Lossy text compression (hackaday.io)

I like this. Would love to see the whole bible done, and online. (I'm not religious btw, just curious).

Interesting project - it could be taken further and apply some form of txt speak dictionary (E.g. Great->gr8 etc)

Since reading doesn’t actually involve a letter-by-letter examination of the material, I wonder how a lossy compression algorithm that used words with the right shape would do compared to lossless compression?