>>10271431 (pb)

Interesting post. "Bean counter" admits that supposed pro-Q "bots" look "bot-like" but do not actually seem to be bots... confirming the conjective of another anti-Q bean counter.

There seems to have been a CONSIDERABLE upsurge in anti-Q bean counting recently (past days/weeks). On my recollection, up to fairly recently, this was almost non-existent. From the time the MSM first started attacking Q in spring 2018, and then for maybe two years, I can remember literally ONE attempt to do some "data analysis" critique of Q and this was VERY early and was by someone looking at interconnections between reddit groups and posters on /r/greatawakening. It was methodologically flawed, but at least he was trying... But then I didn't see almost anything like that at all until fairly recently, and now it seems to be popping up all over. Actually I think a few of these twitter posters have been doing this stuff over time, but they didn't get traction, and now they seem to be getting it.

>>10271526

Should have added this….

This points to what should in principle be an obvious fact… that algorthims that supposedly detect "bots" depend on HUMAN assumptions about what human online activity SHOULD look like. And since much of social media is a bunch of CRAP, these assumptions are derived by generalizing over the activity of people who are acting like morons to begin with.

GIGO!