23Jan/1250

Sustainable

by Jeff

Image text: Though 100 years is longer than a lot of our resources.

This is a graph of the usage of the word "Sustainable" in English in the United States each year.  And as you can see, Randall extends the graph to the point where sustainable will be used as every word.

Sustainable has been increasing in use as people of the US are concerned about making sure that Earth's resources are not totally exhausted in the near future by developing sustainable development.  Sustainable development (using wikipedia here) is a pattern of growth in which resource use aims to meet human needs while preserving the environment so that these needs can be met not only in the present, but also for generations to come.

As Randall mentions in the image text, the 100 years it takes for the word sustainable to get to 100% usage is a lot longer than most of our non-renewable (and non-sustainable) resources will last on the Earth.

Comments (50) Trackbacks (0)
  1. You’d think with the word “sustainable” in every sentence of the explanation, that 2109 were here.

  2. Here you can obtain these numbers: http://books.google.com/ngrams/graph?content=sustainable&year_start=1950&year_end=2011&corpus=5&smoothing=0

    IMHO, a Weibull plot is more apropriate (it never reaches 100%), but will ruin the joke :)

  3. In 2061, the curve is clearly below 1%. Randell must be writing very long sentences.

    • I had a professor from the UK who averaged well over one hundred words per sentence. When I asked another professor why us students were expected to have fewer words per sentence while the UK prof had so many, I was told that the UK schools have different writing norms.

  4. >This is a graph of the usage of the word “Sustainable” in English in the United States each year.

    I don’t mean to nitpick, but what do we have that suggests this is only in the US? I note that from Manso’s link, there is a result distinction between British and US English. However, we don’t know (do we?) which set Randall used, and that said, even American English language texts don’t 100% correlate to uses of “English in the US”. Not a major flaw in the explanation, I just thought I’d point it out.

    • I also stand corrected; there appears to be a general “English” that I presume amalgamates British and American. That said, still not sure why we assume it’s US usage.

      Also, technically he “extrapolates” the trend line of the graph, rather than just “extends the graph”

    • What do we have? Maybe the y-axis, which reads “Frequency of Use of the Word ‘Sustainable’ in *US* English Text, as a Percentage of All Words, by Year”.

      While your argument that American English texts do not perfectly match English text in the US remains valid, I do not know how the Google Ngraphs application works; I would expect it to search English books published in the US, but I may be wrong.
      In one way, Google’s use of language (in the search criteria) would be flawed; in the other way, it’d be Jeff’s. Who can tell for sure?

      • Fair enough with the “US” english text. I did not notice that detail. As I said, it was simply a minor point in any event. Thanks for noting that.

    • That is one of my top two all time favorites. I share it all the time. At cocktail parties…at family parties…at wedding receptions. All the time.

  5. PS: Fun facts from that ngram search (cool page) – I note that it is case sensitive. This can skew results a bit from words that start more sentences than others.

    * The word “the” has consistantly been just over 5% of all word usage in American English since 1800, although since about 1970, it has dropped a bit to just above 4%. Not sure what would cause that to occur. Similar trend with “and” between 2 and 3%, although the slight drop occurs earlier.

    I wonder if it’s just a symptom of the general attempt to be more wordy these days? Frankly I would have expected less wordy books these days written in more casual English than in the 1800s in which formal English with more obscure words would be common.

    * The words “thee” “thou” and “thy” all have notable drops since 1800, although none were very highly used even at 1800. The word “you” has shown not much increase since 1800 in British English.

    * On an aparently unrelated note, in American English, the word “you” has nearly doubled since the 70s. I have few explanations for most of these, but I find them interesting.

    * There is a very big spike in the words “raptor” and “raptors” around 1991 and increased usage through the 90s.

    * “McFly” sees virtually no usage until the 80s, but it really spikes in 1991, 1994 and 2002

    * “pizza” sees an increase from pretty much nothing starting around 1950

    * The (capitalized) names of various presidents jumps significantly around their elections (esp. in American English), as might be expected.
    ** The use of Bush jumps up some what in the 90s, but way more in the 2000s
    ** Clinton has very jumpy usage from 1800 on. It seems to be used as much in 1842 as in its 1994 peak, although it settles down during the 1900s before Clinton’s election
    ** Eisenhower rises from virtually nothing around 1940. Reagan rises from virtually nothing in the 70s. Nixon rises from a low amount in the 50s, but even Nixon some consistant (minimal) usage since 1800.
    **”Washington” has seen a steady increase since 1800, but nothing comes close to its peak usage in the late 1770s where it peaked at a whopping 0.1% (I see that pre-1800 stats are a bit skewed though. Perhaps political documents make up a significant sampling pre-1800. In any event, all results pre-1800 seem to be very jumpy from year to year.
    **”congress” peaked between 1800-1840 and has significantly dropped off, whereas “Congress” (caps) increased since then and stayed steady until a dropoff around 1970 (not sure why). Similar story for “senate” and “Senate”
    * There doesn’t seem to be any modern-day jump for “Colbert”, although there are peaks in 1837, 1865 and 1939.
    * “nuclear” takes off around 1950, but it is notable to me that there is usage climbing from 1880 onward, and even some apparent blips from around 1820.

    I could go on, but I really ought to get moving on other things. Cool toy!

    • Also, “computer” climbs from zero starting with a slow increase from 1940 to a sharper climb in the mid-50s to an early 80s peak. Since 2001, it actually appears to be dropping notably. The word “pirate” bottoms out around 1980 before climbing quite a bit again in the 2000s. Different kind of piracy, I suspect (although the high-seas-variety did see a bit of a news comeback in the last few years).

      • One more fun one. “internet” has this beautifully smooth upward curve starting slowly from the mid-70s until it skyrockets in the mid-90s and then nearly flatlines in the early-00s. But for some reason, there’s a minimal usage around the turn of the century – by which I mean around 1900! Weird!

        • Just a heads up: Jesus’s numbers dropped throughout the 1900s, but have been climbing again since the 80s. Look out 2nd coming…

    • I hate to say this, but TL;DR :(

  6. Dan Loeb, Sentences will have to get longer to help compensating for the ethropizing effect of random “sustainable”s thrown in. The need for embedded, in-sentence, word-based hash values will arise. Between 2100 and 2108, “Light speed” will be the only other word in use, which will make our language somewhat binary, for example xkcd will become:
    sustainable,light speed,light speed,light speed,light speed,sustainable,sustainable,sustainable,sustainable,light speed,light speed,sustainable,light speed,sustainable,light speed,light speed,sustainable,light speed,light speed,sustainable,sustainable,sustainable,light speed,light speed,sustainable,light speed,light speed,sustainable,sustainable,light speed,sustainable,sustainable,

    (Original research, Source:)
    http://books.google.com/ngrams/graph?content=light+speed&year_start=1900&year_end=2008&corpus=5&smoothing=0

    Since everyone knows that light speed≠sustainable in non-quantum physics, there will be no more communication errors. i, for one, am looking forward to this, and am considering to remain in a stasis chamber the next 50 years. When i get out, i’ll inform the world of my copyright on any combinations of the words “light speed” and “sustainable”, which will make me the boss of the world (or another equally important position).

  7. What I find interesting is that, although the x-axis increases steadily at a rate of 20 years per unit, the y-axis increases by one order of magnitude per unit. Presumably because a straight line scatter graph is more impressive than an exponentially shaped scatter graph. Would you call the y-axis a logarithmic scale, then?

    Also, I’d love an explanation of what an ngram is…

    • Yes, the y-axis is logarithmic; otherwise the graph would get too large in the y direction to show it. If y were spaced evenly, there would be no way to show the later years (2109, etc.) without completely losing the earlier years (1960, etc.).

    • A logarithmic scale on the y-axis is used to make exponential growth look linear. Anyone can look at a graph and visualize how to extrapolate a straight line, but other regression models are less obvious. So it helps to use logarithmic scales, to turn whatever regression you’re doing into linear. Exponential regression uses a logarithmic y-axis, logarithmic regression uses an exponential x-axis, and power regression uses logarithmic scales on both axes.

    • Not the first time Randall used that scale.
      http://xkcd.com/482/

  8. Has this comic been drawn in 2009? Why else pick 2109 in combination with the image text?

    • Presumably, Randall meant “approximately 100 years.” 2109 is simply the year his extrapolation hits 100%.

      • Yes, it would not surprise me if math-geek Randall actually input all of the actual data points for each year into Excel (or other) and plotted the real numbers to find that 2109 would be the point when 100% is reached.

      • Note the rise is not monotone (mathematically speaking) so there is an error margin, which is probabely much larger than 2-3 years.

        And the comic itself hints that the rise must flatten at some point anyway.

  9. So apparently we had email back in the 1800’s, it died out and was reinvented in the 1980’s…

    http://books.google.com/ngrams/graph?content=email&year_start=1800&year_end=2008&corpus=5&smoothing=10

  10. Am I the only one wondering what happened in ~1966?

  11. Sustainable. Sustainable sustainable, sustainable.

  12. “developing sustainable development” I always say it like this and then cringe afterwards reading it back.

  13. The amount of growth this chart predicts is not sustainable!

    • it makes sense – after 2109 we’ll just go back and rewrite previously published books to only contain the word sustainable. I’m actually working on the now.

      “Sustainable sustainable Sustainable” by Leo Tolstoy

  14. Sustainable from Sustana-January 2100,

    Sustainable sustainable sustainable sustainable, sustainable sustainable sustainable. Sustainable sustainable? Sustainable!!! Sustainable sustainable sustainable :)

  15. Sustainable Hollister, Aero Aeropostale sustainable Malkovich Mal Malkovich.
    Pika? PikaCHU!

  16. Malkovich Malkovich Malkovich?
    Malkovich Malkovich Malkovich Malkovich: Malkovich Malkovich
    Malkovich Malkovich!
    Malkovich Malkovich Malkovich MalkovichMalkovich Malkovich Malkovich

    ~Malkovich

  17. I assume the comic is poking fun at the studies which show that all of our resources will be exhausted in 100 years.


Leave a comment


Anti-Spam Protection by WP-SpamFree

No trackbacks yet.

Pages

Facebook

Blogroll

Categories

Meta