[aur-general] Deletion of vi-vim-symlink and neovim-symlinks
Recently vi-vim-symlink and neovim-symlinks were deleted from the AUR. These are used by a lot of people to set up their systems, and a lot of people want to know why [1]. They do not violate any AUR rules and have been available in the AUR for a very long time, well over 6 years for neovim-symlinks (about the time neovim became a thing) and over a decade for vi-vim-symlink, but I can't date that one as it predates the current AUR system by a long ways. The reason given for this was:
This is more appropriately managed via the user shell's PATH/configuration. This is completely bogus. First off, PATH has nothing to do with anything here. Second, the shell's configuration doesn't work. Making aliases doesn't work when other programs call vi or vim. It doesn't even work for something like `sudo vim`. This is not a substitute in any way.
Just because one TU doesn't find it personally useful or doesn't understand why someone would use a package doesn't mean it shouldn't exist. This is a similar situation to the -bin package issue, except there's not even a rule to misinterpret on this one. If you don't want to use a package, don't, but please don't force your views on everyone. [1] https://www.reddit.com/r/archlinux/comments/u8o31u/why_was_vivimsymlink_from...
On 2022-04-22 09:58, Doug Newgard via aur-general wrote:
Recently vi-vim-symlink and neovim-symlinks were deleted from the AUR. These are used by a lot of people to set up their systems, and a lot of people want to know why [1]. They do not violate any AUR rules and have been available in the AUR for a very long time, well over 6 years for neovim-symlinks (about the time neovim became a thing) and over a decade for vi-vim-symlink, but I can't date that one as it predates the current AUR system by a long ways.
The reason given for this was:
This is more appropriately managed via the user shell's PATH/configuration. This is completely bogus. First off, PATH has nothing to do with anything here. Second, the shell's configuration doesn't work. Making aliases doesn't work when other programs call vi or vim. It doesn't even work for something like `sudo vim`. This is not a substitute in any way.
Just because one TU doesn't find it personally useful or doesn't understand why someone would use a package doesn't mean it shouldn't exist. This is a similar situation to the -bin package issue, except there's not even a rule to misinterpret on this one. If you don't want to use a package, don't, but please don't force your views on everyone.
Hi, Doug! Firstly, please consider that I was not the one to have accepted the request I made, so at least *two* TUs didn't find it useful ;). Secondly, please remember that TUs exist to actually "force" our views onto everyone via quality control. These packages fall into personal configuration territory with a rather heavy-handed approach. I'd encourage you to become more familiar with the shell environment so that you may configure your system to your liking [1]. There are simple ways to set things on a global scale (I do it myself, too!), so this article may be useful in helping you achieve these goals. A package having a modicum of popularity does not qualify its inclusion! Remember, Arch cares about correctness, not convenience. These packages largely replicate personal workstation configuration through symlinks rather than, say, aliases in the global shell. Thirdly, I've taken a look at your posting history - I would encourage a little more calm when approaching the Arch community as your aggression does your otherwise intelligent speech a disservice. Hope this helps, and have a great rest of your day! [1] https://wiki.archlinux.org/title/Bash
On Fri, 22 Apr 2022 08:31:27 -0700 Brett Cornwall <ainola@archlinux.org> wrote:
On 2022-04-22 09:58, Doug Newgard via aur-general wrote:
Recently vi-vim-symlink and neovim-symlinks were deleted from the AUR. These are used by a lot of people to set up their systems, and a lot of people want to know why [1]. They do not violate any AUR rules and have been available in the AUR for a very long time, well over 6 years for neovim-symlinks (about the time neovim became a thing) and over a decade for vi-vim-symlink, but I can't date that one as it predates the current AUR system by a long ways.
The reason given for this was:
This is more appropriately managed via the user shell's PATH/configuration. This is completely bogus. First off, PATH has nothing to do with anything here. Second, the shell's configuration doesn't work. Making aliases doesn't work when other programs call vi or vim. It doesn't even work for something like `sudo vim`. This is not a substitute in any way.
Just because one TU doesn't find it personally useful or doesn't understand why someone would use a package doesn't mean it shouldn't exist. This is a similar situation to the -bin package issue, except there's not even a rule to misinterpret on this one. If you don't want to use a package, don't, but please don't force your views on everyone.
Hi, Doug!
Firstly, please consider that I was not the one to have accepted the request I made, so at least *two* TUs didn't find it useful ;).
Secondly, please remember that TUs exist to actually "force" our views onto everyone via quality control.
And where did you come up with this? Rules violations, sure, but forcing your personal preferences via "quality control", being completely subjective.
These packages fall into personal configuration territory with a rather heavy-handed approach.
So you're saying this falls under: https://wiki.archlinux.org/title/AUR_submission_guidelines#Rules_of_submissi...
Make sure the package you want to upload is useful. Will anyone else want to use this package? Is it extremely specialized? If more than a few people would find this package useful, it is appropriate for submission. The AUR and official repositories are intended for packages which install generally software and software-related content, including one or more of the following: executable(s); configuration file(s); online or offline documentation for specific software or the Arch Linux distribution as a whole; media intended to be used directly by software."
Note that this specifically says that configurations *are* allowed, as long as they are useful to others. The popularity that you dismiss prove that these packages qualify. As such, I think I'll be re-uploading unless you can show me where this violates the rules.
On 2022-04-22 10:42, Doug Newgard via aur-general wrote:
[...] As such, I think I'll be re-uploading unless you can show me where this violates the rules.
Without going into the details of this dispute, I will point out that this is not how the AUR works: the Trusted Users are the moderators of the AUR, they have the final power of decision over whether a package is allowed or not. Naturally this requires some degree of interpretation of the existing rules, which can sometimes be a somewhat subjective process: the submission guidelines, as the name implies, are just a set of relatively loose guidelines, not a rigid legal definition of what is allowed and what is not. The decisions made by Trusted Users are not irrevocable of course. Therefore asking for clarification if you disagree with a certain action is perfectly fine; unilaterally re-uploading a deleted package without awaiting consent from the Trusted Users however is certainly not. Best, Jonas -- Jonas Witschel Arch Linux Developer, Trusted User and security team member PGP key: FE2E6249201CA54A4FB90D066E80CA1446879D04
On 22-04-22 10:42, Doug Newgard via aur-general wrote:
I think I'll be re-uploading unless you can show me where this violates the rules.
No need, both packages have been restored. They do not have a maintainer, so feel free to pick them back up. -- George Rawlinson
On 2022-04-22 08:31:27, Brett Cornwall via aur-general wrote:
Secondly, please remember that TUs exist to actually "force" our views onto everyone via quality control.
Is that so? In my understanding the TUs job is to enforce the criteria laid out in the submission rules not your personal opinion. Otherwise you might just as well close the AUR, because there won't be many packages left if you remove everything that someone does not like for whatever reason. On 2022-04-22 08:31:27, Brett Cornwall via aur-general wrote:
These packages fall into personal configuration territory with a rather heavy-handed approach.
Yes, this is configuration, but imho the approach is rather elegant and there is nothing wrong with putting configuration into a package and share it with others. On 2022-04-22 08:31:27, Brett Cornwall via aur-general wrote:
I'd encourage you to become more familiar with the shell environment so that you may configure your system to your liking [1]. There are simple ways to set things on a global scale (I do it myself, too!), so this article may be useful in helping you achieve these goals.
It is perfectly fine to do it that way, but others want to do it differently and that is perfectly fine as well. On 2022-04-22 08:31:27, Brett Cornwall via aur-general wrote:
A package having a modicum of popularity does not qualify its inclusion!
From https://wiki.archlinux.org/title/AUR_submission_guidelines Make sure the package you want to upload is useful. Will anyone else want to use this package? Is it extremely specialized? If more than a few people would find this package useful, it is appropriate for submission.
This talks about usefulness above all else, and only later about the amount of users. By the way according to pkgstats the popularity of neovim-symlinks is 1.49, which should be more than enough to clear the "more than a few people" bar. On 2022-04-22 08:31:27, Brett Cornwall via aur-general wrote:
Remember, Arch cares about correctness, not convenience. These packages largely replicate personal workstation configuration through symlinks rather than, say, aliases in the global shell.
In my opinion it is more correct to use a package for this than to put this in the (global) shell config, especially in the absence of a alternatives system. Kind Regards Marcel
participants (5)
-
Brett Cornwall
-
Doug Newgard
-
George Rawlinson
-
Jonas Witschel
-
Marcel Röthke