Hacker News new | past | comments | ask | show | jobs | submit login
[dupe] Resigning as a Debian systemd maintainer (err.no)
127 points by martinml on Nov 17, 2014 | hide | past | favorite | 28 comments



This part resonates well for me:

"Now, how did I, standing stout and tall, get forced out of my tribe? I've been a DD for almost 14 years, I should be able to weather any storm, shouldn't I? It turns out that no, the mountain does get worn down by the rain. It's not a single hurtful comment here and there. There's a constant drum about this all being some sort of conspiracy and there are sometimes flares where people wish people involved in systemd would be run over by a bus or just accusations of incompetence.

Our code of conduct says, "assume good faith". If you ever find yourself not doing that, step back, breathe. See if there's a reasonable explanation for why somebody is saying something or behaving in a way that doesn't make sense to you."

It's easy to say to ignore the comments and the haters, but if you keep shit thrown at you by people who automatically assume the worst possible for everything, even an elephant skin gets pierced at some point.


This is part of a larger cultural problem in the programming / sysops community. Too many people assume the worst, and way way too quickly. The Redis author's "This is why I can’t have conversations using Twitter" (http://antirez.com/news/82) article describes this pretty clearly.

Even if you can write an awesome system, or an awesome article, with one tiny flaw, immediately an army of people will jump on that single tiny flaw and declare you stupid and incompetent, or worse, that you should be shot.

This kind of attitude does not bode well for the open source community, where cooperation and collaboration is important.


At this point in time i fear the *gate trolls have latched onto this issue and if using it for their "lulz".


I enjoyed the post for teaching me what teamwork really is: assume good faith, ok to disagree, do not escalate even when the stakes are high.



They have seen a couple of departures lately. Joey Hess[1] blamed the way decisions are made within the project[2].

[1] https://joeyh.name/blog/entry/on_leaving/

[2] https://lists.debian.org/debian-devel/2014/11/msg00174.html


Can you elaborate why he considers the Debian constitution a toxic document?


Because the guiding committee can decide to call a vote and stop all discussion about the subject. In this case, he thought that 3 days of debate over a decision was not enough, but that's what the committee called for. Joey had some ideas he wanted to present before the vote, but the way the committee acted prevented him from presenting his opinion.


I believe it's about referring technical decisions in subgroups to project-wide political votes (general resolutions = GR).

I can see why trying to keep everything as a pure technical meritocracy is appealing, but I don't think it's realistic. Libre software is inherently political.


And yet, its only happened basically once in 20 years, due to a project wide coup lead by the desktop environment people basically taking over the distro.

I don't want to leave the mistaken impression that every time someone does a git commit to debian-policy or lintian that a GR is auto-proposed. Virtually all historical changes to Policy or TC "cabal" decisions have been extremely calm and compared to the recent coup. Debian minus systemd is pretty calm and well behaved. If anything, cracks are forming because its such an earthquake compared to all previous debate and discussion.


a project wide coup lead by the desktop environment people basically taking over the distro

Maybe you should read the OP, for example the "assume good faith" parts. Really.


As a strategy for keeping systemd out of Debian, harassing the maintainers of systemd until they quit the project is extreme, immoral, and effective.

(This is not an endorsement of the tactic!)


I'm not sure it's all that effective. Yes, they've managed to harass this guy into leaving the systemd team for Debian. But I don't think that's enough to keep systemd out of Debian. And the way things are going, nothing will.

Debian is an operating system, but the amount of code that comes from people actually part of the Debian project is very, very small in comparison to the amount of code that comes from upstream projects. Some of those projects have people on them who are involved in Debian, many do not. In a growing number of cases, those upstreams are starting to rely on functionality that systemd provides, partly because it's becoming common to have systemd available on Linux systems, partly because it's providing functionality that nobody else is (like logind replacing the aged and unmaintained ConsoleKit[1]).

So if upstream packages are going to rely upon systemd, what is Debian to do? They have a limited set of options. They are:

1) Ship the packages the way the upstream built them -- dependent on logind, for instance. Ship any systemd compatability shims that exist for people who want to make a go of it without systemd. 2) Stop shipping packages that require systemd, or rewrite parts of thost packages until they don't anymore.

The problem is that a lot of volunteers who maintain packages for Debian are convinced that number one is the best -- it's the least effort for them, and provides the best user experience for anyone on systemd. People who want to keep systemd out of Debian are pushing for number two, and they're trying to do it politically -- in other words, force maintainers to behave how they want to. In a volunteer project, that's rather dicey.

But there's really no avoiding it -- if the TC had decided against systemd as the default init, GNOME would still depend on logind, and there's still be a fight over who had to do the work to get GNOME running without systemd.

(This is also why I'm really annoyed at people who keep screaming that Linux is about choice -- yes, but it's not about coerced choice. Your choice not to use systemd doesn't mean you get to force others to work on code to support your choice.)

[1] Yes, someone forked ConsoleKit into ConsoleKit2. That happened about a month ago, way past the point where all this debate started, and probably past the point where most packages are going to care.


I don't know why you got downvoted, but I fear it may be because the people doing that don't consider it "immoral" at all, because they are doing God's work, so ejecting the heathens is by definition right.

Edit: Found this in the other thread: https://news.ycombinator.com/item?id=8616490


I have no desire to be a package maintainer. But if it goes as far that systemd support in Debian gets threatened, I'd step up, as the alternative would be to switch distributions. I'm sure others would feel the same.

It'll not be effective.

If anything, I suspect the long term effect will be to make it harder for those who don't like systemd to get heard.


Thanks for your work on Debian and systemd! Sorry to hear that you had to endure flame wars.


Sorry for the offtopic, but that's a great domain :) first "vanity" domain that made me smile in a long time.


And then you have ANOTHER high profile departure like Russ Allbery from the Debian Technical Committee:

https://lists.debian.org/debian-ctte/2014/11/msg00071.html

All of this (as a non-Debian but Fedora user) tells me that Debian is going through a really rough time. For the future of Debian and free software in general, I do hope they figure out how to remove all of the stop energy and get back to making a world class distribution. Just like Redhat, there are so many projects maintained by actual Debian Developers and Debian has written an epic amount of man pages, just to name a few small contributions from Debian as a distribution. It would be a very sad day if Debian devolved into another Gentoo when Daniel Robbins left and it all went to crud, birthing (more or less) Arch Linux in it's place.



Very sad to see


Crap, and Russ Allbery resigns from the technical committee too:

https://lists.debian.org/debian-project/2014/11/msg00054.htm...


"I think the agile philosophy got a few things right: find ways to reduce the cost of change, empower individuals to make choices and act on them, and reduce the cost of failure and embrace iteration instead of trying to prevent failure in advance."

This reads as a stealth attack against systemd, as that's a pretty good summary of the opposite of the systemd development philosophy as currently implemented. Although he claims in his post that line is about the Debian decision making process, which is perhaps true although very meta. Everything systemd is about is implementing the exact opposite of the quoted line above.

My one line summary of the overall situation is the systemd project is a coup of the overall ecosystem by the "desktop environment" people ignoring the needs of all other users, and the "universal OS" true believers are not overly supportive of that coup.

I'm reacting to the replacement of "universal OS" with "we're only going to be a gnome desktop" by simply moving everything, desktops and servers, to freebsd, other people are flaming, others are making threats, and of course others are moving to openbsd, whatever.


FYI, you're quoting an email from Russ Allbery, who voted for systemd: https://lists.debian.org/debian-ctte/2014/02/msg00283.html

To quote Russ vote: "D U O V F", so systemd first and the current System V init last ("F" was for further discussion).

In the same email from Russ you're quoting, there's this: "I would appreciate it if people would ask for clarification rather than making assumptions, as assumptions about other people's motives are one of the things that I find the most demoralizing about the Debian project right now."

Worth a though IMHO.


I don't understand what do you intended with this post besides adding fuel to the fire.


> https://lists.debian.org/debian-devel/2014/11/msg00133.html

This is a good letter about managing any kind of community dealing with a disagreement. It's also good to get this out in front before any decisions are made.


On the subject of bad attitudes in social groups with long-term continuity of any kind, be it weirdos at a hackerspace or misogyny in STEM or death threats over Twitter:

I've found that it's so hard to manage and avoid these sorts of issues because the roots of it are laid long before it ever becomes vocalized as a problem. It usually starts with one person who misbehaves in some way and everyone else ignores/allows it because they don't want to be the bad guy who starts the conflict. But it grows from there--the original bad actor grows more bold and/or attracting other bad actors, and previously good actors start to become resentful, start to demonstrate their own bad behavior, especially now that they recognize the environment as tolerating it.

By the time someone says anything about it, passive-aggression has become a large part of the group. Pruning such a large (though not necessarily majority) part of the group could leave the group without the day-to-day support it needs to remain viable.

Note that I'm not calling this out as a failure of any particular group involved in this particular case. The vast majority of organizations either don't have people who are dedicated enough to stick around, year after year, or fall into this habit because they feel (most likely correctly) they lack the authority to do anything about it. Most organizations are not setup from the start to expect bad behavior, and therefor don't have the mechanisms in place to take care of it.

I don't know how to clean up an already existing problem. I suspect that it either takes admitting personal defeat and moving on to something new. There might be something in going after the minor offenders, first, to undermine the support that the long-established people enjoy in these sorts of conflicts, but I doubt such a strategy is not so transparent to throw up red flags.

If you're starting a new organization, I plead to you: always be friendly, always be fair, but always be firm. The first time something even remotely inappropriate happens, you have to say something directly to the offender. I can't even count how many times I've had to chide people for using the N-word (let alone other words that should be as obvious by now). Do it in private, there is no need to embarrass them. Public embarrassment should be reserved for 2nd-time offenders, because if they think they can get away with it a second time, it means they think they are in an environment that will support that behavior, so there are probably others who think the same way and just haven't acted yet.

You'll probably piss someone off in the process, "how dare they censor me?" But you don't need to please everyone. You need only the good people who can play nicely with others.

"Oh, I'm doing it ironically" is not an excuse, even if (in the exceedingly remote case) it's true. Because context is important. You never know when someone new is coming along and doesn't understand the person is "just joking". And there are two ways the new person can react, neither which are good for your organization. In the best case, they will be put off and not return. In the worst case, they will think they've found a sympathetic environment in which they can peddle their own bad behavior.

In the process, you absolutely must not stoop to their level. You must always portray an outward appearance of calmness, even friendliness, regardless of how angry you might be. It has to be clear that the issue is the behavior and not a personal conflict.

And you have to be fair. You have to equally reprimand bad behavior, even when you agree with the actor's thesis. There is no other way to make it clear that you mean business.

You're not going to have a problem with the vast majority of people. You're going to encounter one or two really obvious nogoodniks and it's up to you to be the adult, stand up and inform them that the sort of behavior they've demonstrated is not welcome there. If you don't ignore the elephant in the room when he shows up, you can avoid the much more insidious problems further down the road.


“If you're starting a new organization, I plead to you: always be friendly, always be fair, but always be firm. The first time something even remotely inappropriate happens, you have to say something directly to the offender.”

Agree, but I would suggest that even then the elephant in room is a culture lacking diversity.

Look around the culture and see how much diversity is there.

It is much easier to insulate against harrasing and derogatory remarks when those remarks and attitudes are directed at people you respect, admire, and value.


Yes, people say horrible things when they think they are in an environment that will tolerate saying those things. But I'm going to give the management of whatever hypothetical org of which we're speaking the benefit of the doubt and say they don't want to start that way. I hope that, for new organizations who have yet to establish their own culture, that tendency is a projection of expectations for the industry as a whole. My advice is for people who are serious about being fair and equitable in their dealings with diverse groups of people. If we don't have that baseline, I can't help them.

But with that baseline often comes a person who is specifically trying to be nice to others and not hurt other peoples feelings. This is, unfortunately, not going to keep bad actors from getting in and ruining the soup. Thus the need to be firm and resolute. One must establish early, often, and unequivocally that the org is not going to be party to exclusionary behavior.

It's easy to say, "you have to increase diversity", but that's a goal, not a plan. Being a part of the "majority", we're not going to be able to increase diversity by ourselves. That's rather much the whole point of the idea of "diversity", that the homogenous group can't solve all of the problems for all of the people of the world. That rather much includes our own damn problems and the problems of our own damn industry.

For example, take these "<TargetedDemographic>CanCode Campz" things. I'd like to see numbers about them. To me, they've always seemed very forced. I've also heard first-hand complaints from said people in the target demo that the concept is perceived as patronizing. I applaud the organizers for putting in the effort, and their heart is definitely in the right place, but I wonder if they aren't doing more harm than good.

It almost sounds like a chicken-or-egg paradox, that I'm suggesting you can't foster diversity without already having diversity, but that itself is a symptom of the problem. There are very dedicated people out there who are trying to work and trying to be accepted. I've seen first-hand how easy it is for one, single, isolated incident to drive away the very people who would help introduce our organization to a broader audience. We have to stop being blind to the people who are already here.

All that it takes is to not let the vocal, hopefully-minority-of-bad-actors drive good people away. All that it takes is the shocking, controversial policy of just treating "them" not like "them"s.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: