But that's not even the greatest flaw. The greatest problem is expecting people to do a bunch of research to dig through past archives (thousands and thousands of jokes?) to determine if any of those are a substantial match with this one.
There are MANY flaws with Bennett Haselton's idea. There always are. That's the problem with his posts getting front page placement.
Anyway, let's deal with them in order of time.
1. Who is going to sign up to read a list of JOKES just because it promises "new" jokes in exchange for your work?
2. Who is going to SUBMIT jokes FOR FREE to that list to reach those readers?
3. Who is going to do the work FOR FREE of documenting duplicate jokes?
4. Who is going to do the work FOR FREE of verifying the documentation of #3?
That's why I think sock puppets will be the biggest exploit of those flaws. Anyone incentivized to reach the people in #1 will be incentivized to game it by creating sock puppets. It won't take many sock puppets because most people will soon give up on the effort to document a joke / moderate the documentation of a joke.
Eventually you will end up with sock puppets documenting jokes and sock puppets moderating that documentation. And the list of jokes will devolve into whatever personal vendetta the guy with the most sock puppets has.