Don't Alienate Open Source Contributors
linkedin.com
linkedin.com
I had an opposite experience trying to report an issue to a certain Apache project I won't mention. Gave the exact version of their program in use, all platform information, a sample file with the issue and the exact command to run. Everything to reproduce.
I received an incredibly derisive response about "why nobody thinks to include a stack trace because we're not mind readers, duh" and the issue was closed promptly. Another dev re-opened the ticket and actually ran the command with the file to reproduce, but it already soured my experience. Needless to say neither I or my colleagues will ever consider contributing to that project again.
Not being an asshole is obviously the right basic mindset. But if you're a maintainer of open source, there are a number of other considerations to keep in mind. You really want your project to be as accessible as possible.
People talk sometimes disparagingly about "drive-by pull requests", but these are good things! Your core contributors can only do so much, and you're likely to get only so many people capable of making core-level contributions. You want to take advantage of the long tail as much as possible.
This is especially true when you consider the pool of potential contributors is often much wider than seasoned open source contributors. You'll get people who have never squashed commits before, or never even submitted a pull request. A welcoming mindset is the first step. But you need to be thinking hard about how difficult it is for a new contributor to onboard onto your project.
I know it's a lot of work for maintainers who are already stretched thin. If this is a topic that interests you and you want to chat, feel free to email me. I'd love to help figure out how to make this easier for more maintainers.
In fact, we try to encourage them as best we can and I know that I personally try to be polite and positive about any attempted contributions even if they can't be accepted.
I figure just working out how to get, build and submit patches for LibreOffice takes quite a big of time (build time alone is massive, though getting better) and let's be honest, if you've never used git or source control it can be pretty intimidating to even clone a repository!
So I agree. Someone has taken the time to even attempt to make your software better, and even if that might be misguided it's good to remember that.
Working with a Linux distribution probably eases the burden, but even with all the dev tools and build requirements at the ready a clean build might not arise until the environment is a perfect replica. These are things that regular contributors probably take for granted, they've set this all up previously and the order of magnitude is greatly reduced.
I've always enjoyed the challenge and healthy debate around open source tools, it's been an important part of my life in computing, and ethics are at the heart of it. I also like the restrictions that the tools impose, it's a constraint on creativity, but learning to be intuitive with constrains is also a creative outcome. In turn I've always tried to commit quality bug reports, and patches where I feel I understand the environment enough to fix my own problem. The time that goes into these things is prohibitive and takes a level of patience and good will that most people simply wouldn't bother with.
A projects culture really matters, and I think that open-source projects often are tainted by vicious personalities. But it's far more difficult to hire and fire in this world, a dominant contributor is still important even when their personality sucks, and the amount of work hours devoted to removing them by a community is often not worth the effort of just accepting an ego and warning contributors of the problem. This just continues to poison the pot. It only gets worse when the contributor becomes a lead.
There are projects that you might personally find valuable and that's a good reason to get involved, but everyone has a life outside of open-source. If you're lucky the project is what makes you money and it's in your best interest to quit complaining and do something. If you're contributing in your free time then your relationship will probably be fleeting at best. It's hard not to have a life.
I'd love to work in a culture of expected mutual respect, but it's not a perfect world.. In the meantime let's support peoples initiative for what it's actually worth, be positive and hopefully grow community by encouraging more people to get that little bit more involved. They've probably already taken extraordinary steps in that direction.
I'm not one who thinks that Linus Torvalds is completely out of hand -- in fact, whenever I read the messages and the context for whatever blowup he's allegedly stoking, I find him to be quite reasonable. And maybe that would be the case here after knowing both sides of the story...but the wording of the message received by the author really, really rubbed me the wrong way:
> "You were asked to correct this in 2013, failed to even reply, and it has been archived. Please explain your inaction and lack of the most basic manners."
Sure, I don't know how CRAN works, and maybe the author is an unreliable narrator who is leaving out the part where his "inaction" caused CRAN to experience a week-long downtime...but barring that kind of disaster, the message the author received struck me to be not just callously judgmental, but just...sad. Has this person never been on the receiving end of a missed email? Does this person not realize that life happens, even to people who contribute to CRAN? And that the author may have experienced some serious downtime in his life, such as a death of a loved one, or unemployment, and just simply missed the email? Why the fuck would you harass someone who is actively seeking out to update his contribution in response to users' pleas? How jaded must you be to accuse someone through email of lacking "the most basic manners" when this person is offering help, rather than thinking that just maybe you should give him the benefit of the doubt?
Maybe the author was an inattentive jerk. But there's no reason to assume that right off the bat. Yeah, I know this is a relatively small scuffle in the scheme of things...but that's what really bothers me. One of R's disadvantages is that it doesn't have enough contributing developers...but small communities have their own advantages, and I feel that's why my interactions with R-users have always made me feel good about the language. But the way this CRAN maintainer allegedly acted...it just makes me sad...if someone in a community this small can be so cold to another member, then it's a downer to think how things in much bigger communities that lack the personal connectedness.
But good on the author for taking it in stride...he probably doesn't need R as much as R needs more people like him.
As a maintainer it's not your job to provide free therapy sessions to other contributors but despite the inhumanity of the medium it's still important to remember that on the other end sits a human like you with their own problems, worries and distractions.
There's nothing wrong with archiving the package for inactivity. But the unnecessarily hostile and judgemental tone of the message is inexcusable. There was also nothing wrong with turning down the "iGiveHead" patch someone else mentioned here -- it was simply a matter of how it was rejected.
Open source projects don't have to accept every contribution and they don't have to be endlessly forgiving when contributors don't fulfil their self-imposed duties. But showing at least a minimal level of respect should be basic human decency.