[will_hughes]

> Their server could trivially query the url and store the result hash.

I'm not sure what you mean here? I've not seen a site tell you "We don't use this query parameter" if you stick an additional param on there.

[antsar]

I think the idea is to detect when different URLs contain the same content. That defends against duplicate entries like example.com/?foo and example.com/?bar (which are the same page).

[mod]

Fetch the page, hash the contents, compare hashes.