Jeremy Stanley <fu...@yuggoth.org> writes: > Removing a package from the archive purely on the grounds that it > contains objectionable text, if such is the reason for not distributing > it, is making a value judgement of that text. The concerns that have > been raised so far for objecting to the content of the package in > question are applicable to quite a number of other packages in Debian as > well. Hyperbolic perhaps, but it doesn't seem that far separated as > analogies go.
I think the analogy adds a lot of heat, but I think this is getting to the heart of the problem with this recurring discussion. I think one can make an argument that Debian is not intended to be a collection of random data that someone finds interesting. There are a lot of things in the world that are available under free-software-compatible licenses and that we could theoretically package, such as the entire contents of the Gutenberg Project. We by and large don't package such things, in part because that feels a bit afield of the purpose of the project. We could potentially make a firmer decision that we only package software and not large data files, and that would be a reasonable and relatively straightforward decision we could build criteria around. That would imply that we wouldn't package large fortune databases, the Bible, doc-rfc, and so forth. It would probably also affect a lot of other things that I haven't thought about. I'm dubious there's a consensus to make that sort of change, but I think that would at least be a clear boundary to draw around the purpose of Debian that would be consistent with some of the other comments on this thread. But if we're not going to do that, I think these decisions get a lot more subjective. Currently, what we're roughly doing in practice is that people package whatever they feel would be useful in a package, with some check by the archive team that this has some plausible connection to Debian and doesn't consume excessive Debian resources. (For instance, I suspect they wouldn't be willing to accept the complete Gutenberg Project just because it's quite large.) Things that have a more direct tie to software that we're also packaging are going to be more readily accepted, something that applies to both fortune databases and to the text of the Bible, since in both cases they are packaged alongside software in Debian designed to consume and manipulate that data format. We could adopt an additional subjective decision process around how much we like the content, but, well, this feels to me like a recipe for constant contentious project arguments of exactly the type we're both not great at handling and that burn huge amounts of emotional energy for everyone involved. I think it's possible to draw a very clear distinction between the contents of packages and our interactions with each other (the subject matter the project Code of Conduct is designed for). The primary distinction is *context*: communications with each other are directed at specific people and happen within the context of an ongoing working relationship. We're trying to accomplish something together, that requires communicating with each other, and that communication should therefore support that joint effort and reflect the ideals of the *community* that we're trying to build to work on the project. This is generally not the context of Debian packages. I could think of cases where it was, and where the Code of Conduct would apply, but that would involve packages that are being uploaded to specifically target other people in the project, which thankfully isn't the case here. Or it would involve using the package to communicate directly to users, such as enabling an offensive fortunes database by default to show quotes during the Debian installation process, something we obviously wouldn't do. Absent that sort of intent, packages are very much like books in a library. The person checking out the book brings their own context; the presence of the book does not impose a context in the way that direct communication between project members do. For example, I would find it very off-putting for someone to prosletyze at me within a Debian collaboration context. If I asked them to stop, I'd expect them to stop. But the presence of the text of the Bible in the archive doesn't do that; it's a lot like a book in a library. And even if one disagrees quite strongly with things that it says, it still serves a separate purpose in a separate context: as a reference, as a research tool, etc. The fortunes database is kind of a trivial instance of this. The stakes are pretty low, which is probably part of why the argument is so heated (to borrow the old trope about academic debates). It's essentially only a proxy fight over larger principles. But I think the larger principles weigh against embracing some sort of project-wide content-based decision on package vetting, and certainly against applying the Code of Conduct to something that does not have at all the same context as what the Code of Conduct was designed to address. If we were going to write a project content policy (which I'm dubious we really need to do, or that it would be worth the emotional effort required), I think it would look much different than the Code of Conduct because it would have different goals. It wouldn't be about building a community or encouraging productive collaboration, because the contents of our archive don't need to do either of those things. Lots of people use Debian who are not members of any shared community, and this is a feature. -- Russ Allbery (r...@debian.org) <https://www.eyrie.org/~eagle/>