Editing 856: Trochee Fixation

{{w|Huffman coding}} is a lossless data compression algorithm that works by organizing characters into a tree structure (called a Huffman tree) with the most used characters in a string closer to the top. The characters in the string are then replaced by the sequence of bits representing their place in the tree, allowing for characters that are used very often to be represented with only a handful of bits compared to the 16 or 32 bits usually needed (depending on the character set used). In highly repetitive data this can cut down the file size immensely, which is what Randall is implying by saying you would only end up with 30–40 bytes. Most of the "[[221: Random Number|random]]" stuff said on the Internet has been said before, and isn’t particularly random either, following predictable patterns.

{{w|Huffman coding}} is a lossless data compression algorithm that works by organizing characters into a tree structure (called a Huffman tree) with the most used characters in a string closer to the top. The characters in the string are then replaced by the sequence of bits representing their place in the tree, allowing for characters that are used very often to be represented with only a handful of bits compared to the 16 or 32 bits usually needed (depending on the character set used). In highly repetitive data this can cut down the file size immensely, which is what Randall is implying by saying you would only end up with 30–40 bytes. Most of the "[[221: Random Number|random]]" stuff said on the Internet has been said before, and isn’t particularly random either, following predictable patterns.

[http://blog.xkcd.com/2011/02/04/trochee-chart/ On the blag], Randall published statistics about the occurrence number of certain combinations (now obviously inaccurate).

[http://blog.xkcd.com/2011/02/04/trochee-chart/ On the blag], Randall published statistics about the occurrence number of certain combinations (now obviously inaccurate).