How to BURN OUT Open Source devs: After SENDING a patch, IBM asks me to also do a TESTCASE for FREE!

[ivankmeto]

I burned out and stopped caring years ago. You can report the issue, file proper and verbose bug report, fix the issue yourself, comply with all the coding guidelines, send in the patch and at the end of the process is some person who just rejects it because of some very stupid reason or he tells you there's more work you have to do (for free) if you want it merged. In other cases you won't get reply for years. Just say no. (F)OSS should be more open and somewhat fun to work on, instead you get coding bureaucrats who drain life out of you. No thanks, it's not worth it.

[zelllers]

You don't know me, this is a very one-sided relationship, but I love you and appreciate all you do.

[maze_]

It's always nice to have fixes accepted upstream, because if they change something it will probably conflict with your local fix, however it does take a lot of effort to get things accepted.
Just continue sending your fixes and state that you will not follow up on it. If they care, they'll take care of it. If they don't care, they don't deserve any more effort on your part anyway.
Thank you for your work!

[Angryblueflamingo]

Yeah that's a ridiculous ask, you would think a well-paid presumably pretty distinguished IBM engineer would be able to fix whatever nitpicks he has about your patch in an hour or two lol. These guys get paid a lot of money and are very skilled, they can do this themselves, especially when someone just gave you the solution. It doesn't get any easier than that, refactoring a working solution.

[Bruno_Haible]

Many FOSS projects, from GCC to xz, are understaffed. Their request for patches via the mailing list, for a unit test, etc. is meant to ramp you up as a contributor and co-developer. When I have no interest in becoming a co-developer of that project, I tell the people explicitly: I have done my duty by reporting the bug, now all the rest, from the unit test to the doc update, is your turn. It works! When you give them a finger and they want to pull your whole hand, just say NO!

[kng-gx]

To be honest as a student who likes tinkering with os, compilers, linux, this is really so demotivating I hope this video (and others) make positive impact in workings of these organisations.
I never imagined It could take such a huge toll on mental health.

Hey, take care wishing you good luck.

[fluffyunicorn]

My solution is to have as few dependencies as possible. If you have a dependency where upstreaming bugfixes is impossible or takes too long, try to replace it.

There are cases where this does not work, of course. I wrote iOS software ten years ago and found bugs in Apple's libraries. In the beginning, I filed them with the official bug tracker. I never got any response. After a while, I discovered this was normal and experienced Apple developers would never open a bug in the Apple bug tracker.

[austist]

i havent utilized any of your work, but as a viewer I respect what you do and appreciate it. You do good work Rene, and you're a selfless soul. Do what brings you peace and contention .

[jonathanhirschbaum]

I love those beautiful oddballs of open source community. Unconditionally. I'm pretty sure Rene, you have no idea how many people owes you endless thanks and moral support

[FlamerOHR]

good on you for still caring and pushing it upstream.
I've tried contributing to open source projects, which were also commercially backed by a company, and the attitude I get back from them is exactly like this - they expect our efforts to match their's in every level and overlook the very premise of the contribution, to fix the bug found!
I gave up after a year, I'm not even getting a thank you, nevermind getting paid for contributing.

[ThePredator]

This is so disrespectful 😤. Thanks for making this video. So much essential labour isn't paid.

[valentinziegler]

Next level multi billion dollar cooperation: pay premium support fees for them to even consider your bug reports (i.e., Microsoft model)

[JTordur]

Someone should organize a FOSS Strike, I don't think the suits in the big companies actually have any idea of the amount of volunteer work they benefit from.
Striking would really show them that.

By the way, FOSS developers working these projects for free, you really are the giants upon which shoulders we stand on.

[xyzabc-ol]

open source software developers need to become open source hardware developers and even open source manufacturers of simple chips eventually.

it is too inefficient and unpredictable to work with big corporations these days

[michaelthompson]

the thing with these corporations is the people maintaining these projects probably only superficially care about open source (like me) and cannot empathize with anything you say in this video. your frustration was really well communicated in this video

[CaptainWumbo]

This is interesting on two points:
I remember being asked in an interview once if I had ever submitted to open source before. I told them no, I don't work for free. Not that I don't create and share things for free sometimes, but I don't do work for others for free.

2. What you're experiencing is actually how these companies work internally (and all the same criticisms are valid). They are doing the review the same way they do internal pull requests, with stupid nitpicky comments included. And what you suggested actually does make a lot of sense, if a reviewer wants changes to the code they should make the changes themselves and ship it. Because that would realign motivations where nobody wants to waste their own time on stuff that doesn't matter but they are happy to waste someone else's time.

[MatthewHolevinski]

I would of talked so much trash in that bugzilla, ask someone to CC other people, lolwut? No they can monitor the tracker themselves, oh I would of went off.

[Schadowofmorning]

Every time I hear stuff about how linux is maintained & developed a part of me dies.
This bureaucracy around getting one thing fixed would make the soviets proud.

[roberth]

Yeah, that does seem inappropriate. Worse than ignoring value (which decays like bitrot), they are not even aware of the value... which means they don't actually have a wholistic view of their own product (only those parts that are critical to their business). When they do finally address the matter, it's like they are sacraficing their time to do a code review, and addressing you as a lazy employee not following procedure, or an intern that should know better and is not following their onboarding directions.

[foobarf]

I feel you, gave up on rust on ppc64 due to SIGILL on power5 because the AltiVec instructions are not mandatory in the arch, just common (embedded power seems special/rare).