Scores text based on how obscure the words are.
Get the site/obscure_score.zip. unzip it (to a .html).
Open the html in a browser.
Type some text in.
Hit score.
We use the words.txt from http://number27.org/assets/misc/words.txt.
Each word has a rank, the higher the rank, the more obscure.
We add up the ranks.
We put the entire words.txt in the .html. We didn't try to make it smaller. Ouch.
Makes the .html:
make
Clean everything except the downloaded words.txt:
make clean
Refetch words.txt
make refetch
- use a perfect hash
- maybe compress/uncompresss the hash data in memory
- phase 2 is backend etc.