in reply to
Recursive search for duplicate files
Much better way is to use MD5 for file comparison.
If you care about performance, you might just want to hash the first 5% (or the first 1k or whatever) and see if there are any collisions, and if there are you can still look at the entire file.
The same name on every site (unless it is taken already)
A different name for every site
The same name for sites I don't care about, but different names on sites I do care about
Different names for sites I don't care about, but the same name on sites I do care about
The anonymous account
Someone else's username
Root, after hacking into the site
Results (254 votes),