The cruel crucible of open source development

The open source development community is not the vicious hellhole some describe, but what goes around often comes around

open neon sign
Credit: tribalicious via Flickr

It was with great interest that I read Lennart Poettering’s missive on open source software development and his experiences in the field last week. There is some truth to his claims that discussions can get heated and things can be said that are perhaps uncouth and salty. Still, I can't agree with his wholesale characterization of open source development as "sick" or “full of a**holes.”

I think, perhaps, that Lennart has been exposed to more troublesome technical discussions and descriptions than many in this field, but not because he is an innocent target. Rather, he conducts himself in such a way that it evokes this kind of reaction. To paraphrase an old saying, “If you run into a jerk in the morning, you ran into a jerk in the morning. If you run into jerks all day long, then maybe you’re the jerk.” I think it may in part apply here.

Technical discussions aside, a failure of public relations has surrounded systemd, alongside Poettering and a few other developers. In most cases, we attribute open source development and big open source projects to a team of people, with some names occasionally moving into higher headspace, such as Linus Torvalds. But with Poettering (and Kay Sievers to a lesser extent), their collection of public interactions and issues with prior projects have appeared to paint them in an increasingly bad light, and they are becoming the focal point of a groundswell of disapproval regarding a number of rapid, fundamental changes surrounding Linux and mainstream Linux distributions. 

A large number of veteran *nix admins view fundamental changes such as systemd with skepticism due to a slew of technical reasons, as well as the fact it was adopted so quickly and widely in the face of significant opposition, lack of long-term testing, and poor documentation. When a key systemd developer like Sievers has been banned from contributing to the Linux kernel by Linus Torvalds for poor code and failure to address problems, it doesn't help the cause.

There is a strongly held feeling that Red Hat is exerting undue power over other distributions and Linux itself in this battle, essentially becoming a unilateral enforcer of the functions of every Linux distribution. In essence, it's their way or the highway. This is borne out in postings by Poettering and others in mailing lists, blog posts, and bug reports.

To put it simply, in a world that prides itself on ease of adaption, openness, and choice, Poettering has become the poster child for removing choice, ignoring discussion, and limiting compatibility. To many, it seems no technical discussion can be had with Poettering where he is not right, which means there is no discussion to be had at all. It is this state that engenders spiteful and derogatory remarks at his expense.

I've seen some community members speak out on his behalf, decrying the terrible state of open source development discussions and the damage wrought. Well, it isn’t too terrible or damaging if you’re not trying to run roughshod over the very salient and reasonable objections of your peers. If you consistently treat others as fools and inferiors, ignoring their valid issues and points of view, it shouldn’t be surprising when they begin to dislike you.

This is more a commentary on human nature and Internet communications than it is about open source development. In terms of people acting badly during technical discussions, you could say the very same about reactions to this column. Writing is not my full-time vocation, but I do it when I can -- for more than a decade now -- to discuss the vagaries of deep IT across the wide variety of technical disciplines I’m involved in on a day-to-day basis.

However, when I discuss a topic that challenges the status quo (such as excising desktop dependencies from a mainstream Linux distribution), I get all kinds of responses that are heavy on the antagonism and light to non-existent on technical details. These are people that clearly do not understand the discussion, but think they do and thus are quick to attack. I’ve received plenty of emails that simply state “F*** you,” then get nasty. It happens. That’s all from a few columns, a few thoughts not forced upon the Linux world in general.

The fact is that Poettering has a habit of ignoring criticism and real problems, dismissing them and those who report them as user error, or overlooking the issues altogether. Scores of other open source projects do not suffer from this, and they'e run by people or teams that can handle criticism and technical discussions, without consistently disparaging or snubbing their users and contributors. That's generally been my experience with open source development, at least since this field started a few decades ago.

So forgive me if I am less than sympathetic to Poettering’s depiction of open source development as a hellhole of evil where tasks can be accomplished only through abuse and high tempers. It's not Sunnybrook Farm for sure, but that is not an accurate representation for most of us.

You reap what you sow, Lennart, and you may want to begin sowing seeds of cooperation and understanding. Your current methods are clearly not doing you any favors.

From CIO: 8 Free Online Courses to Grow Your Tech Skills
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.