4 ms·
>If I want to contribute to an open source software project I need to share the politics of the people that wrote it I disagree. Software is meant to be merito
by krapp 10y ago
>If I want to contribute to an open source software project I need to share the politics of the people that wrote it
I disagree. Software is meant to be meritocratic - all that should matter is the quality of your commits, if you're a contributor. It shouldn't be necessary or even desirable to have an explicit code of conduct to expect civility.
> There's a daily printed and online publication in the UK called the Daily Mail. It's a nasty, divisive, dishonest paper and it's done a lot of damage to our country. But they published a software library that I wanted to use. I was genuinely conflicted.
>The environment under which the library was written supports division, inequality, victimisation of minorities
Would using the library support division, inequality or victimize minorities?
Would not using the library help minorities or promote unity and equality?
Fritz Haber developed a process[0] for synthesizing ammonia as a means of manufacturing poison gas for use in World War 1 - a process which led directly to the development of the gas chambers used during the Holocaust. However, the Haber process also revolutionized agriculture worldwide, making it possible to grow food in places where it would otherwise have been impossible. Does that mean modern agriculture is imbued with violent German nationalism or anti-Semitism? Of course not.
In the same way, code has no political, social or moral dimension of its own, and the political environment in which it's written need have no relevance at all to how its used, or by whom.
[0]https://en.wikipedia.org/wiki/Haber_process https://en.wikipedia.org/wiki/Haber_process
- wolfgke 10y ago> I disagree. Software is meant to be meritocratic - all that should matter is the quality of your commits, if you're a contributor. How do you come to the idea that software is meant to be meritocratic? Many people start an OSS software project so that the code implements the founder's vision. So for example I committed some (objectively; in this case "objectively" can objectively be justified [pun intended]) high-quality patches to some OSS project (which project is of no concern here). They were refused because they diverted from the vision that the founder and maintainer of this OSS project has for it.
- krapp 10y agoIf that "vision" relates only to the design or structure of the software, then I believe that meritocracy means conforming to the project owner's standards. But if it's about politics or social issues taking precedence over the code, then I would be opposed to it. There are plenty of angry, uncivil, misogynistic and racist 'thought leaders' in tech who are respected not for their political views (although sometimes, yes) but for the quality and popularity of their projects, so I think the premise of software being meritocratic is fairly well established.
- wolfgke 10y ago> If that "vision" relates only to the design or structure of the software, then I believe that meritocracy means conforming to the project owner's standards. At least in this case "meritocracy" does not mean "the opinion (about code structure etc.) that is (necessarily) shared by the majority of the contributors".
- krapp 10y agoWell, yeah, many if not all open source projects are dictatorships by necessity - not everyone's opinion can matter when only one or a few people are actually able to accept pull requests. Even the ones that accept the will of the majority are doing so by an arbitrary fiat. But if the majority of contributors disagree with the owner about fundamental design issues, that's what forks are for.
- afandian 10y agoSoftware is just ones and zeroes. The only thing it's "meant" to do is execute on a computer. If you mean "collaborative open source software projects", the idea of "meritocracy" is an explicitly political idea; it's not innate. Open Source is also an explicitly political idea. Competing political ideas in software development include 'not invented here syndrome' which is based on the tribal affiliation of the contributor. If I don't share the political principles of open source (e.g. copying closed code in contravention of the license), I'm going to have a hard time contributing to an open source project. If I don't share the political ideas of a commercial closed-source project (e.g. copying open code in contravention of the license) then I'm going to have a hard time contributing to a closed commercial project. And I wasn't making an assertion about using libraries published by the Daily Mail so much as posing a question. I can't fully quantify the effect of any action I make, everything is interrelated. I'm perfectly happy to admit that it was a matter of personal pride that I didn't want to be associated with the Daily Mail in any way because I don't like what they stand for. Identity politics. I'm only human. I wouldn't (I hope) force anyone else not to use the Daily Mail library. And I wouldn't try to force someone out of a software project because they had different politics. Maybe that broadens out to the original topic: the reason some people don't seem to like CoCs (and I'm undecided) is because they can be used to enforce tribalism. We've seen examples of people being attacked in software projects because they have contrary political views to some other people. One of the arguments here seems to be whether or not that is driven by tribal instinct, and whether CoCs provide a vector for that.