[derefr]

Hackers, in my experience, have more shared experience in being naive enough socially to allow themselves to be baited into making statements which are later used against them, than in actually saying things (or even associating with others who would say things) which are intentionally malicious.

We all understand fixing someone's computer and then being blamed for any problems it has from then on; or for making a software estimate and being held to it as a requirement. In short, hackers are used to being scapegoats, so the first conclusion we jump to when one of us is accused of something is that they're being scapegoated too.

This can look like "victim-blaming", but it's really more-than-anything a yearning for people to keep their standards of evidence high, and to avoid jumping to conclusions. Counterarguments and counterfactuals ("what else could have happened") aren't presented as "this is obviously what happened instead", but just to decrease confidence that the evidence presented thusfar is strong enough to prove anything much. If evidence A fits alleged narrative X, but also random equally-likely stories Y or Z, then X is no longer implied directly by A.

[rsynnott]

> This can look like "victim-blaming", but it's really more-than-anything a yearning for people to keep their standards of evidence high, and to avoid jumping to conclusions.

That really depends, now, doesn't it, on what the subject is? :) HackerNews seems, in general, to have a very low standard of evidence for things like "the NSA is doing something evil", and tends to be quite credulous on such things. It seems to have a very, very, _very_ high standard of evidence for "bigotry exists" though.

[derefr]

That's certainly one way to frame the reference classes involved. Let me try an alternate one:

> HackerNews seems, in general, to have a very low standard of evidence for "[large organizational structure] is doing something evil", and tends to be quite credulous on such things. It seems to have a very, very, _very_ high standard of evidence for "[individual] is doing something evil" though.

I think this succinctly predicts more of HN's behavior, personally. HN is just generally afraid to condemn individuals for much at all.

That doesn't exclude your conclusion from also being true, though: HN can be both resistant to condemning individuals, while also being biased against believing that bigotry is a thing that exists. But assuming that either one is the "whole cause" of any particular effect, will cause effects to seem weirdly large, because of the confounding effect of the other.

[pekk]

As a generalization this might be true, or not... but likely a lot of people commenting here just have a globally high standard of evidence. Why are they deleted from the picture?

[erikpukinskis]

What is sexist is that I think many people have not actually invested the energy in figuring out what the distribution of X, Y, and Z are. I think X is actually much more common in this kind of scenario than Y or Z, and honestly I think anyone who has actually gained the trust of a few women in tech and heard stories about what it is like is pretty likely to see that too.

Anyone who throws their hands up and says "well I guess X, Y, and Z are equally probable" to me clearly has not actually spent time understanding the problem.

[derefr]

You're doing exactly the thing people-who-are-oft-scapegoated are afraid of: equating "extremely probable" with "incontrovertibly proven."

In an actual trial, for a conviction to take place, evidence would have to be presented that leaves no room for reasonable doubt--which is to say, the evidence would have to effectively take Y/Z/etc. out of consideration.

We chose this standard because we, as a society, decided that there was something more important than making sure "justice is served" in all cases: making sure nobody has the power to condemn innocents at a whim by making up an "extremely probable" accusation.

But the court of public opinion has no such strict standard, even though the punishments it hands out can be far worse.

[erikpukinskis]

No, probable is probable. I'm happy to talk about unlikely corner cases, in an appropriate sidebar. I think anyone pushing them as a primary talking point has their priorities misaligned and is tilting at windmills.

[jd0]

This can look like "victim-blaming", but it's really more-than-anything a yearning for people to keep their standards of evidence high

I don't buy it. The "hacker type" despite self-professing as a logical bunch falls prey to all of the logical fallacies and cognitive biases that any other opinionated group does. The evidence is already out there, this is one of many incidents following a trend of sexism in this industry that has been going on for decades.

Github does not exist in a vacuum and real life social interaction doesn't use variables beyond a distant abstract analysis.