On Thu, May 11, 2023 at 3:23 AM Ralf Corsépius <rc040203@xxxxxxxxxx> wrote: > > > > Am 11.05.23 um 09:09 schrieb Zbigniew Jędrzejewski-Szmek: > > On Thu, May 11, 2023 at 12:22:09AM +0200, Kevin Kofler via devel wrote: > >> Kevin Kofler via devel wrote: > >>> So you folks were not happy with the feedback on the mailing list, so you > >>> just made up a poll somewhere else. > >> [snip] > >>> Was that poll ever announced on the mailing list? Or did you just expect > >>> people to magically notice it has popped up in the FESCo ticket? > >> > >> Sorry, looking at the FESCo ticket, I see the stats are actually just your > >> subjective interpretation of the mailing list replies. Since I do not know > >> who you counted in what category, I cannot tell whether your interpretation > >> is correct or not. > > > > Mattdm put in the work to go over replies and count people's sentiment. > > If you think he did a bad job, you can do your own calculation. I > > doubt you'd get wildly different results. Please post them here. > > I am lacking words to express my sentiments about how MattDM and FESCO > are treating Fedora, to say the least. > As I've said in the FESCo discussion, I actually don't like this idea at all. I have tremendous reservations around forcing development discussions of any kind into Discourse. But at the same time, I also recognize that if we don't actually *try* it once, this will keep coming up. This has been a background discussion from Matthew Miller for at least five years now (and I'm on record for being in opposition to it for just as long)[1]. If we don't actually try it with something, we're not going to see what the outcomes are going to be. To be honest, I think for the vast majority of the non-engineering side of Fedora, mailing lists have been a terrible model. For example, many of the SIGs that report up to the Mindshare Committee tend to communicate with rich media, something that most people here don't use even though email supports it. And the restrictions on email attachments make it incredibly difficult to deal with too. Discourse has been a boon for most of them because the communication mechanism allows them to more fully represent their work as they're developing it. And when it comes to user support, mailing lists are horrible because it's quite difficult to represent the problems people are having well for basically the same reason. I've personally bemoaned the fact that Fedora hasn't had a first-class forum system for users for most of the time I've been here. Pretty much all of our competitors do (Ubuntu, openSUSE, Manjaro, Arch, etc.) and we've always outsourced this to fedoraforum.org (not that they're bad, but it makes no sense for them to be third-party). I deeply dislike the way Discourse works as a forum, but what else is there now? I guess Vanilla, but that's a solution where there's a mismatch in the "cloud" version and the self-hosted open source version (at first glance, they might be completely different codebases now!). At least in theory we should be able to self-host Discourse at a moment's notice, provided we prove we can do that and test it regularly. And porting to another forum software now is rife with peril without significant investment. For the engineering side that reports to FESCo? I think Discourse is a much harder sell. Data durability and portability matters a lot more, as our discussions are often used to inform others of the background of our decision-making. We all know that the Fedora Changes process (and its predecessor, the Fedora Features process) was designed to make it very easy for Red Hat to review how Fedora Linux evolves so when they branch Fedora Linux to make a Red Hat Enterprise Linux release, they can maintain or revert significant Changes as desired. But it's also used quite often by other distributions as a way to identify things worth implementing. As someone who works in multiple Linux distributions, I'm very mindful of how important Fedora is to the larger Linux distribution community and so I want high-fidelity archives of those discussions. For better or worse, email-based communications are the only format with that property. But that doesn't matter if we don't have anyone around to actually *do* anything, and if Matthew is right that we will get higher quality discussions with more engagement on Discourse, it might be worth it. More participants doesn't necessarily mean better conversation, and if we're exchanging one group for another, that might be worse. But we won't know these things until we try it. It was probably a mistake to say "Fedora 40 and onward" will use Discourse for Change discussions, as it assumes success (which I don't want to assume after many experiences in the past proving otherwise). I would have preferred the trial limiting to just Fedora 40 and reviewing it afterward. But we can also just force that review after Fedora 40's development window closes anyway. As it currently stands, this is a *trial* to see how it goes. If we get crickets or low-quality discussion there, we'll put it back and hopefully this won't come up again for another five years. :) But if it works out... Well, we'll see. [1]: https://lwn.net/Articles/768483/ -- 真実はいつも一つ！/ Always, there's only one truth! _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
