Hacker News Reader

Comment by kashyapc | Hacker News Reader

Probably that can be a useful insight in some specific circumstances as you describe, but it falls flat as a generalization. Allow me a snark: maybe a group of venture capitalists (many of whom call themselves "angels") might not admit a mistake. But many groups (not as grandiose as the military, though) jolly well admit and own up non-trivial mistakes.

I participate in several open source projects over 13 years now. And as jancsika mentions in this thread, various times I've seen mature groups of contributors and maintainers admit, and articulate, really difficult mistakes in public. Granted, these are small groups ranging from six to twenty-ish. Still, it takes courage and wisdom to do it gracefully.

View on HN

I have yet to make a mistake, but when I do I plan on publicly admitting it.

It seems to miss the mark in other regions, but here in Denmark that is considered a bullseye joke :)

> I have yet to make a mistake

For some carefully chosen definition of mistake.

I believe that was humor, my friend

It's getting dangerously reddit in this thread.

HN has been Reddit for a while now, there's no functional difference.

As far as I'm concerned there's not much reason to care either.

I think that might’ve also been humor? Some dry, mathy humor

The post you replied to was also humor, btw.

I think what separates the two camps is a positive incentive to admit failure, which is where the rarity comes in, as highlighted by other posters. If the group's incentive is integrity and trust, or loyalty and retention, it makes sense as there's a capital (political, monetary, etc.) incentive in doing so.

That's where I think the insight cleanly falls apart and yet holds up pretty well for the vast majority of situations. Rarely are groups actually positively incentivized to admit failure, and therefore they do not, as a cohesive unit, actually admit failure as there's a greater incentive not to, barring force.

Groups can get a life of their own, like a virus. Those who do admit fault can be replaced with others who are vocal about not admitting fault.

The more polarizing, non falsifiable the beliefs of the group, the harder the change.

> I've seen mature groups of contributors and maintainers admit, and articulate, really difficult mistakes in public.

And do it right here on HN. Just the other day a member of the Python core team said the Python2 to Python3 transition was be a never to be repeated mistake.

> Granted, these are small groups ranging from six to twenty-ish.

And big groups do it too, apparently. In community groups I'm in saying things like "we tried this but it didn't work" is a regular occurrence. They make for entertaining, self depreciating war stories, letting newbies know things aren't judged too harshly around here.

In fact it happens so often, I wondering if all the posts here saying "of yes - that's right" is some sort of group confirmation bias at work. I wonder if it will be self correcting - and the group later come to the conclusion it was wrong.

> And do it right here on HN [...] And big groups do it too, apparently [...]

Indeed; all this is old news for anyone who has worked in open source communities for a decent chunk of time (say, six years at least). It's the "angels" that are disconnected from real world that tend to stretch a small observation into a sweeping generalization, in the name of "insight". The cynic in me says, "they're doing it for the clicks and brand-building."

Admitting mistakes is one thing. Admitting that what the organization has set out to do is unachievable is another.

Reply on news.ycombinator.com