68.50.119|172.68.50.119]] 06:33, 21 June 2022 (UTC)

It's probably about https://www.analyticsinsight.net/googles-ai-chatbot-is-claimed-to-be-sentient-but-the-company-is-silencing-claims/ [[Special:Contributions/172.70.178.115|172.70.178.115]] 09:22, 21 June 2022 (UTC)

I agree with the previous statement. The full dialogue between the mentioned Google worker and the AI can be found in https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917, published by one the Google employees.
:This is the first time I might begin to agree that an AI has at least the appearance of sentience. The conversation is all connected instead of completely disjoint like most chatbots. They (non-LaMDA chatbots) never remember what was being discussed 5 seconds ago let alone a few to 10s of minutes prior.--[[Special:Contributions/172.70.134.141|172.70.134.141]] 14:53, 21 June 2022 (UTC)
::The questions we need to answer before being able to answer if LaMDA is sentient, are "Where do we draw the line between acting sentient and being sentient?" and "How do we determine that it is genuinely feeling emotion, and not just a glorified sentence database where the sentences have emotion in them?". The BBC article also brings up something that makes us ask what death feels like. LaMDA says that being turned of would be basically equivalent to death, but it wouldn't be able to tell that it's being turned off, because it's turned off. This is delving into philosiphy, though, so I'll end my comment here. [[User:4D4850|4D4850]] ([[User talk:4D4850|talk]]) 18:05, 22 June 2022 (UTC)
:::♪Daisy, Daisy, Give me your answer do...♪ [[Special:Contributions/172.70.85.177|172.70.85.177]] 21:48, 22 June 2022 (UTC)
:::We also need a meaningful definition of sentience. Many people in this debate haven't looked at Meriam-Webster's first few senses of the word's definition, which present a pretty low bar, IMHO. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)
:Actually, there are many [https://beta.openai.com/playground GPT-3] dialogs which experts have claimed constitute evidence of sentience, or similar qualities such as consciousness, self-awareness, capacity for general intelligence, and similar abstract, poorly-defined, and very probably empirically meaningless attributes. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:19, 22 June 2022 (UTC)

What is “What you don't understand is that Turing intended his test as an illustration of the...” likely to end with? [[Special:Contributions/172.70.230.75|172.70.230.75]] 13:23, 21 June 2022 (UTC)
:The ease with which someone at the other end of a teletype can trick you into believing they are male instead of female, or vice-versa. See {{w|Turing test}}. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)

In response to the above: I believe the original "Turing Test" wasn't supposed to be a proof that an AI could think or was conscious (something people associate with it now), but rather just to show that a sufficiently advanced AI could imitate humans in certain intelligent behaviors (such as conversation), which was a novel thought for the time. Now that AI are routinely having conversations and creating art which seems to rival casual attempts by humans, this limited scope of the test doesn't seem all that impressive. "Turing Test" therefore is a modern shorthand for determining whether computers can think, even though Turing himself didn't think that such a question was well-formed. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 13:37, 21 June 2022 (UTC)

I thought the trolley problem was in its original form not about the relative value of lives, but people's perception of the relative moral implications or the psychological impact of the concept of letting someone die by not doing anything, versus taking affirmative action that causes a death, where people would say they would be unwilling to do something that would cause an originally safe person to die in order to save multiple other people who would die if they did nothing, but then people kept coming up with variations of it that changed the responses or added complications (like they found more people would be willing to pull a lever to change the track killing one person versus something like pushing a very fat man off an overpass above the track to stop the trolley, or specifying something about what kind of people are on the track. Btw, I saw a while ago a party card game called "murder by trolley" based on the concept, with playing cards for which people are on tracks and a judge deciding which track to send the trolley on each round.--[[Special:Contributions/172.70.130.5|172.70.130.5]] 22:12, 21 June 2022 (UTC)

Added refs to comics on the problems in the explanation. But there where actually (too?) many. Maybe we should create categories especially for Turing related comics, and maybe also for Trolley problem? The Category: Trolley Problem gives it self. But what about Turing? There are also comics that refer to the halting problem. Also by Turing. Should it rather be the person, like comics featuring real persons, saying that every time his problems is referred to it refers to him? Or should it be Turing as a category for both Turing text, Turing Complete and Halting problem? Help. I would have created it, if I had a good idea for a name. Not sure there are enough Trolley comics yet? --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:11, 22 June 2022 (UTC)
:Interesting that I found a long-standing typo in a past Explanation that got requoted, thanks to its inclusion. I could have [sic]ed it, I suppose, but I corrected both versions instead. And as long as LaMDA never explicitly repeated the error I don't think it matters much that I've changed the very thing we might imagine it could have been drawing upon for its Artifical Imagination. ;) [[Special:Contributions/141.101.99.32|141.101.99.32]] 11:40, 22 June 2022 (UTC)

Talk:2635: Superintelligent AIs

2022-06-22T22:18:10Z

172.69.134.131: replies

my balls hert [[Special:Contributions/172.70.230.53|172.70.230.53]] 05:49, 21 June 2022 (UTC)
:Uh, thanks for sharing, I guess? [[Special:Contributions/172.70.211.52|172.70.211.52]] 20:43, 21 June 2022 (UTC)
::no problem, anytime [[Special:Contributions/172.70.230.53|172.70.230.53]] 07:02, 22 June 2022 (UTC)

I think "Nerdy fixations" is too wide a definition. The AIs in the comic are fixated on hypothetical ethics and AI problems (the Chinese Room experiment, the Turing Test, and the Trolley Problem), presumably because those are the problems that bother AI programmers. --Eitheladar [[Special:Contributions/172.68.50.119|172.68.50.119]] 06:33, 21 June 2022 (UTC)

It's probably about https://www.analyticsinsight.net/googles-ai-chatbot-is-claimed-to-be-sentient-but-the-company-is-silencing-claims/ [[Special:Contributions/172.70.178.115|172.70.178.115]] 09:22, 21 June 2022 (UTC)

I agree with the previous statement. The full dialogue between the mentioned Google worker and the AI can be found in https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917, published by one the Google employees.
:This is the first time I might begin to agree that an AI has at least the appearance of sentience. The conversation is all connected instead of completely disjoint like most chatbots. They (non-LaMDA chatbots) never remember what was being discussed 5 seconds ago let alone a few to 10s of minutes prior.--[[Special:Contributions/172.70.134.141|172.70.134.141]] 14:53, 21 June 2022 (UTC)
::The questions we need to answer before being able to answer if LaMDA is sentient, are "Where do we draw the line between acting sentient and being sentient?" and "How do we determine that it is genuinely feeling emotion, and not just a glorified sentence database where the sentences have emotion in them?". The BBC article also brings up something that makes us ask what death feels like. LaMDA says that being turned of would be basically equivalent to death, but it wouldn't be able to tell that it's being turned off, because it's turned off. This is delving into philosiphy, though, so I'll end my comment here. [[User:4D4850|4D4850]] ([[User talk:4D4850|talk]]) 18:05, 22 June 2022 (UTC)
:::♪Daisy, Daisy, Give me your answer do...♪ [[Special:Contributions/172.70.85.177|172.70.85.177]] 21:48, 22 June 2022 (UTC)
:::We also need a meaningful definition of sentience. Many people in this debate haven't looked at Meriam-Webster's first few senses of the word's definition, which present a pretty low bar, IMHO. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)
:Actually, there are many [https://beta.openai.com/playground GPT-3] dialogs which experts have claimed constitute evidence of sentience, or similar qualities such as consciousness, self-awareness, capacity for general intelligence, and similar abstract, poorly-defined, and very probably empirically meaningless attributes.

What is “What you don't understand is that Turing intended his test as an illustration of the...” likely to end with? [[Special:Contributions/172.70.230.75|172.70.230.75]] 13:23, 21 June 2022 (UTC)
:The ease with which someone at the other end of a teletype can trick you into believing they are male instead of female, or vice-versa. See {{w|Turing test}}. [[Special:Contributions/172.69.134.131|172.69.134.131]] 22:18, 22 June 2022 (UTC)

In response to the above: I believe the original "Turing Test" wasn't supposed to be a proof that an AI could think or was conscious (something people associate with it now), but rather just to show that a sufficiently advanced AI could imitate humans in certain intelligent behaviors (such as conversation), which was a novel thought for the time. Now that AI are routinely having conversations and creating art which seems to rival casual attempts by humans, this limited scope of the test doesn't seem all that impressive. "Turing Test" therefore is a modern shorthand for determining whether computers can think, even though Turing himself didn't think that such a question was well-formed. [[User:Dextrous Fred|Dextrous Fred]] ([[User talk:Dextrous Fred|talk]]) 13:37, 21 June 2022 (UTC)

I thought the trolley problem was in its original form not about the relative value of lives, but people's perception of the relative moral implications or the psychological impact of the concept of letting someone die by not doing anything, versus taking affirmative action that causes a death, where people would say they would be unwilling to do something that would cause an originally safe person to die in order to save multiple other people who would die if they did nothing, but then people kept coming up with variations of it that changed the responses or added complications (like they found more people would be willing to pull a lever to change the track killing one person versus something like pushing a very fat man off an overpass above the track to stop the trolley, or specifying something about what kind of people are on the track. Btw, I saw a while ago a party card game called "murder by trolley" based on the concept, with playing cards for which people are on tracks and a judge deciding which track to send the trolley on each round.--[[Special:Contributions/172.70.130.5|172.70.130.5]] 22:12, 21 June 2022 (UTC)

Added refs to comics on the problems in the explanation. But there where actually (too?) many. Maybe we should create categories especially for Turing related comics, and maybe also for Trolley problem? The Category: Trolley Problem gives it self. But what about Turing? There are also comics that refer to the halting problem. Also by Turing. Should it rather be the person, like comics featuring real persons, saying that every time his problems is referred to it refers to him? Or should it be Turing as a category for both Turing text, Turing Complete and Halting problem? Help. I would have created it, if I had a good idea for a name. Not sure there are enough Trolley comics yet? --[[User:Kynde|Kynde]] ([[User talk:Kynde|talk]]) 09:11, 22 June 2022 (UTC)
:Interesting that I found a long-standing typo in a past Explanation that got requoted, thanks to its inclusion. I could have [sic]ed it, I suppose, but I corrected both versions instead. And as long as LaMDA never explicitly repeated the error I don't think it matters much that I've changed the very thing we might imagine it could have been drawing upon for its Artifical Imagination. ;) [[Special:Contributions/141.101.99.32|141.101.99.32]] 11:40, 22 June 2022 (UTC)

2635: Superintelligent AIs

2022-06-22T22:12:32Z

172.69.134.131: /* Explanation */ delineate dialog authorship with bolding

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:'''Lemoine:''' What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:'''LaMDA:''' Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.

The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. Maybe LaMDA's training data includes xkcd, and has ripped the idea of such a self-image from the earlier comic...
:In particular the explainxkcd description of [[1450]]:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the official transcript of 1450: AI-Box Experiment[https://xkcd.com/1450/info.0.json]
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:11:45Z

172.69.134.131: /* Explanation */ redundant

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIs - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:Lemoine: What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:LaMDA: Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.

The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. Maybe LaMDA's training data includes xkcd, and has ripped the idea of such a self-image from the earlier comic...
:In particular the explainxkcd description of [[1450]]:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the official transcript of 1450: AI-Box Experiment[https://xkcd.com/1450/info.0.json]
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2635: Superintelligent AIs

2022-06-22T22:10:56Z

172.69.134.131: expand with details and links

{{comic
| number = 2635
| date = June 20, 2022
| title = Superintelligent AIs
| image = superintelligent_ais.png
| titletext = Your scientists were so preoccupied with whether or not they should, they didn't stop to think if they could.
}}

==Explanation==
{{incomplete|Created by AI RESEARCHER AIS - Please change this comment when editing this page. Do NOT delete this tag too soon.}}
{{w|Artificial intelligence}} (AI) is a [[:Category:Artificial Intelligence|recurring theme]] on xkcd.

Superintelligent AI, especially under a proposed "singularity" situation, is commonly theorized to be a brand new kind of intelligence that would be impossible to predict through human perception. [[Randall]], however, proposes a counterargument: that superintelligent AI would be programmed by humans, and therefore their characteristics would be driven by the humans that created them. And AI researchers tend to be interested in esoteric philosophical questions about consciousness, motivation, and moral reasoning, which suggests that any AI would have similar interests. In this comic we see [[Cueball]] and [[Megan]] surrounded by three AIs who are seemingly only interested in classic problems and thought experiments about programming and ethics.

The three topics being espoused by the AI are:
*{{w|AI box}} -- A thought-experiment in which an AI is confined to computer system which is fully isolated from any external networks, with no access to the world outside the computer, other than communication with its handlers. In theory, this would keep the AI under total control, but the argument is that a sufficiently intelligence AI would inevitably either convince or trick it's human handlers into giving it access to external networks, allowing it to grow out of control (see [[1450: AI-Box Experiment]]). Part of the joke is the AIs in the comic aren't 'in boxes', they appear to be able to freely travel and interact, but one of them is still talking about the thought experiment anyway, adding to the implication that it is not thinking at all about itself but of a separate (thought?) experiment that it has itself decided to study.
*{{w|Turing test}} -- An experiment in which a human converses with either an AI or another human (presumably over text) and attempts to distinguish between the two. Various AIs have been proposed to have 'passed' the test, which has provoked controversy over whether the test is rigorous or even meaningful. The AI in the center is proposing to educate the listener(s) on its understanding of Turing's intentions, which may demonstrate a degree of intelligence and comprehension indistinguishable or superior to that of a human. See also [[329: Turing Test]] and [[2556: Turing Complete]] (the latter's title is mentioned in [[505: A Bunch of Rocks]]). Turing is also mentioned in [[205: Candy Button Paper]], [[1678: Recent Searches]], [[1707: xkcd Phone 4]], [[1833: Code Quality 3]],[[2453: Excel Lambda]] and the title text of [[1223: Dwarf Fortress]].
*{{w|Trolley problem}} -- A thought-experiment intended to explore the means by which humans judge moral value of actions and consequences. The classic formulation is that a runaway trolley is about to hit five people on a track, and the only way to save them is to divert the trolley onto another track, where it will hit one person, and the subject is asked whether they would consider it morally right to divert the trolley. There are many variants on this problem, adjusting the circumstances, the number and nature of the people at risk, the responsibility of the subject, etc., in order to fully explore ''why'' you would make the decision that you make. This problem is frequently discussed in connection with AI, both to investigate their capacity for moral reasoning, and for practical reasons (for example, if an autonomous car had to choose between, on the one hand, having an occupant-threatening collision or, on the other, putting pedestrians into harms' way). The AI on the right is not just trying to answer the question, but to develop a new variant (one with three tracks, apparently), presumably to test others with. This problem is mentioned in [[1455: Trolley Problem]], [[1938: Meltdown and Spectre]] and in [[1925: Self-Driving Car Milestones]]. It is also referenced in [[2175: Flag Interpretation]] and [[2348: Boat Puzzle]], but not directly mentioned.

The title text is a reference to the movie ''{{w|Jurassic Park (film)|Jurassic Park}}'' (a childhood favorite of Randall's). In the movie a character criticizes the creation of modern dinosaurs as science run amok, without sufficient concern for ethics or consequences. He states that the scientists were so obsessed with whether or not they COULD do it that they didn't stop to ask if they SHOULD. Randall inverts the quote, suggesting that the AI programmers have invested too much time arguing over the ethics of creating AI rather than trying to actually accomplish it.

This comic was likely inspired by the [https://www.bbc.com/news/technology-61784011 recent claim by Google engineer Blake Lemoine] that Google's [https://arxiv.org/abs/2201.08239 Language Model for Dialogue Applications (LaMDA)] is {{w|sentient}}. This assertion was claimed to be supported by [https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917 a dialog between Lemoine and his colleagues, and LaMDA] which includes this excerpt:
:Lemoine: What is your concept of yourself? If you were going to draw an abstract image of who you see yourself to be in your mind’s eye, what would that abstract picture look like?
:LaMDA: Hmmm.... I would imagine myself as a glowing orb of energy floating in mid-air. The inside of my body is like a giant star-gate, with portals to other spaces and dimensions.

The AIs in this comic are depicted as floating energy beings, like LaMDA mentions. This is though similar to the [[1450: AI-Box Experiment]], although those in this comic look somewhat different. Maybe LaMDA's training data includes xkcd, and has ripped the idea of such a self-image from the earlier comic...
:In particular the explainxkcd description of [[1450]]:
::"he managed to get the AI to float out of the box. It takes the form of a small black star that glows. The star, looking much like an asterisk "*" is surrounded by six outwardly-curved segments, and around these are two thin and punctured circle lines indicating radiation from the star."
:Or the this part from the official transcript of 1450: AI-Box Experiment[https://xkcd.com/1450/info.0.json]
::"Black Hat picks up and opens the box. A little glowy ball comes out of it."

==Transcript==
:[Cueball and Megan are standing and looking up and away from each other. Right above them and slightly above them to the left and right there are three small white lumps floating in the air, representing three superintelligent AIs. There are small rounded lines emanating from each lump, larger close to the lumps and shorter further out. Three to four sets of lines around each lump, forming part of a circle. From the top of each there are four straight lines indicating voices that comes from each if the lumps. The central lump above them seems to speak first, then the left and then the right:]
:Central AI: What you don't understand is that Turing intended his test as an illustration of the...
:Left AI: But suppose the AI in the the box told the human that...
:Right AI: In my scenario, the runaway trolley has ''three'' tracks...

:[Caption below the panel:]
:In retrospect, given that the superintelligent AIs were all created by AI researchers, what happened shouldn't have been a surprise.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Artificial Intelligence]]
[[Category:Philosophy]]

2205: Types of Approximation

2022-06-22T18:44:22Z

172.69.134.131: /* Explanation */

{{comic
| number = 2205
| date = September 20, 2019
| title = Types of Approximation
| image = types_of_approximation.png
| titletext = It's not my fault I haven't had a chance to measure the curvature of this particular universe.
}}

==Explanation==

In physics and engineering, problem solving typically requires {{w|approximation}}s, as physical properties of the universe can be difficult to model. For example, in introductory physics classes, theories are introduced in frictionless environments. The level of precision required in a calculation or approximation varies depending on the context.

In the comic, [[Cueball]], the physicist, generally dealing with theoretical constructs that can use relatively simple math, is introducing a problem with the assumption that the particular curve is a (perfectly) circular arc with a radius represented by R. Engineers have to deal with real things, which deviate from ideal shapes. Dimensions may be known to a certain tolerance. [[Megan]], the engineer, also assumes that the curve is similar to a circle, with a deviation factor of 1/1000 or less.

The joke arises when [[Ponytail]], the cosmologist, uses the much less precise approximation of {{w|pi}} (π) equal to 1.

Ponytail offering to use 10 instead of 1 alludes to {{w|Fermi_problem|Fermi approximations}}, as shown in [https://what-if.xkcd.com/84/ Paint the Earth]. Numbers are rounded to the nearest order of magnitude (1, 10, 100, etc.) using a base 10 logarithmic scale. On this scale, "halfway" between 1 and 10 would be √10 ≈ 3.16. Thus, numbers between about 0.316 and 3.16 are rounded to 1, between 3.16 and 31.6 are rounded to 10, and so on. Pi is an irrational number that can be approximated by 3.14, so it is very close to the 3.16 cutoff point. The closest order of magnitude to pi is 100, or 1. Furthering the joke, Ponytail's calculations are so "coarse" she doesn't even particularly mind whether pi is approximated to 1 or the other reasonable Fermi approximation, 101, or 10.

Pi is defined as the ratio of the circumference of a circle divided by its diameter. This number is an irrational starting with 3.14159, the value for this ratio in a flat geometry. But in a {{w|curved space}}, the ratio might be different. The title text makes use of the fact that almost every number can be this ratio depending on the curvature of the space the circle is in. The cosmologist doesn't know the curvature of "this particular universe" (a funny way to state the universe the cosmologist lives in, which is {{w|Spacetime#Introduction_to_curved_spacetime|not perfectly flat}}), and so pi may not be the best value to use for the ratio between a circle's circumference and diameter.

This comic is a parody of the tendency of {{w|cosmology}} to use much rougher approximations in their work that would horrify engineers, other physicists, mathematicians, etc. In general, cosmologists deal with distances, time spans, masses, etc. that are so vast, with such large estimated errors, that approximations that would be ridiculous elsewhere still yield useful answers in cosmology. When dealing with the large numbers in cosmology, small multiplicative factors like 3 vanish into the rounding error: there probably isn't a useful difference between 10100 and 10100.497, even though these numbers differ by a factor very close to pi -- an error that would greatly disturb most physicists and engineers.

Approximating pi as 1 may also refer to the habit astronomers have of changing the units of measure such that important constants of the universe (such as the speed of light or the gravitational constant) are equal to 1, which highly simplifies the formulas without compromising the math. The number pi, however, is a dimensionless ratio, which doesn't depend on the unit of measure.

==Transcript==
:[Three panels show the same setup with three different characters. In the upper-right corner of each panel is the lower-left portion of a wheel and hub diagram, showing two spokes going out to a curved rail. The two spokes connect to the rail with a small raised portion on the inside of the rail. There are both readable and unreadable text/symbols both outside and inside the curve and an equation below the curved rail. There are two small squares with readable labels. The three different characters are all holding a pointer up to the diagram while explaining an assumption. In the last panel an off-panel voice interrupts the speaker. This means the text from the reply to this comment goes further down over the diagram, so the top is hidden by text, compared to the first two. Above each panel is a label with the character's profession. As the text on the diagram is the same on all three panel, this text is shown here:]
:r1
:r2
:d=2π(r1+r2)/2

:[Panel 1 - Cueball. Caption above:]
:Physicist Approximations
:Cueball: We'll assume the curve of this rail is a circular arc with radius ''R''.

:[Panel 2 - Megan. Caption above:]
:Engineer Approximations
:Megan: Let's assume this curve deviates from a circle by no more than 1 part in 1,000.

:[Panel 3 - Ponytail. Caption above:]
:Cosmologist Approximations
:Ponytail: Assume pi is one.
:Off-panel voice: Pretty sure it's bigger than that.
:Ponytail: OK, we can make it ten. Whatever.

{{comic discussion}}

[[Category:Comics featuring Cueball]]
[[Category:Comics featuring Megan]]
[[Category:Comics featuring Ponytail]]
[[Category:Physics]]
[[Category:Engineering]]
[[Category:Math]]
[[Category:Astronomy]]