Re: Distribution of longest common hash prefixes

Linus> Well, neither of the the two objects at 9 bytes may not be (and Linus> probably aren't) commits and of the 32 8-nibble cases who knows Linus> how many are actually commits (probably none), so an 8-byte SHA1 Linus> is *probably* unique at least if you just look at commits.

Linus> Remove the "--objects" to find out.

That makes for:

0: 1: 2: 1 3: 23320 4: 25431 5: 2259 6: 134 7: 8

But is the kernel large enough to be sufficiently representative?

Did Jon complete an import of the 'zilla src?  

Has anyone tried to import OOo?

-JimC