I think this sounds interesting in a lot of ways, but I question your assumption that a team of professional lawyers will agree that you "don't have to read the legalese to understand what your contracts are saying". Presumably the lawyers had a reason for putting the language in the contract to begin with. You might be able to summarize a contract in this way, but in the general case I don't think you can just "translate" the contract into plain English without glossing over parts of it, and parts that lawyers probably think are important details. It's a bit like software or mathematics in that way, the formal language serves some purpose.

- However, maybe the parts that get glossed over are covering edge cases that often don't matter?

Worthwhile (if difficult) idea to implement. In addition to decrypting the legalese speak it would be helpful to standardise your output so that contracts could be compared side by side.

I like the idea too

there are too often terms in contracts which can cause you big problems but which you don't notice before they actually cause you problems, when everything has already been signed.

good luck with that

this idea looks good to me.

may be to validate this further, you can try doing this first for the popular software license terms - Apache, MIT, LGPL, etc. and see how the HN community use it.

sounds like what these guys do : http://www.pramata.com