I have observed a concerning pattern: many package maintainers exhibit a very negative attitude toward user feedback, especially on RHBZ. Numerous maintainers never respond to reports submitted there. As a user, I don’t know what I’m supposed to do. I’m not even sure whether checking RHBZ is part of a maintainer’s obligations. All I’m left with is a sense of helplessness—seeing issues, sometimes even identifying solutions, yet having nowhere meaningful to send them.
I’ve tried submitting merge requests, but I’m not a professional developer, and often my changes don’t meet the expected standards. As a result, submitting a merge request ends up replacing the act of reporting an issue on RHBZ. I don’t believe this is the intended workflow.
Here is a specific example: I sent a DM invite to a package maintainer asking for clarification on a commit for a package, and they rejected the invitation.
Feel free to point out if I’ve done something wrong, because I genuinely don’t know who else to ask about such a minor question except the author of the commit directly.
@fxzxmicah, they always receive e-mail on bugs filed at RHBZ. If you believe that your reports are being ignored, I believe that a process exists to notify a higher authority that the maintainer is inactive, but am not familiar with the process. Misusing the MR feature merely causes more mail for the maintainer to sift through.
I agree that there is a common lack of response, however. I find that filing an issue upstream, where possible, citing an RHBZ bug, assists.
The issue isn’t whether the maintainer is active — I can see they’re still making commits. They simply don’t respond on RHBZ, which doesn’t mean they’re inactive. That said, there are indeed cases where the primary maintainer is no longer active, and other contributors are helping maintain the package.
If the issue originates upstream, I report it upstream. When I report an issue to Fedora, it means the issue is specific to Fedora and not an upstream issue.
In RH bugzilla you see response time from mere days to up to 10 years, so it varies a lot. I don’t think it is negative attitude at all, but more like no time or simply interests has changed.
There is also the cases where the half-yearly rebuild everything results in failed builds. If no maintainer takes care of those, you can see packages being dropped from the repositories. For example this happened to grub-customizer.
As a user, the best thing is to explain how to reproduce the issue, and then it is up to the maintainer to try to reproduce it.
Also, I’m not just speaking for myself — many other people’s bug reports receive no response either. Sometimes, bug reports remain untouched for years. In some cases, you can see that the maintainer was still responding on RHBZ a few months (or years) ago, but then suddenly stopped, even though they continue actively submitting commits.
Most rhbz reports are abrt generated useless spam.
Users need to include the required steps to reproduce the issue, if they can’t they shouldn’t waste the maintainers time!
The implicit condition of this post is: the maintainer has not responded to any issues on RHBZ — including closing invalid issues or acknowledging valid ones. Just like a package with no one maintaining it.
Also, by your logic, should we abandon RHBZ altogether? After all, most of it is useless spam anyway.
Are abrt-generated reports tagged as such? If not, that would surely go a long way to filtering them from all else. They are at least prefixed with [abrt], so they should be fairly trivial to ignore if they’re considered to be so.
@fxzxmicah, which tracker you utilise isn’t per se relevant to this, insofar as you provide enough information for the package maintainer to understand what package you’re utilising, and have reason to believe that they support downstream packages. KDE’s is an example: at bugs.kde.org, the user is able to choose which distribution’s packages are affected.
Many — and personally I would even say most — software developers simply don’t have the capacity to handle issues that are limited to a specific downstream. Some issues aren’t even problems in the strict sense, but enhancement requests, such as splitting test components from the main package.
Of course, I understand that you’re offering a possible path to resolving issues more quickly. But I still believe that problems specific to Fedora should be resolved within Fedora itself.
Just like I don’t believe using merge requests as a substitute for RHBZ is a reasonable workflow, but the reality is that maintainers often ignore RHBZ while still responding to MRs. (Of course, there are cases where they don’t respond to MRs either—usually when the primary maintainer is no longer active and the package is being maintained by other contributors.)
Not responding is often a consequence of the employer’s priorities , so it is irresponsible to assume a “negative attitude” on the part of the maintainer.
Linux maintainers are a scarce resource. Some are approaching retirement age. Some work in a large organization that has been reducing staff and reorganizing, so job responsibilities and priorities are in flux.
You’ve misunderstood — when I said “negative,” I didn’t mean it in a derogatory sense. It was meant in the same way as terms like “lazy loading,” describing an objective state. It’s simply the opposite of an “active” response. “Active” isn’t a positive judgment either; both terms are factual descriptions.
After all, it’s also possible that a bug in the RHBZ mail system caused them not to receive the message, isn’t it? I wouldn’t casually engage in any kind of “attack” on anyone. I simply hope the community can take note of this issue — either explain why it happened, or work to improve the situation as much as possible.
For me, the opposite of active/timely response is lack of (a timely) response. This is most probably because people work on other things (related to Fedora or not).
Negative attitude implies that maintainers would actively respond negatively (or with animosity) to user feedback. This is something completely different and my understanding is that it doesn’t seem to be the case. So I was also very confused by the current wording and initially completely misunderstood what issue is discussed in this topic.
In my view, the only possible ambiguity in my statement lies in whether it carries a derogatory tone. I have already explained this, so that aspect ends here.
As for your statement that “Negative attitude implies that maintainers would actively respond negatively (or with animosity) to user feedback,” you are not entirely mistaken. No response is also a form of response; no response is, in fact, an active way of responding negatively.
However, this has almost deviated from my original topic and turned into a post correcting my phrasing. I would like this aspect to end here as well.
Packages that are maintained by Gnome folks remain quite close to upstream, and from what I know the maintainers prefer bugs to be opened directly upstream instead of on Fedora’s bugzilla. See this discussion, for example:
For the particular case you’ve given as example, in general, we avoid private messages in the community. We “default to open”, so queries of this type would, for example, be asked in the public channels—the devel list or matrix channels.
I wouldn’t perhaps accept private DM invitations from people I don’t know either—there’s also been a lot of (quite appalling) spam on the channels where individuals/bot accounts invite us to DMs and then send us really bad images.
I’ve removed the screenshot because it pointed to an individual community member. We don’t do that.
(If there’s a case where a CoC violation has occurred (which is not the case here), then we follow a process where the various individuals involved are required to be named if necessary)
I wasn’t specifically referring to maintainers of GNOME-related packages.
If that’s the case, then why does he still leave personal contact information in his profile? To me, leaving contact info gives the impression of: “I’m open to receiving private messages.”
One more important thing—next time I have a small question like this, where and to whom should I direct it?
I can’t speak of other package maintainers since it depends on individuals. The “policy” i pointed to seems to be limited to Gnome.
Ah, no, please do not make that assumption unless some one has explicitly said so. One can also find me and everyone that is logged in to Matrix with their Fedora account—it uses our Fedora usernames. That does not mean that we must accept and respond to all direct messages. We may accept some, we may not accept others. There can be many reasons for either decision, but we do have the right to choose.
Anyway, in general, please “default to open”. That’s the community way.
I would use the -devel channels where all the package maintainers hang out. Whoever can help, will.