Maybe. But wouldn't you say this flaw is more likely with a tech platform that breaks down a debate into 140 character long pieces and makes it difficult to see the whole context?

Look at the three tweets Antirez mentions. The second one starts

"the 99% percentile is bad [...]"

Antirez mentions someone took this to literally mean that percentiles are a bad metric. To me, this doesn't particularly require bad faith or malicious cherry-picking; it just requires a bit of carelessness, enabled by the lack of context and no indication this was part #2 of a 3-long series of tweets. And apparently, once the "rebuttals" start piling up, they generate a snowball effect. Not entirely Twitter's fault, but definitely made worse by it, which is what (I think) Antirez is arguing:

> "Once upon a time, people used to argue for days on usenet, but at least there was, most of the times, an argument against a new argument and so forth, with enough text and context to have a normal condition. This instead is just amplification of hate and engineering rules 101 together."