Did you guys run any sort of analysis like this, on comment content in aggregate, during your tenure? I would guess with all of that data it'd be fun for hackathons to come up with interesting conversation starters like this one, despite the flaw you point out.
Glad folks are putting together these datasets now, too, if informally.
No; we were usually too busy just keeping the site up and trying to generate revenue.... though actually, there were occasional times where we said "fuck that" and took a break to work on something fun, like an April Fools stunt.
If anyone did such an analysis, it would have been ketralnis.
What percent of such threads contain a comment that mentions potatoes, or bismuth, or juggling?