AUR guideline on `Duplicate` packages, and Package Maintainers
Recently there's been a debacle about the wechat packaging in AUR. Essentially there were two competing packages wechat and wechat-bin [1] which are both popular. Package maintainer K(type user in AUR) of wechat took unbelievable measures to try to get rid of wechat-bin from AUR to begin with and was rebuked[6] by a PM A(type "Package Maintainer" in AUR). But eventually the same maintainer K succeeded when another PM M(type "Package Maintainer" in AUR) sided with him and took control of wechat-bin with a total revamp of the packaging. Package maintainer K's actions are bizarre, hostile and malicious overall. There's absolutely no reasonable explanation as to why K had to take over the control except for the guideline which is flawed. Plenty of users submitted their negative views[2] towards such takeover. And it's also my observation that PM M did not do due diligence on this matter and just arbitrarily made a decision suppressing users' voice in the meantime. The situation got worse under PM M's dealing with this matter that users feel ignored and betrayed that many **wrong** deletion requests[3] have been filed towards both wechat and wechat-bin, presumably by angry and frustrated users from both camps. The chilling effect can also be seen here[5] when the most upvoted wechat related package is discussing a backup plan in case the same thing happens to it and the possibility of retreating to github, when one PM dictates the course of action and there's no way to appeal. If we also look at the roles PMs play in this debacle, it also draws ire from K. K is not happy when his attempt was denied by PM A either. https://blog.kimiblock.top/2024/12/08/do-not-waste-time-on-aur/ https://blog.kimiblock.top/2024/08/23/aur-moderation/ The questions are: 1. Is there enough room for more than one packaging in AUR or does it have to favor the first package maintainer in system log?[1] What's wrong with choices letting vote/popularity work and users the freedom to choose? 2. What can we do about it when the package maintainer ignores legitimate technical issues and won't budge until after a PM steps in and orders such? [4] 3. What can we do about it when one PM takes over, calling shots and suppressing users' voices? Is there an appeal process? [1] wechat started on 6/30/2024 with an essentially nil placeholder commit. Actual and meaningful commits started on 11/5/2024 with nothing in between. https://aur.archlinux.org/cgit/aur.git/log/?h=wechat&ofs=50 wechat-bin started on 11/5/2024 https://aur.archlinux.org/cgit/aur.git/log/?h=wechat-bin As such wechat has a earlier start date in the system than wechat-bin. Yet if we look into the detail by no means can K claimed that his package arrived first in the repo. PM A expressed[6] his doubt towards K's legitimacy in his response to K's attempt. Quote "Claiming legitimacy over the other existing packages in a forced and unexpected manner like you're doing is not helpful." I'm a wechat-bin user and am sure it collected 50+ votes and not sure about wechat which currently also tallies 50 votes. That said, if anything, both packages have the support and deserve to stay instead of merging them in a haste. [2] Most comments can be found under packages' comment page https://aur.archlinux.org/packages/wechat-bin https://aur.archlinux.org/packages/wechat Select negative user comments towards K: ernest https://aur.archlinux.org/packages/wechat#comment-1004581 etoyz https://aur.archlinux.org/packages/wechat#comment-1004579 timefaker https://aur.archlinux.org/packages/wechat#comment-1004546 JoveYu https://aur.archlinux.org/packages/wechat#comment-1004540 envolution https://aur.archlinux.org/packages/wechat?O=30#comment-1003718 pr0m1x https://aur.archlinux.org/packages/wechat?O=40#comment-1003602 wszqkzqk https://aur.archlinux.org/packages/wechat?O=40#comment-1002784 Keep-Silence https://aur.archlinux.org/packages/wechat-bin#comment-1004718 JoveYu https://aur.archlinux.org/packages/wechat-bin#comment-1004832 duguyipiao https://aur.archlinux.org/packages/wechat-bin#comment-1004833 [3] https://lists.archlinux.org/archives/search?q=wechat&page=1&mlist=aur-requests%40lists.archlinux.org&sort=date-desc [4] After K's takeover, packaging quality took a dive and won't even launch properly. User provided feedback and K simply cited his deal with PM M to provide only vanilla packaging and won't budge, until M stepped in and ordered K to correct such and K complied. It's noticeable that PM M chose to ignore this important attitude detail that K has shown towards this package and simply summarized it as "Still only a minor issue (in the upstream desktop file) was reported and it was promptly fixed in less than one day." This kind of sweeping-it-under-the-rug approach is very much problematic and troublesome. This brings into question PM M's capability in fairly addressing inquiries. [5] https://aur.archlinux.org/packages/wechat-universal-bwrap#comment-1004422 [6] PM A had this to say in rebuke to maintainer K, quote "This indeed is an inappropriate usage of the "replaces" array. This should be a "conflicts" array at best. As you know (since you started the related discussion on the ML), the "wechat packaging situation" in the AUR is a bit controversial and difficult to deal with. Claiming legitimacy over the other existing packages in a forced and unexpected manner like you're doing is not helpful. I switched the "replaces" array to a "conflicts" one, please do not switch it back and give time for the AUR staff to deal with the pending requests about wechat related packages. Thanks for your comprehension & collaboration." https://aur.archlinux.org/packages/wechat?O=50#comment-1001449
Update [4] On Thu, Jan 2, 2025 at 9:57 AM S Umple <sumple@gmail.com> wrote:
Recently there's been a debacle about the wechat packaging in AUR. Essentially there were two competing packages wechat and wechat-bin [1] which are both popular. Package maintainer K(type user in AUR) of wechat took unbelievable measures to try to get rid of wechat-bin from AUR to begin with and was rebuked[6] by a PM A(type "Package Maintainer" in AUR). But eventually the same maintainer K succeeded when another PM M(type "Package Maintainer" in AUR) sided with him and took control of wechat-bin with a total revamp of the packaging.
Package maintainer K's actions are bizarre, hostile and malicious overall. There's absolutely no reasonable explanation as to why K had to take over the control except for the guideline which is flawed. Plenty of users submitted their negative views[2] towards such takeover. And it's also my observation that PM M did not do due diligence on this matter and just arbitrarily made a decision suppressing users' voice in the meantime.
The situation got worse under PM M's dealing with this matter that users feel ignored and betrayed that many **wrong** deletion requests[3] have been filed towards both wechat and wechat-bin, presumably by angry and frustrated users from both camps.
The chilling effect can also be seen here[5] when the most upvoted wechat related package is discussing a backup plan in case the same thing happens to it and the possibility of retreating to github, when one PM dictates the course of action and there's no way to appeal.
If we also look at the roles PMs play in this debacle, it also draws ire from K. K is not happy when his attempt was denied by PM A either. https://blog.kimiblock.top/2024/12/08/do-not-waste-time-on-aur/ https://blog.kimiblock.top/2024/08/23/aur-moderation/
The questions are: 1. Is there enough room for more than one packaging in AUR or does it have to favor the first package maintainer in system log?[1] What's wrong with choices letting vote/popularity work and users the freedom to choose? 2. What can we do about it when the package maintainer ignores legitimate technical issues and won't budge until after a PM steps in and orders such? [4] 3. What can we do about it when one PM takes over, calling shots and suppressing users' voices? Is there an appeal process?
[1] wechat started on 6/30/2024 with an essentially nil placeholder commit. Actual and meaningful commits started on 11/5/2024 with nothing in between. https://aur.archlinux.org/cgit/aur.git/log/?h=wechat&ofs=50
wechat-bin started on 11/5/2024 https://aur.archlinux.org/cgit/aur.git/log/?h=wechat-bin
As such wechat has a earlier start date in the system than wechat-bin. Yet if we look into the detail by no means can K claimed that his package arrived first in the repo. PM A expressed[6] his doubt towards K's legitimacy in his response to K's attempt. Quote "Claiming legitimacy over the other existing packages in a forced and unexpected manner like you're doing is not helpful." I'm a wechat-bin user and am sure it collected 50+ votes and not sure about wechat which currently also tallies 50 votes. That said, if anything, both packages have the support and deserve to stay instead of merging them in a haste.
[2] Most comments can be found under packages' comment page https://aur.archlinux.org/packages/wechat-bin https://aur.archlinux.org/packages/wechat
Select negative user comments towards K: ernest https://aur.archlinux.org/packages/wechat#comment-1004581 etoyz https://aur.archlinux.org/packages/wechat#comment-1004579 timefaker https://aur.archlinux.org/packages/wechat#comment-1004546 JoveYu https://aur.archlinux.org/packages/wechat#comment-1004540 envolution https://aur.archlinux.org/packages/wechat?O=30#comment-1003718 pr0m1x https://aur.archlinux.org/packages/wechat?O=40#comment-1003602 wszqkzqk https://aur.archlinux.org/packages/wechat?O=40#comment-1002784 Keep-Silence https://aur.archlinux.org/packages/wechat-bin#comment-1004718 JoveYu https://aur.archlinux.org/packages/wechat-bin#comment-1004832 duguyipiao https://aur.archlinux.org/packages/wechat-bin#comment-1004833
[4] After K's takeover, packaging quality took a dive and won't even launch properly. User provided feedback and K simply cited his deal with PM M to provide only vanilla packaging and won't budge, until M stepped in and ordered K to correct such and K complied. It's noticeable that PM M chose to ignore this important attitude detail that K has shown towards this package and simply summarized it as "Still only a minor issue (in the upstream desktop file) was reported and it was promptly fixed in less than one day." This kind of sweeping-it-under-the-rug approach is very much problematic and troublesome. This brings into question PM M's capability in fairly addressing inquiries. PM M had this to say "Stop this useless spam and provide valid reports if any, or else keep your personal PKGBUILD elsewhere if you don't want to use this package." as if M owns this community https://aur.archlinux.org/packages/wechat-bin#comment-1004808 and opposition are just spammer and the only option is to flee. I'm pretty sure that's against the value of OSS and M does not repesent the overall PM group and just an isolated case.
[5] https://aur.archlinux.org/packages/wechat-universal-bwrap#comment-1004422
[6] PM A had this to say in rebuke to maintainer K, quote "This indeed is an inappropriate usage of the "replaces" array. This should be a "conflicts" array at best. As you know (since you started the related discussion on the ML), the "wechat packaging situation" in the AUR is a bit controversial and difficult to deal with. Claiming legitimacy over the other existing packages in a forced and unexpected manner like you're doing is not helpful. I switched the "replaces" array to a "conflicts" one, please do not switch it back and give time for the AUR staff to deal with the pending requests about wechat related packages. Thanks for your comprehension & collaboration." https://aur.archlinux.org/packages/wechat?O=50#comment-1001449
On Thu, Jan 02, 2025 at 10:11:31AM -0500, S Umple wrote: <all snipped> JFC. All y'all sound like a bunch of petulant teens that got their phones taken away. Simple solution to all of this BS drama: 1. Copy PKGBUILD to your machine. 2. Edit said PKGBUILD to your needs. 3. Build and install it using the Arch provided tools. 4. 'Problem' and drama solved, and maybe the SNR on this list will improve immeasurably. Merell -- He who laughs last didn't get the joke.
Hello, First, there's no need to shorten our names. It's not something like a prohibited word. aur/wechat was wechat-uos-bwrap and wechat-uos-qt respectively before. They are merged into wechat on that very day because I'm busy doing other stuff before that and rushing will probably result in a broken package. The guidelines was already written well before. *I'd say you can't only support the rules when you benefit from it*, after all rules are rules. It exists to reduce confusion and prevent duplicated effort spent, which sadly did not happen to this app. Hostile actions are taken by some problematic users including cursing, spamming and harassment towards maintainer and her family. You have chosen to hide the fact that comment & requests spam is happening everyday on wechat and, the guidelines have moderation effect. Some users reported to me that multiple voting requests towards wechat-bin has happened in WeChat's official QQ group presumably to gain votes maliciously and mislead people to spam wechat's comment area, though this is not verified by me. My blog posts are indeed expressing negative feeling of prolonged action delay, it was not targeted at anyone and you can safely consider it as a rant and not some harassment that actually happened on the AUR. Regarding the replaces array, that happened before moderators sort out whether -bin is a duplicated package. There was nothing "silenced". Even harassment comments are still there as you have posted several of them. And you seems to be ignoring the fact wechat-bin still contains no modification. To me those users is simply having too much conformity and somewhat a cult of the leader. I hope this can be proven wrong though. Have a nice day. -- Sincerely, Kimiblock
it's quite sus that such great effort was taken just for the sake of being a maintainer (which is a non-paid volunteer role) of some package. i think some sort of security audit should take place to fully clarify this situation, because such pattern strongly remind some sort of social engineering hack attempt, for example for including some unwanted payload to one of "concurrenting" package, especially considering that wechat app itself is covered with rumors about spyware On Thu, Jan 2, 2025 at 4:57 PM Kimiblock Moe <kimiblock@icloud.com> wrote:
Hello,
First, there's no need to shorten our names. It's not something like a prohibited word.
aur/wechat was wechat-uos-bwrap and wechat-uos-qt respectively before. They are merged into wechat on that very day because I'm busy doing other stuff before that and rushing will probably result in a broken package.
The guidelines was already written well before. *I'd say you can't only support the rules when you benefit from it*, after all rules are rules. It exists to reduce confusion and prevent duplicated effort spent, which sadly did not happen to this app.
Hostile actions are taken by some problematic users including cursing, spamming and harassment towards maintainer and her family. You have chosen to hide the fact that comment & requests spam is happening everyday on wechat and, the guidelines have moderation effect. Some users reported to me that multiple voting requests towards wechat-bin has happened in WeChat's official QQ group presumably to gain votes maliciously and mislead people to spam wechat's comment area, though this is not verified by me.
My blog posts are indeed expressing negative feeling of prolonged action delay, it was not targeted at anyone and you can safely consider it as a rant and not some harassment that actually happened on the AUR.
Regarding the replaces array, that happened before moderators sort out whether -bin is a duplicated package.
There was nothing "silenced". Even harassment comments are still there as you have posted several of them. And you seems to be ignoring the fact wechat-bin still contains no modification. To me those users is simply having too much conformity and somewhat a cult of the leader. I hope this can be proven wrong though.
Have a nice day.
-- Sincerely, Kimiblock
I don't understand what you mean. Kimiblock has been the maintainer of wechat-uos-qt and wechat-bwrap-something since I started using arch a year ago. The merged "wechat" is just one less package, and the vanilla “原汁原味儿” wechat-bin was already restored 4 days ago. I don't understand all the uproar. If you don't want the wechat sandbox fixes, then just don't use that. -- Cheers, Aᴀʀᴏɴ
On 1/2/25 3:57 PM, S Umple wrote:
Recently there's been a debacle about the wechat packaging in AUR.
"Recently" is an interesting word choice, this "debacle" has been going for months [1]...
Essentially there were two competing packages wechat and wechat-bin [1] which are both popular. Package maintainer K(type user in AUR) of wechat took unbelievable measures to try to get rid of wechat-bin from AUR to begin with and was rebuked[6] by a PM A(type "Package Maintainer" in AUR). But eventually the same maintainer K succeeded when another PM M(type "Package Maintainer" in AUR) sided with him and took control of wechat-bin with a total revamp of the packaging.
Package maintainer K's actions are bizarre, hostile and malicious overall. There's absolutely no reasonable explanation as to why K had to take over the control except for the guideline which is flawed. Plenty of users submitted their negative views[2] towards such takeover. And it's also my observation that PM M did not do due diligence on this matter and just arbitrarily made a decision suppressing users' voice in the meantime.
As said above, this situation has been going for months and is particularly difficult to deal with. We have multiple separate mail threads about the overall situation and most comments on the AUR are not written in English which makes it non-trivial to deal with from a moderation standpoint. Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
The situation got worse under PM M's dealing with this matter that users feel ignored and betrayed that many **wrong** deletion requests[3] have been filed towards both wechat and wechat-bin, presumably by angry and frustrated users from both camps.
Which is a totally unjustified and unacceptable behavior that resulted in a waste of time for everyone (both users and staff...).
The chilling effect can also be seen here[5] when the most upvoted wechat related package is discussing a backup plan in case the same thing happens to it and the possibility of retreating to github, when one PM dictates the course of action and there's no way to appeal.
That's also a non-English comment, which doesn't help... It's important that comments, contributions & discussions remain intelligible to most people (including the staff). As for PMs "dictating" the course, keep in mind that Arch Linux is not run as a democracy [2]. While the staff shall always attempt to implement universally peaceful solutions, they have the final word (which, unfortunately, cannot always please everyone).
If we also look at the roles PMs play in this debacle, it also draws ire from K. K is not happy when his attempt was denied by PM A either. https://blog.kimiblock.top/2024/12/08/do-not-waste-time-on-aur/ <https://blog.kimiblock.top/2024/12/08/do-not-waste-time-on-aur/> https://blog.kimiblock.top/2024/08/23/aur-moderation/ <https:// blog.kimiblock.top/2024/08/23/aur-moderation/>
I'll do as if you didn't share those links... This goes beyond the scope of the AUR, which this mailing list and this overall situation is about. Of course, I could just throw that "I'm not paid enough to deal with such public rant about my decision making" (since this is all voluntary work) and bail out, but I frankly don't (want to) care.
The questions are: 1. Is there enough room for more than one packaging in AUR or does it have to favor the first package maintainer in system log?[1] What's wrong with choices letting vote/popularity work and users the freedom to choose?
Duplicate packages are not allowed, this is a clearly stated rule [3]. As for the second question, the users are not the ones maintaining and moderating the platform. Keep in mind that the AUR is a platform entirely maintained and moderated by the staff on a voluntary basis. Some rules have to be established, otherwise the platform itself cannot live (or rather not reliably). Users are still free to maintain their own PKGBUILD locally or in an unofficial repository [4] though.
2. What can we do about it when the package maintainer ignores legitimate technical issues and won't budge until after a PM steps in and orders such? [4]
You could raise it to the moderation (via email or AUR requests) until a PM steps in (which is currently happening).
3. What can we do about it when one PM takes over, calling shots and suppressing users' voices? Is there an appeal process?
Short and rough answer: Nothing [2]. This situation is difficult and controversial enough that it feels that, whatever the final decision is, people will be unhappy anyway. I'd like to remind everyone that the AUR is a platform entirely maintained and moderated on a volunteer basis for the sole purpose of providing a freely accessible and community maintained package repositories. We expect the community to do their part to contribute to a healthy community and environment. I'll be completely transparent here, the only thing that came out from the recent events* is that we started discussing another potential outcome to all of this: banning any wechat related packages from the AUR altogether. After all, it seems like there isn't any other distro shipping wechat [5] and seeing the controversy & the overall pain it causes from a moderation stand point to our staff, that's an actual possibility we are eventually considering. Please, work _with_ us. *The erroneous deletion requests repeatedly submitted out of anger, impolite and disrespectful behaviors [6], shit shows in comments, wechat related packages repeatedly re-submitted after being deleted by the staff and so on... [1] https://lists.archlinux.org/hyperkitty/list/aur-general@lists.archlinux.org/... [2] https://terms.archlinux.org/docs/code-of-conduct/#respect-the-staff [3] https://wiki.archlinux.org/title/AUR_submission_guidelines#Rules_of_submissi... [4] https://wiki.archlinux.org/title/Unofficial_user_repositories [5] https://repology.org/project/wechat/versions [6] https://lists.archlinux.org/archives/list/aur-requests@lists.archlinux.org/m... -- Regards, Robin Candau / Antiz
On Thu, 2025-01-02 at 18:05 +0100, Robin Candau wrote:
As said above, this situation has been going for months and is particularly difficult to deal with. We have multiple separate mail threads about the overall situation and most comments on the AUR are not written in English which makes it non-trivial to deal with from a moderation standpoint.
Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
I would very much like to second this statement. Mulfone and Antiz did not make their decisions in isolation; I approve of their decisions and conduct. The submission guidelines [1] are clear on the rule against duplicates. Sometimes variants make sense, but that doesn't mean an endless line of small tweaks should be in the AUR. You are free to choose a slightly different set of options, but there is no reason the resulting PKGBUILDs have to live in the AUR. I would also like to emphasize the remarks about writing in langauges other than English. We are a very international community and as such when discussing we should write in a language that is intelligible for everyone. We cannot moderate what we do not understand. The team of Package Maintainers tries to keep the AUR useful and welcoming to all, but it is sometimes hard to please everyone. In situations like this, where many people are claiming different things and there isn't a clear "correct" answer, decisions have to be made, and we do our best. As repeatedly stated, this is a volunteer position. Cheers, Bert. [1] https://wiki.archlinux.org/title/AUR_submission_guidelines
It's good that this is an open discussion so we can get all the facts and misinformation presented for everyone to see and judge. Let's get the main points addressed first. Details are next. This is the subject of this discussion and the main points are about: ** AUR guideline on `Duplicate` packages, and Package Maintainers ** ** Firstly: AUR guideline on `Duplicate` packages This is the current rule:
Check the AUR if the package already exists. If it is currently maintained, changes can be submitted in a comment for the maintainer's attention. If it is unmaintained or the maintainer is unresponsive, the package can be adopted and updated as required. Do not create duplicate packages.
This rule *is* the problem. This particular problem between wechat and wechat-bin could have been avoided if this rule were not listed as such. This is the basis for K and M. Rule is rule when it's in effect. But don't forget rules can be changed, even though changing a rule can be hard and it takes courage. This rule as of now is apparently not perfect yet works relatively well. Are AUR staff willing to discuss and amend it to best cover more cases? Or perhaps it's not worthy of the effort to gain marginal improvement. By the way I believe my original question 1 remains unaddressed even if this rule stays the same in the foreseeable future. ** Secondly: Package Maintainers I give up on this. Details is below and in first message of this thread(I don't think it's read carefully or how can someone agrees that M did a good job there):
Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
Users are still free to maintain their own PKGBUILD locally or in an unofficial repository [4] though.
Duplicate packages are not allowed, this is a clearly stated rule
Some rules have to be established
voluntary work
Arch Linux is not run as a democracy
(which, unfortunately, cannot always please everyone).
cannot always please everyone
Please, work _with_ us.
PMs: Just be fair as much as you can. It's not democracy, PMs/staff have the final say. Your house, your rules. My way or the highway etc I get it. I'm not asking any of that, any more. Just be fair. You manage the domain and it has responsibilities. Also bear in mind that it goes without saying you are probably doing 98% of a good job keeping AUR the way it is, but at the same time you have to be able to stand will-intended criticism be it true or false, improve if you can on the true ones and not let the false ones get to you. It's not easy or pleasant to do this. For all I care wechat-bin is the 2nd AUR package I use and I can live without it. I am volunteering the same as you guys spending the time and effort to help perhaps make things are a bit better, albeit not at a level you guys are contributing and hiding behind curtains. I am just a user I have no right I did my part to raise a flag. The rest is up to you folks. You guys are in charge. "arbitrary" in my opinion is a neutral word and if it's not acceptable especially given what M has done I'd say it's not my problem. Someone needs to reflect on what's been said and done. Same goes for K who whines about all of a sudden stones being thrown in a certain direction. Since CoC is quoted, apply any penalty as you fit.
Which is a totally unjustified and unacceptable behavior that resulted in a waste of time for everyone (both users and staff...). I meant to say "wrongful" instead of "wrong". Excuse my english.
After all, it seems like there isn't any other distro shipping wechat [5] and seeing the controversy & the overall pain it causes from a moderation stand point to our staff, that's an actual possibility we are eventually considering. wechat is a big deal in China, south-east Asia and it will keep expanding. It's available in popular distro UOS and Deepin appstores in China before until not long ago Tencent finally produces official binary based on QT which is a very positive gesture well received by any Chinese linux user. Ask around. But hey if staff decides to ban it based on misinformation, so be it.
for the sole purpose of providing a freely accessible and community maintained package repositories. Yet second opinion or choices nuanced as a `duplicate` package is not an option in AUR due to various practicality reasons. What an irony.
Lastly, 50 votes gets you around top1500 among ~90k AUR packages. It is significant in my view. And I told K the same thing: RESPECT. Or just pull out the rule book, enforce the rule and wipe it clean, because K brought it to your attention because of this xxxxxx rule. K and you had to do it this way. What a world. Peace. On Thu, Jan 2, 2025 at 12:44 PM Bert Peters <bertptrs@archlinux.org> wrote:
On Thu, 2025-01-02 at 18:05 +0100, Robin Candau wrote:
As said above, this situation has been going for months and is particularly difficult to deal with. We have multiple separate mail threads about the overall situation and most comments on the AUR are not written in English which makes it non-trivial to deal with from a moderation standpoint.
Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
I would very much like to second this statement. Mulfone and Antiz did not make their decisions in isolation; I approve of their decisions and conduct.
The submission guidelines [1] are clear on the rule against duplicates. Sometimes variants make sense, but that doesn't mean an endless line of small tweaks should be in the AUR. You are free to choose a slightly different set of options, but there is no reason the resulting PKGBUILDs have to live in the AUR.
I would also like to emphasize the remarks about writing in langauges other than English. We are a very international community and as such when discussing we should write in a language that is intelligible for everyone. We cannot moderate what we do not understand.
The team of Package Maintainers tries to keep the AUR useful and welcoming to all, but it is sometimes hard to please everyone. In situations like this, where many people are claiming different things and there isn't a clear "correct" answer, decisions have to be made, and we do our best. As repeatedly stated, this is a volunteer position.
Cheers,
Bert.
[1] https://wiki.archlinux.org/title/AUR_submission_guidelines
(sorry about the formatting. resend) It's good that this is an open discussion so we can get all the facts and misinformation presented for everyone to see and judge. Let's get the main points addressed first. Details are next. This is the subject of this discussion and the main points are about: ** AUR guideline on `Duplicate` packages, and Package Maintainers ** ** Firstly: AUR guideline on `Duplicate` packages This is the current rule:
Check the AUR if the package already exists. If it is currently maintained, changes can be submitted in a comment for the maintainer's attention. If it is unmaintained or the maintainer is unresponsive, the package can be adopted and updated as required. Do not create duplicate packages.
This rule *is* the problem. This particular problem between wechat and wechat-bin could have been avoided if this rule were not listed as such. This is the basis for K and M. Rule is rule when it's in effect. But don't forget rules can be changed, even though changing a rule can be hard and it takes courage. This rule as of now is apparently not perfect yet works relatively well. Are AUR staff willing to discuss and amend it to best cover more cases? Or perhaps it's not worthy of the effort to gain marginal improvement. By the way I believe my original question 1 remains unaddressed even if this rule stays the same in the foreseeable future. ** Secondly: Package Maintainers I give up on this. Details is below and in first message of this thread(I don't think it's read carefully or how can someone agrees that M did a good job there):
Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
Users are still free to maintain their own PKGBUILD locally or in an unofficial repository [4] though.
Duplicate packages are not allowed, this is a clearly stated rule
Some rules have to be established
voluntary work
Arch Linux is not run as a democracy
(which, unfortunately, cannot always please everyone).
cannot always please everyone
Please, work _with_ us.
PMs: Just be fair as much as you can. It's not democracy, PMs/staff have the final say. Your house, your rules. My way or the highway etc I get it. I'm not asking any of that, any more. Just be fair. You manage the domain and it has responsibilities. Also bear in mind that it goes without saying you are probably doing 98% of a good job keeping AUR the way it is, but at the same time you have to be able to stand will-intended criticism be it true or false, improve if you can on the true ones and not let the false ones get to you. It's not easy or pleasant to do this. For all I care wechat-bin is the 2nd AUR package I use and I can live without it. I am volunteering the same as you guys spending the time and effort to help perhaps make things are a bit better, albeit not at a level you guys are contributing and hiding behind curtains. I am just a user I have no right I did my part to raise a flag. The rest is up to you folks. You guys are in charge. "arbitrary" in my opinion is a neutral word and if it's not acceptable especially given what M has done I'd say it's not my problem. Someone needs to reflect on what's been said and done. Same goes for K who whines about all of a sudden stones being thrown in a certain direction. Since CoC is quoted, apply any penalty as you fit.
Which is a totally unjustified and unacceptable behavior that resulted in a waste of time for everyone (both users and staff...).
I meant to say "wrongful" instead of "wrong". Excuse my english.
After all, it seems like there isn't any other distro shipping wechat [5] and seeing the controversy & the overall pain it causes from a moderation stand point to our staff, that's an actual possibility we are eventually considering.
wechat is a big deal in China, south-east Asia and it will keep expanding. It's available in popular distro UOS and Deepin appstores in China before until not long ago Tencent finally produces official binary based on QT which is a very positive gesture well received by any Chinese linux user. Ask around. But hey if staff decides to ban it based on misinformation, so be it.
for the sole purpose of providing a freely accessible and community maintained package repositories.
Yet second opinion or choices nuanced as a `duplicate` package is not an option in AUR due to various practicality reasons. What an irony. Lastly, 50 votes gets you around top1500 among ~90k AUR packages. It is significant in my view. And I told K the same thing: RESPECT. Or just pull out the rule book, enforce the rule and wipe it clean, because K brought it to your attention because of this xxxxxx rule. K and you had to do it this way. What a world. Peace. On Thu, Jan 2, 2025 at 12:44 PM Bert Peters <bertptrs@archlinux.org> wrote:
On Thu, 2025-01-02 at 18:05 +0100, Robin Candau wrote:
As said above, this situation has been going for months and is particularly difficult to deal with. We have multiple separate mail threads about the overall situation and most comments on the AUR are not written in English which makes it non-trivial to deal with from a moderation standpoint.
Decisions have not been taken arbitrarily, Muflone is just trying to deal with this situation and I'm personally very thankful that someone is actually willing to try to solve this mess (excuse the term, but frankly that's what it is at that point). The rest of the staff approves Muflone's decision / judgment (which is purely about trying to make the package work, regardless of the maintainer). Such assumptions and judgments about staff decision's making are not welcomed here [2]. Keep in mind that this is voluntary work.
I would very much like to second this statement. Mulfone and Antiz did not make their decisions in isolation; I approve of their decisions and conduct.
The submission guidelines [1] are clear on the rule against duplicates. Sometimes variants make sense, but that doesn't mean an endless line of small tweaks should be in the AUR. You are free to choose a slightly different set of options, but there is no reason the resulting PKGBUILDs have to live in the AUR.
I would also like to emphasize the remarks about writing in langauges other than English. We are a very international community and as such when discussing we should write in a language that is intelligible for everyone. We cannot moderate what we do not understand.
The team of Package Maintainers tries to keep the AUR useful and welcoming to all, but it is sometimes hard to please everyone. In situations like this, where many people are claiming different things and there isn't a clear "correct" answer, decisions have to be made, and we do our best. As repeatedly stated, this is a volunteer position.
Cheers,
Bert.
[1] https://wiki.archlinux.org/title/AUR_submission_guidelines
Hey, Packages are only considered duplicate if there aren't any actual differences. Packages that have real differences (see official electron app -bin's vs versions that adapt to use Arch's electron) won't be deleted. However, the recent new packages created on the AUR other than Kiwiblock's are nearly all duplicates of wechat-bin. BTW, welcome to an Arch mailing list! Please take a look at [our guidelines][https://wiki.archlinux.org/title/General_guidelines#Mailing_lists]. In particular, people don't like it when you do what is called "top posting"—quoting the email you reply to at the *end* of your email message. (Also, to the other readers of this thread: As a Chinese speaker, I don't think S Umple meant to sound sarcastic here, and such apparent tone is probably a translation error; except for the last two paragraphs, that is.) -- Cheers, Aᴀʀᴏɴ
participants (7)
-
Aaron Liu
-
Actionless Loveless
-
Bert Peters
-
Kimiblock Moe
-
Merell L. Matlock, Jr.
-
Robin Candau
-
S Umple