Hacker News Reader

Comment by turbinerneiter | Hacker News Reader

I have a theory without data.

Is there a difference in burnout-rate between the new school of Open Sourcers and the old school of Free Software people?

I feel like the stories of people quitting, either due to the insane workload or because of the abusive "community", are mostly people who publicly hack on stuff that is on GitHub.

The old folks who quietly drop their code in an SVN repo only the distributors know how to find seem fine.

Obviously, I have no data on that, but to me it seems the whole "social coding" thing on GitHub ... sucks for the people who do the actual work.

View on HN

I think this long predates github. Sure, it might be less of an issue if it's just a one off code drop - but there've been maintained open source projects LONG before github.

To me this is much more about reaching a (sub-) project size where there's no realistic way to keep up with everything you feel you ought to. The guilt of not getting around to review somebody's patch (even though that's how you got started), the guilt of not fixing a bug in code you last touched / wrote, not keeping up with subsystems you created, is profound.

I think this more likely an issue for people with personally traits that make them more likely to become maintainers.

I've seen this in companies too - the difference is that in open source it's often the "relaxation" time that's taken over by the guilt/work.

Edit: repair some autocorrect damage

I think the culprit is GitHub notifications. If you are not careful they will pull you in all sorts of directions.

There are some days where I set myself to a task. Open GitHub. Quickly check the notifications because the blue dot grabs my attention. Three hours later it's time to go to bed and I haven't done anything satisfying.

It's fine to fix other people's problem but it also prevents from building deeply meaningful things. Most of the interactions on GitHub are superficial issue discussions.

> I have a theory without data.

Me too :D

I've noticed signs of fanaticism within the Rust community. In response to many, if not most, articles about: new languages, new operating systems, or even old languages, or old software projects... I see folks commenting about Rust. Often times, it's pretty rude -- e.g. asking "Why isn't this done in Rust" in response to an operating system written in 2014. It smacks of zealotry.

And then we see people quitting due to burnout, and from abuse within the community. This makes a lot of sense if the community is, in fact, being driven by zealotry -- such communities tend to consume themselves.

> Is there a difference in burnout-rate between the new school of Open Sourcers and the old school of Free Software people?

From the old-school communities I've been part of, no. New-style communities are just more public. Old-school "stepping away from burn-out" posts went out over private mailing lists and IRC channels. These days, they go over blog posts and GitHub issues, so the public has a lot more visibility into them.

My suspicions are compatible with yours. Last year I jotted down some thoughts about a possible solution:

> Zero-obligation communities could probably extract more value from their contributors (including converting passive bystanders into contributors) by fostering a culture of voluntoldism and getting maintainers to put in some asks of their own. Right now, people are half-insisting that there's something intrinsic to open source that is placing a high burden on maintainers. Mostly, I regard this as the result of bad practices originating from the culture of GitHub, not intrinsic to open source. But either way, it's widely reported that maintainers are burning out. Users are showing up and asking for too much. How about reversing polarity and start conditioning maintainers into asking for things from users and less involved contributors? The idea is for maintainers to give consideration to the most effective way a project would benefit if it had at its disposal a voluntary group of mechanical turks.

https://www.colbyrussell.com/2019/02/15/what-happened-in-jan...

There is also the expectation these days that when your project grows big enough, the unicorns are supposed to step in with funding. Considering MSFT is evaluating and using Rust now, paying a couple maintainers would hardly unreasonable for them, the community dropped the ball here. Mozilla and the Rust community need to step up their fund-raising game, losing product managers left and right is simply terrible for the ecosystem. Just a fortnight ago we lost the lead developer of one of the fastest web frameworks in existence over ideological conflicts. For good or for bad, libraries don't code themselves, every effort needs to be celebrated and applauded.

There is no Rust foundation.

I think that the old guard of open-source greybeards were generally a lot less responsive to public feedback, and more apt to tell people that were being obnoxious to go pleasure themselves, rather than pivoting into PR mode and trying to placate.

Yes. And now this whole code-of-conduct business would probably have these perfectly reasonable greybeards banned from their own projects. Gone are the good old days of open source.

Reply on news.ycombinator.com