Replies: 29 comments 14 replies
-
this is simply github's interface being bad at showing a manual merge. the PR was applied in commit 1c2891d |
Beta Was this translation helpful? Give feedback.
-
That is good to know, thank you, I was not aware. |
Beta Was this translation helpful? Give feedback.
-
|
Any how I am not here to cast stones at anyone, I am really looking to try to brainstorm ideas on how to improve communication between users and maintainers. One idea is to add a mechanism to xbps to deliver a message to all users of a package in a really hard to miss way, without pushing a change to the package. There is no reason this system has to only apply to a specific package, this mechanism could also be a good way to reach out to all users. This could be nice for when trying to gauge consensus on a topic, a poll with real numbers could help a lot when making choices. |
Beta Was this translation helpful? Give feedback.
-
|
Just subscribe to issues of the repository. |
Beta Was this translation helpful? Give feedback.
-
That is not a bad suggestion, but I see some flaws that are not a good for maintainer to user communication. Not all users have a github account (or want one), it requires users to be familiar with github to know to do this and it does not provide good granularity. The reason I suggested adding this to xbps the package manager, is because 100% of void users have it. |
Beta Was this translation helpful? Give feedback.
-
|
You are painting a negative picture of how bad maintainers have been. Your issue is that in two instances packages have been removed from the repository. In one instance the packages were later added back because, there was not enough communication between maintainers and contributors and while most of them either agreed with the removal or abstained from it, some stepped up and took over maintenance. This could have been done better by tagging all contributors and maintainers of the relevant packages. For the minecraft launcher, I don't see any issue at all. The maintainers treated you right and did nothing wrong. They explained why the package was removed and suggested or agreed that an install message would have been good. I deleted my earlier response since it failed to address the whole picture. I don't think there is any issue that has not been addressed already and I definitely don't see a pattern of issues. |
Beta Was this translation helpful? Give feedback.
-
|
With the minecraft launcher the problem is not that it was removed, that would of been the correct choice! The problem is that I installed "PolyMC" onto my system, I did not install "Prism". An "update" to "PolyMC" quietly replaced one set of software with another. This happened quietly and without any communication. I only found out about this after I executed a program on my computer. A post install message would of been an improvement, but that is still not great move. If the concern was security then PolyMC should of been removed and not replaced, a new package should of been created for Prism and explicitly installed by the user. This may seem like "nitpicking", but lets imagine a slightly different scenario where the fork added an anti feature that is really important to some users to not have. Lets imagine "Prism" added extra telemetry. In this theoretical scenario a user who is careful about what they add to there system would of been exposed to a package without ever having a choice in it. |
Beta Was this translation helpful? Give feedback.
-
|
If an update removes a package that I use, I would appreciate a post install message telling me so, telling me why and what can I do instead (telling me to manually install an alternative package for example). |
Beta Was this translation helpful? Give feedback.
-
|
With the crypto PR, I am upset that out of the blue void is taking an anti crypto stance & then doing a wide purging of packages it's users depend on, regardless of how well they where maintained or if there was any willing maintainers.
I do not understand why it would be removed. The PR adds new language to the CONTRIBUTING.md file, clearly saying crypto is not welcome. Unlike the segment about browser forks, there is no maybe, it also includes no explanation as to why, and is worded in a more strict way. Why will a browser fork be generally not accepted and the crypto flat out "are not accepted"? If there is no intent on being neutral, then please make your political or social stance explicit and clear, if crypto people are not welcome in the void ecosystem then please just tell us that. the bit on crypto: |
Beta Was this translation helpful? Give feedback.
-
The packages are being listed for removal, you still have to actually accept the updates before the package is removed. A post install message can only be shown by package that are still installed, new messages can't be put in old packages. If there is something that pulling in the old or the new version, we can put an install message in it, like it was done with There are a lot of packages that are being removed all the time, for good reasons. Without removing them "automatically", users won't be able to update their system or install packages at all if they cause dependency conflicts by depending outdated libraries or packages. The "automatic" (user accepted) removals make it possible to update and remove packages at the same time, without breaking any dependencies. Users would have to manually remove the breaking packages while breaking potentially other currently installed packages that depend on them using If you really don't like the "automatic" removals, there are two options for you.
If you do that, and you run into any troubles that you can't solve yourself, make sure you mention that you do this and first try to check if that is the cause.
There was a single person that somewhat objected it in the pull request. And we already accepted that this wasn't good and more maintainers and contributors should have been pinged.
Nobody linked or showed any logs, because there are no links, the channel is public and everyone is welcomed to join. Nobody here or in the PR afterwards said that the IRC discussion was enough to accept the removal. Nothing was or is being hidden, it was a public pull request by a contributor that was open for 12 days, then it was merged after some approval. I don't see why why have to discuss this at length, nobody disapproved of bringing the packages back when contributors and maintainers wanted to bring them back.
Literally just because some contributor opened a pull request and there was some chatter on IRC about the packages being outdated and not well maintained or of interest for anymore. And again it was already reverted, and people accepted that it wasn't sufficient enough without pinging/tagging more maintainers.
Find a web3 based distributions with NFTs and shit, "crypto people" lol. Its probably generally a good decision to not accept any "new" crypto currency packages anymore, would've been even better if that had been done earlier. The wording is what it is, there was simply no big discussion and I don't think the wording is wrong in any way. |
Beta Was this translation helpful? Give feedback.
-
It's awesome that some chatter in a non officially logged IRC chat can create a lot of chaos /s, the communication process failed and it wont be changed magically, this will not be the last time. Also this wonderful new precedent will be used in the future, like how a bad action was used to justify the crypto removal PR. Even if after the fact it was "accepted" that it wasn't good. Also the claim in the PR opening comment is false, not all the PRs referenced where dead or tiny unimportant projects.
See this is what I was fearing, an entire technology will now be marked as irredeemable, because morons and grifters found out and decided to get rich quick and don't even know what it is that they are buying. As for web3 yes most of it is a dumpster fire, a lot of it is actually a privacy nightmare worse then what we have now. But I also do not want to toss out the baby with the bath water. For example Namecoin & ENS are interesting technologies, they provide a solution that gives you all three properties of Zooko's triangle. The current system of centralized DNS is problematic when it comes to censorship. Any how thank you for insulting me directly, its so nice lol!
So Void is a not a general purpose operating system and is instead a playpen for only the stuff a few people find interesting? (note: this post has edits) |
Beta Was this translation helpful? Give feedback.
-
Nothing is being justified, the communication process is appropriate for 99% of daily changes that are merged. This does not set any precedent, other than that yes changes like that should require more approval.
Awesome another made up bad interaction you can fit your narrative.
Merging packages that are going to be build on void linux build servers and are in the official repository are merged at the void teams discretion. If they don't get actively maintained anymore, if they need rebuilds due to dependencies, they will fall back on people who are regular contributors. What a shocker, every single open source project and linux distribution works like that. |
Beta Was this translation helpful? Give feedback.
-
|
alright I get it, I have escalated this too far, nothing good will happen, but how else was this intended to be taken, how could this be interpreted in a positive way, if no negative implication was intended?
|
Beta Was this translation helpful? Give feedback.
-
|
Yes I made fun of you calling yourself a crypto person. |
Beta Was this translation helpful? Give feedback.
-
|
Well I am a bit of a clown here, I created a issue titled "Better Communication" and I did not do a good job at starting this conversation, my initial post was not written in a way conducive to positive change, its function as more of a vent for negative nonsense, I should of taken a different angle.
That is the best wording I could come up with for people with an interest in cryptocurrency technology. Why should people that hold that interest be ridiculed & lumped in with scammers? Is there a better term you would like to suggest, that is less silly? |
Beta Was this translation helpful? Give feedback.
-
|
I do not think you, or anyone else is getting any value from this interaction, and I am not getting anything good out of this too. The way the PRs where handled where not good & a structural change could help there. Act then explain/correct if asked later is not a good experience, its a very negative one. The perceived lack of communication do not inspire trust & make it look like that the intent was to make big changes go unnoticed. I now understand I should of not immediately assumed malice, that is on me, I am sorry. I do not wish to craft a fake narrative. If this issue is un-salvageable, It might be best to close this issue. Even if we close this issue, I still urge that a conversation should be taken to improve the situation. Maybe having it in private between usual members without a nutjob like me will hopefully result in something good. Opensource is often thankless and this issue will sadly be a example of that, Keep up the wonderful work on Void, hopefully things will improve. |
Beta Was this translation helpful? Give feedback.
-
As the person who wrote the PR, what claim are you talking about, and where are they contained in the PR's text? Note that no edits were made to the PR |
Beta Was this translation helpful? Give feedback.
-
|
To be transparent, this post #45892 (comment) was written with anger, I was attributing malice to your PR and was "reading in between the lines", that was wrong of me.
I will attempt to explain my thoughts at that moment. I was viewing the entire PR from an adversarial prospective, I believed the PR never had attempted consensus and that the initial post was simply a cover story to remove stuff that was not liked. So discarding the IRC statement as unverifiable, there was no more text that could explain the justification for what was in my eyes a huge change. Lastly this PR was huge touching all crypto, usually most PR I have looked at touch a few packages. So that read to me as a attempt to do this quickly before people would notice. To be fair, you did call your branch (edit: typo fix) |
Beta Was this translation helpful? Give feedback.
-
|
I do not want to blame any individual, adversarial or not I do not think your PR is the problem. The problem is that it was merged. Also #44778 is not an acceptable compromise, the reviewer should be neutral & this is a joke https://github.com/void-linux/void-packages/pull/44778/files/3c353891313291f29d219a339dfb68db1ff2f44d As is, this looks more like damage control at best. |
Beta Was this translation helpful? Give feedback.
-
|
I understand your frustration, but herewith a few comments:
|
Beta Was this translation helpful? Give feedback.
-
|
I think some of these anxieties could be alleviated if the handling of removed packages was changed, i.e., if the package is removed from the repos, it shouldn't be uninstalled from users systems. I've seen this mentioned a few times but I've no idea (and won't pretend to) of the feasibility or appetite to change this. |
Beta Was this translation helpful? Give feedback.
-
|
It is often feasible for packages to be removed from the repositories and left on users systems for awhile, but that falls apart quickly when library dependencies roll on and need to be upgraded. Users wishing to upgrade will be hit with "unresolvable shlib" error messages because the package manager will fail to update a library when the update removes a version required by some still-present-but-officially-removed dependent. Perhaps this message could better highlight the nature of the error and avoid some support requests, but the underlying problem remains. It is because of this that we added Users are always welcome to ignore or hold the We generally expect that users read the list of updated packages and think for a moment before approving the actions of XBPS, but it is much more common that people blindly approve updates (even adding At least XBPS keeps a cache of packages that users can reinstall in this case. |
Beta Was this translation helpful? Give feedback.
-
|
That's a very clear explanation, thanks!
Yeah, this always confuses me. |
Beta Was this translation helpful? Give feedback.
-
This is exactly how I too perceived these events. |
Beta Was this translation helpful? Give feedback.
-
|
Why the "cryptoshit" branch named by @0x5c and merged basing on some IRC discussed consensus did not arrive to Restricted Packages? "Void offers some packages that are officially maintained, but not distributed. These packages are marked as restricted and must be built from their void-packages template locally." https://docs.voidlinux.org/xbps/repositories/restricted.html |
Beta Was this translation helpful? Give feedback.
-
Isn't it the maintainers job to maintain it? I mean, say John submits a package xyz. That means only staff members can maintain it? |
Beta Was this translation helpful? Give feedback.
-
Once the maintainer inevitably vanishes, someone has to take the slack. Which is barely even possible currently with the existing catalog of packages. |
Beta Was this translation helpful? Give feedback.
-
You made that branch name before I was even involved, so no I don't think it can be taken as a personal attack, nothing about it targets me as a person. That would make no sense. But this comment can be taken as that, I am not your "bud" and I don't appreciate you trying to deflect this point and instead resorting to insulting me to not address the last remaining "issue". Wilkart has already adressed this, For example, if you did the same thing for Gimp and called the branch "FuckGimp", I would question your actions. The Void members (especially ahesford, thank you) have done a good job explaning what happended, everything they have said makes sense, even if its not what I would of liked #46087 (comment) it was all in good faith. You are correct there is no "no point arguing", there is no way to make this look good. (edit: I mixed up usernames, I meant ahesford not Anachron) |
Beta Was this translation helpful? Give feedback.
-
The fact that you're still hung up on a branch name, still think it relevant to ping me in every single message you send even when I'm already in the thread, and your defensive tone very much proves my point. Blocking people ends receiving notifications from them regardless of what they do, so no need to add pings to your messages now. Also, Anachron is not a void member. |
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
|
Let me give my I think what is good and important is that Maintainers have the ability to make big decisions and let the distro move forwards. What I don't approve is decisions that are rushed. My suggestion would be:
I don't think this topic should've been exploded the way it has, but that's not something I want to talk about. |
Beta Was this translation helpful? Give feedback.
-
|
Several of you seem confused about some basic facts:
Such is the nature of Void. Members are given broad powers and wide berth to contribute as they deem appropriate, subject to the expectation that they ensure that their contributions continue to be well maintained and functional. We do our best to provide sound maintenance of the distribution. While we do think about the experience of end-users, this is always counterbalanced by the time and effort it requires to carry packages indefinitely. After all, this distribution is built atop volunteer effort and nothing more. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you, this is a well formulated comment that clearly communicates the situation and addressees facts that are not instantly obvious to outsiders. While point 5 makes sense, and I believe what your saying is in good faith, I think that should be changed in the future, As demonstrated here this can lead very bad optics. The merging creates an impression that the maintainers agree with the PR. Most of this was unintentional and I now understand the more anarchic nature of Void, but the eyes of the wider public judge by first impression and not from the view point of the people involved. The home page invites people to create a community around Void, it has links to Twitter, IRC & Reddit and is written in a way that makes it look like that attention is wanted. Public relations matter if your going to invite the public in. But that is just my thoughts on it. My opinions on point 5 do not matter, It's perfectly fine if nothing is changed, I am not a member of Void. |
Beta Was this translation helpful? Give feedback.
-
|
Sad to see that one of my favorite distros became unusable with this joke of a PR #44422. Been a Void user for years who depended on the Monero related packages, uninstalled today after realizing everything Monero related is gone after an update. |
Beta Was this translation helpful? Give feedback.
-
|
I can read that you're frustated, but please bring something useful to the discussion. |
Beta Was this translation helpful? Give feedback.
-
|
Just tried to install Monero and it didn't work. I don't like Linux distros telling me how to use my computer. That's exactly why I use free software. Uninstalling void and moving to gentoo. |
Beta Was this translation helpful? Give feedback.
-
I am not tying to spam and I am not reaching out in bad faith, the text below is from a comment I made on a pull request, but I have a feeling no one is going to read a post on a PR that was merged months ago.
I have lost all trust in the current maintainers of this project. I still like this distribution and really wish I could continue using it without worrying. I want things to improve. The usual disgruntled response is "Yeah, well, I'm gonna go build my own distribution. With blackjack and hookers!", but I don't have the desire, skill or time to maintain a gnu+linux distribution or a repo. Instead I am making my voice known and maybe something positive can happen.
I believe the main issue is one of communication. I think solving communication between users and maintainers would do void a lot of good. I would like to get an opinion from the maintainers on this.
Below this line, is the original post, it describes my frustrations. its from the 44422 PR https://web.archive.org/web/20230903050120/https://github.com/void-linux/void-packages/pull/44422
Every encounter I have had with the maintainers has been a negative one for the last few years. Look I understand opensource work pays nothing and is often a thankless job. But I have interacted with maintainers for other project and it never was this bad.
Consistently moves are made with zero attempt to communicate them before they are done. This is not a one-off, this is a pattern, I will list the last 3 times that upsetting actions have been done before communicating them from my memory.
I do not appreciate the maintainers making life style choices for me, Whats next? Are you going to remove TOR, that allows you to access and do bad things. Maybe void should remove everything with encryption, it's what bad people use to hide things.
I use to run xbps-install -Suy weekly, keeping my system up to date, Now hesitate to update, every time I do so there is a chance something annoying will be done by the maintainers and that I should go double check.
Beta Was this translation helpful? Give feedback.
All reactions