On holy wars, and a plea for peace
lkml.org
lkml.org
That is exactly what the people pushing the new CoC are claiming.
Far more is being made of this situation than seems merited by the facts. That it's easy to do I recognize, but that ease is no excuse.
This would present a problem, except the Linux governance model doesn't actually confer power to a person or persons to exclude contributors from the kernel. Linus and his underling maintainers are still individually free to accept or reject patches from anyone. CoC complaints would be handled by the Tech Advisory Board, but they can't issue orders or blacklists to maintainers.
Funny thing is, I suspect that this is by design, to make Linux development resistant to covert or open hostile action from the likes of Microsoft.
Pending a drastic change in the process, you can all keep calm and carry on using Linux.
I'm opposed to racism, sexism, homophobia, and transphobia too, but damnit, if I'm a maintainer and Alex Jones has a good patch, I'm going to clip out the attached verbiage about Hillary Clinton being a literal demon and roll the patch into my next rc. I'd be doing the project a disservice not to.
The OED has entries that were submitted by an insane murderer from the asylum (look up William Chester Minor).
eta: As it happens, I'd take Alex Jones's patch, too. But I would require it be resubmitted anonymously, and without inflammatory verbiage. If not so resubmitted, I would not merge it. Yes, it does the project a disservice to reject good code. But software, like every made thing in the world, is made by people. It is not unreasonable for me to look at the sort of people who attend upon Alex Jones, and then at the sort of people who find his presence so distasteful that they will not associate themselves with anything he's touched however fleetingly, and decide which sort of people I prefer to include, by my actions, in my pool of potential contributors. If the latter group seems to me to be more likely to contribute good code, then I'm not going to put them off by having Alex Jones on my contributor list. If that costs the project a good patch, then that cost is still less than the other - if I've evaluated correctly, and if the patch has merit, someone not so divisive will submit another like it before very long.
Of course, most people don't give a damn either way, and mainly just want to do the work they're doing and land their changes with a minimum of fuss. Which is more or less the point that I'm trying to make here: this is neither an attempt to install a blood estrogen titer in the code review process for Linux kernel contributions, nor to ensure that Linux kernel development is a clubhouse admitting only pale penis-bearing people of power. Everyone attempting to make a substantive contribution, of whatever sort, to the process, is doing so in good faith, out of a genuine belief that their contribution will improve the quality of the result. But I appreciate that's hard to keep sight of, when Twitter and ESR and /g/ join in, in their inimitable fashion to add only heat and no light.
These people are not acting in love. They are acting in hate trying to protect those they love. It is shameful. It is how wars start.
"The CoC is a political document" is a reasonable statement from someone who considers that everything is political. I disagree, and I think a lot of risks attend upon such consideration which Ms. Ehmke and the movement of which she's a part too often fail to recognize. But, disagree though I may with their rationale, at least they have a rationale. Their opposition seems primarily to rely on viciousness.
Forced to weigh people who I think are wrong about some things on the one side, and on the other people who plumb the depths of hatred and vileness without the slightest apparent scruple, I hope I may be forgiven for deciding that my sympathies belong with the former.
The hell of it is, there are worthwhile counterpoints to be made, and people trying to make them. But there's no hearing them through all the garbage, and until we figure out how that changes, I don't see any real prospect of a stable modus vivendi.
If you were conspiracy-minded about it, it almost seems like the SHITCOCK posters are agents for the other side, in order to discredit the opposition. (It's happened. Look up Meg Lanker-Simons, a standard issue SJW stupid enough to get caught sexually harassing herself.)
[0] https://twitter.com/coralineada/status/1029161113848557569?l...
Note that she went on this tirade last month.
Where have I given Ms. Ehmke a free pass? Where has anyone?
So Ms. Ehmke's only regret in Opalgate is that she tipped her hand too much too soon, and wasn't vague and concern-trolly enough to convince the Opal project maintainers to evict the allegedly transphobic contributor seemingly of their own accord.
Part of the problem people have with the Contributor Covenant is that while there isn't any anti-meritocracy language in the document itself, the rationale for the document cites "dogmatic insistence on meritocratic principles of governance" as a problem that must be corrected. It's right there on https://contributor-covenant.org. Previous versions of this document made reference to "the pervasive cult of meritocracy", but again, that was tipping her hand too much so she dialed it back some. Given that meritocracy is the bedrock of open source development principles, chafing at the adoption of a document whose stated purpose is to combat meritocracy is to be expected. Her goal is to make her values open source values, and in the absence of an opportunity to collect scalps, she will look for leverage to twist arms. The strategy is to tone down the language of her CoC and related documents until it's palatable enough to get an in, and then put the screws to the community once they've pledged to uphold her code.
Ever play Pandemic? A winning strategy is to start with benign symptoms like sniffles and a sneeze or cough, and then, only once everybody is sick (including those paranoid holdouts in Madagascar), crank up the nausea, vomiting, internal bleeding and heart failure sliders.
As it happens, the critique of meritocracy is one of the claims I don’t particularly buy. One of many; sympathetic as I am to most progressive goals, I think the movement’s choice of methods is almost maximally foolish. But, leaving aside the question of good faith on which I doubt we will agree, I think you ascribe a lot more power to the pro-CoC faction than that faction is in truth able to command.
In my original post, I said the CoC would be a problem, but the structure of Linux governance appears to be resilient to bad-faith complaints for the time being -- again, likely to prevent entryism from the likes of Microsoft in the 90s, when the threat of infiltration of the process by saboteurs was arguably more real.
Given the author's toxic Twitter comments, I'm suspicious of the motivations here.
This pretty much rules out "good faith".
I consider myself a leftist (although somewhat disenchanted these days, but that's another story). I find it somewhat unjust and illogical that in our world of plenty people still are in want. For example, the USA produces more food than needed to feed everyone, yet people still starve and few have access to healthy affordable food. It's one argument if food really was in 2018 a scarce resource, but since it no longer needs to be, depriving it from people seems unjust to me. Same for clean drinking water, clean air, etc.
So I have that stance when it comes to material resources, but open source isn't really like that. The rewards are "reputation" which isn't a necessary need for life like food or water or healthcare is, so I don't see why a redistribution of reputation is necessary from that left perspective. I suppose contributions to F/OSS places nice footnotes on one's resume, but if that's the reason then this idea of post-meritocracy is attacking the problem from the wrong end, which involves the actual material realities of access to education and basic needs suffered by certain people. And, sorry, let's keep it real, if you have enough clout and are placed in such a situation that you can worry about open source in the first place, you are materially better off than the vast majority of people in the US, so I don't see it as that high on the list in terms of priorities social justice honestly.
So I don't get what this is for other than it being easier than actually materially improving the lives of disenfranchised people.
Political points for its author. Because factually, it is crap.
Where is "the exclusion of underrepresented people in technology"? Providing high-quality software for free, that is inclusive for poor people (both the poor people in rich countries, and the average people in poor countries). A lot of free software supports localization anyone can contribute to, which is inclusive for speakers of minority languages (instead of having to write your own software, or petitioning a huge corporation, you just download a text file and translate a few lines). Free software can be used by everyone, black or white, male or female, heterosexual or homosexual, cisgender or transgender, whatever; all code is available for all people.
Somehow this all means nothing. What we need instead is to make sure there are not too many white cishet guys contributing to all this stuff. Now people whose only contribution is bringing conflict are kicking out those who contributed code, in their free time, for the benefit of everyone.
Further, there's the argument that aslong as something doesn't hinder the "telos", and it has a positive impact on the community, why not do so?
"...[Accusing anti-CoC community members of sexism and the like] can only inflame their sense that the group telos is being hijacked. They make it clear; they signed on to participate in a meritocracy with reputation rewards, and they think that is being taken [away] from them."
Is the telos to produce a kernel, or is the telos to "participate in a meritocracy with reputation rewards"? ESR's piece gives much overt consideration to the question of "what is the telos?", but even granting that that question is not itself disingenuous in the context of LKML, the quoted passage somewhat deftly sneaks in an answer of its own, between all the many waves of hands about how "[that] last paragraph may sound like I have strayed from neutrality into making a value claim, but not really" and the like.
I suspect a large number of people with commits in the kernel trunk would be surprised to learn that they're really there for the meritocracy, rather than for the work...
a) You’re there to work on the kernel, and because that’s the end goal you out up with the CoC;
b) You’re there for the reputation and/or meritocracy, so the CoC feels like a personal attack.
Is that it?
Be clear, though: this is nothing but my reading of ESR's statement of his own perspective on the issue, even all the facts of which are somewhat difficult to identify without ambiguity. I'm not especially interested in trying to make any larger argument than that, and I'm not at all sure that the desire for meritocracy and reputation is incompatible with the adoption of an explicit code of conduct requiring that kernel contributors, while contributing to the kernel, eschew gratuitous misbehavior of the sort which tends to drive other contributors and potential contributors away from the project.
As I read the CoC in question (https://www.kernel.org/doc/html/latest/process/code-of-condu...), I see nothing in it which suggests any intent beyond that. In particular, I see nothing suggesting any intent toward thought policing, or policing of behavior outside the context of interactions in the Linux kernel development community, or declared public representation of same. And the Technical Advisory Board, charged with enforcement of the CoC, is just that - technical. So I really don't know where all this culture war stuff is coming from, or why anyone sees a need for it.
Okay, that's a lie. I do know where it's coming from, and why people on both sides see a need for it. It's just that I wish they didn't - again, on both sides - because the Linux kernel is too important a technological artifact to be made this sort of battlefield. I'm sure it'll blow over in the end, and that the silent majority that invariably constitutes the audience for such engagements will go on doing as they do, not at all ruffled by the explicit adoption of a standard of behavior that really says nothing more than "don't be an asshole, and if you're enough of an asshole, we won't let you work with us any more".
That is to say - the same rule that applies everywhere, whether explicitly stated or otherwise. In my experience, the correlation between worthwhile kernel contributors, and people who enshrine Linus Torvalds' own somewhat infamous former style of comportment as somehow essential to technical excellence in the project, is negligible unto imperceptibility - so I see no reason to imagine that the project itself is imperiled by this current contretemps. That contention is really the only reason why it's achieved as wide a currency as it has, I think; absent the - in my reading highly questionable - idea that a significant fraction of kernel contributors will withdraw their contributions in a way that actively harms further development, there's no reason I can see for so many pundits of ESR's and /g/'s stripe to weigh in so heavily and with such typical respect for the facts and realities of the matters on which they choose so to speak.
This is a bit like asking "is your goal to bake bread, or to use flour?", and when the baker says "to bake bread, obviously", replying "so you don't mind if we just throw away all your flour, right?". I would side with the baker who says that people who go around throwing away the flour are probably doing it to sabotage baking bread, not because they want to improve the quality of bread.
The Linux Documentation Project existed long before these people came. And if "dealing effectively with people" is what we currently observe, I'd say no thanks.
I think this is the root of the issue - the previous state (not having a CoC) allowed contributors to assign their own telos, based on their perception. Creating a CoC creates a framework where none existed, and some people don't agree with that framework.
The CoC - any CoC - requires that all members agree on a telos. That wasn't necessary before.
Does it though? All it says to me, that whilst working towards this "telos", do it in a respectful manner.
Instead, all I'm trying to say is that while you see it in that light, others don't. Those others must make a decision - accept the change, quietly leave the project, or use the legal means at their disposal to disassociate themselves and damage the project in the process. That decision wasn't required of them before the CoC was created.
It does more than that. It allows for a path forward/establishes a slippery slope. Now a group that is most earnestly behind the implementation of the CoC to now point to this as a tipping point where the tyranny of the majority is accepted (and can now make modifications since you already agreed to a CoC, what are some nomic modifications to you?). This runs counter to the existing, unstable hierarchy.
Why don't we all just be people, accept that our best foot is not always the forward one, and accept positive contribution wherever it can be found.
It was actually pretty mild, and did forbid harassment, but its purpose was to highlight how producing the best possible kernel was the ethos of the community.
Some will claim of course that it didn't do enough to highlight what was considered harassment, or that it didn't really have teeth because it didn't specify any consequences for bad behavior.
(The TAB members, for their part, are trying to allay the concerns of critics of the new CoC by essentially claiming that it still has no teeth, because the TAB cannot prevent maintainers from accepting contributions from people who allegedly violate the CoC).
The new CoC has a specifically different primary purpose:
> In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone
In other words, the goal has shifted from producing the best possible kernel to creating a welcoming environment for everyone. Now, even if you believe both are good goals, and even if you believe that these goals can be pursued in a synergistic manner, there's still a question of which is more fundamental to the ethos of the community.
[0]https://www.kernel.org/doc/html/v4.10/process/code-of-confli...