Recently I received this from @KapiX:
I don’t want be here anymore if I am considered “problematic”. I see no delete account option, so I will not log in.
I think this is very unfortunate. Your progress/status reports have been a highlight.
I hope you can find a way to stay motivated to make your awesome contributions to the code and I hope to see you back here again, in time.
I don’t think a warning is needed, it only adds fuel to the fire. That said, a slightly different tone in said discussion would’ve made it a discussion instead of a fight. I personally prefer a more direct approach, but many people don’t. Being offended is something you do to yourself, it’s not entirely the responsibility of the messenger. I suspect language barriers and cultural differences might not have helped in this case.
I get why it might be seen as such, but I don’t interpret the discussion started by X.512 as a personal attack. (neither is giving a warning, which is just a moderation tool) It’s a shame things have to result in quarrel.
14 Likes
I also loved X512’s progress reports and his work on numerous parts: RISCV64 port, GPU acceleration, promises,…
This was really motivating.
I thought this problem had been sorted out and was surprised to hear about it now.
14 Likes
I hope that he continues to contribute code.
The progress he made is amazing.
I think the stuff X512 made and Waddlesplash working in full-time are the two major reasons for Haiku now being more active than ever before.
Without X512 the development would likely slow down again what would be very sad.
12 Likes
This is exactly what I have thought for some time now.
6 Likes
Proposing direct messages when it’s crystal clear that both parties cannot find a common ground (and one side seems to be in position of power) is a really bad idea.
From the outside, it looks very much like trying to hide “bad things” under the rug, caring more about the image of the project than the project itself. And that is what puts me off (people fall in conflicts all the time, everyone makes mistakes, why would anyone be put off by that?).
Anyway, i really liked to read X512’s progress reports on all sorts of developments. It would be a shame to lose that.
8 Likes
Hi,
We received multiple independant complaints about the bad mood set on the forums by these posts from different users and developers, publicly and privately. We have discussed this with the moderation team and decided to tell X512 why his behavior is annoying, leading several people to stop contributing to the project. At no point in this discussion we considered banning him or other further actions, as we expect he can understand how his behavior is problematic and driving other people away, and he can understand why it’s important to follow the existing rules and ways of doing things in the Haiku project.
I don’t think there is anything to hide under the rug here. Several people reported a problem, we have chosen the response that seems to be appropriate (discuss the issue with the person who was reported as causing the trouble). We contacted X512 in private first so he had an opportunity to reply to this.
For completeness, here is the response we received from X512 (since apparently, making this discussion public is ok):
Lets ban me just now.
I don’t think this is a great way to show he is open to discussion to improve things.
7 Likes
From the perspective of a contributor that took some distance after being put off by everything that happened, I contribute to Haiku because it’s a project where I can just work on code and enhance my skills in a friendly environment among people that share similar interests and share a similar goal, which is why I can corroborate @PulkoMandy’s claim and publicly state that I did take some distance from the project as a result of my frustration from the drama in combination with other factors IRL.
All of this is/was avoidable from my perspective, and I’d also like to believe that the way some of the issues were a result of misunderstandings. However, these “meltdowns” (for the lack of a better term) in public are not only toxic for the project, but it is absolutely self-destructive for you (@X512) on an individual level. I don’t believe (and I’m pretty sure that many others can also corroborate) that you should be banned or stop contributing, but calm down, a warning means “cut it off” and not that you are an undesirable person and that we all hate you and secretly wish that you would jump off a cliff, or something like that. Escalating this further makes things worse for both yourself and the rest of us emotionally. Your frankly amazing work and you, as a person/contributor, are desired. It’s just that these sorts of vibes that you are spreading around unfortunately undo all of that in, again, a self-destructive manner and that is, again, absolutely avoidable.
I am not sure if this is the result of cultural differences or if you understand why some people got hurt, but as a person that made many of these mistakes like these as I begun to get affiliated with a bunch of collaborative efforts and communities (local, international and Western-centric) when I was a teenager and without having any sort of authority here and ordering you to do anything, I’d like to give some (unsolicited but hopefully helpful) advice: I would seriously suggest that you should ask any of us what went wrong and issue apologies where they are due. Things don’t have to suck so much, but making up for it is possible, although absolutely up to you. Social stuff is hard, so, I’m making myself available if you need any help with trying to set things right. Take care. ^-^
12 Likes
I agree that the response you quoted was bad. However, he was advised to direct message the other side, in a situation when it was clear they cannot agree at all, and that the other side has a lot more power. It’s like saying: “we don’t care, stop complaining, sort it out”, knowing well that, in practice, he has to give up the case or go away.
Unless he was just trolling (i find it unlikely, given how much work he did for the project), it looks like “going public” and kinda all-dramatic was the only option he had left aside from just quitting.
Maybe the next person to hire should be HR-type of person to smooth things out between strongly opinionated developers :D?
What makes you think that?
There was discussion over this particular change on the mailing list with other developers also stating they disagree. It went without problems. We now all agree that this change is ok for now and it will be fixed later. This did not require any drama or influence fights. Just technical discussion between people behaving like adults and understanding each others’ point of view.
If people cannot behave like this, and instead when there is a technical disagreement they start to roll on the floor like a 5 year old child and threaten to quit, what do you think the HR person in any company would do?
If you want the discussion that happened in the right places, it is on the review for this change:
https://review.haiku-os.org/c/haiku/+/4506 and as a result of this discussion, a ticket was created to track the problem and continue working on it later: #17360 (launch_daemon & BRoster: environment management problems) – Haiku
As you can see, everyone else somehow manage to discuss things and reach a common agreement on what are the next step. The issue is indeed somved as far as the development team is concerned. But the drama on the forum continues.
4 Likes
This is what I never wanted to see here, ever.
You missed a supercar to please so many the bad mood of so many jealous small cars.
Having said that, I recommend a good movie.
The imitation game.
Yes that’s right, without the difficult character, but brilliant pioneer Alan Turing we wouldn’t be here, in metaphor this is a similar situation.
I for one can see the problem from @X512s position: Previously some lengths were taken to justify blocking a merge of his code, when frankly the code in question affected very little, and much of the argumentation behind the block was from someone who is not a project member. But later a much wider reaching change was pushed through in spite of his and others objections. So it may appear that when it comes to merging code there is one rule for @X512 and another rule for some others.
I can only agree that bringing these disputes onto the forum is not appropriate, but wonder if @X512 feels he has no other route to resolution, that his concerns have previously been dismissed or discounted elsewhere, and I wonder if hearing and addressing his concerns may help prevent this occurring again. He is likely now thinking that not only is his code unjustifiably blocked but he is also being pushed back against in the forum where he seeks support.
To be fair I do not know all the details about all of this (nor do I really want to), I am not really sure who is involved exactly and don’t judge anyone to be at fault, this is just an observation about the situation. I’m sure this is at its heart a misunderstanding that may be related in some aspects to differing culture, and hope it can be resolved (privately might be better?) and that everyone can move on.
5 Likes
Quote from the mail he received (pointing to direct messaging), it’s not the first time there is an issue between them, and only one of them being part of the “core” team (and the only paid developer on the project).
I’ve read comments on that patch review before. It does not look good to me, when author gives himself +2 (and merges the patch?), there is no other vote-for, no clear agreement, and there are comments against the change ;), but i’m not a developer on the project - maybe that’s the way things are done here.
To me it looks like the mail was the trigger for today’s drama, after the earlier issues, not the last issue itself. Almost like adding gasoline to the dying-out fire :).
I agree that “blackmailing” the team with a forum drama is not a good solution, but i also think that things could have been improved on all sides. But that’s just my observation from the outside - i may be missing a lot of information (maybe that’s why it might be better to keep things in the open, so people don’t have to just guess? Maybe set up some clear rules for resolving conflicts, at least with 3rd person to cool everyone’s heads?).
1 Like
If said supercar started crashing into smaller cars, it would be damaging them and the road they all drive on. This is what metephorically happened here. Also btw, your analogy is quite dismissive of the small contributions that help keep the project going. Open-source is dependent on contributions big and small alike. Most (if not all) projects won’t last long if there are only a few so-called rockstar developers working on them.
Yes that’s right, without the difficult character, but brilliant pioneer Alan Turing we wouldn’t be here, in metaphor this is a similar situation.
On the other hand, one does not need to be a difficult character to be influential or a genius. Dennis Ritchie is often fondly remembered by his colleagues and his contributions (Unix and C) are much more directly related to Haiku’s existence and most of the modern computing world wouldn’t be here without him.
3 Likes
So we should just do nothing? We did that the first time, it didn’t work.
It took so long because there was a discussion about this.
There is a rule: use haiku-development mailing lists. X512 refuses to do that because it’s inconvenient for him.
1 Like
I think you are reffering to https://review.haiku-os.org/c/haiku/+/4311. If you look closely, you will see that this change was, in fact, merged. There was some discussion to make sure we understand what the change is doing and how it is used (we have all agreed that booting without u-boot would not be officially supported by Haiku). Once everyone agreed, the change was merged. This is what happens when people have a productive discussion, they reach an agreement and things move on.
There is one rule for people who have commit access to the repository (they can do whatever they want) and one rule for people who don’t (they need someone else to +2 and submit their changes). We could make it so no one is allowed to push changes directly to the repository, but it would slow down development even more than it is now. We could allow any random developer to push changes, but it would be a mess. So I don’t really see how we can do otherwise.
The other way to fix this would be if X512 was allowed direct commit access. This was discussed by the development team but we decided to not go with it for two main reasons:
- Behavioral problems which culminated (hopefully, this will not be going further) in this forum topic,
- A lot of the submitted patches did not follow our coding guidelines. In the case of the RISC-V port, kallisti5 has spent a lot of time reworking the patches so we could merge them while keeping our quality standards. This is boring and thankless work that largely goes unnoticed. I am fine with people not willing to do it, but then the downside is they don’t get commit access.
That is exactly what we put in the message, encourage to discuss things privately. And the good news is, we are now discussing this privately with X512.
Yes, that’s how things are currently done. I don’t agree with it (and I voiced my concerns about it with the other members of the developer team several times), but it’s the current state of things. It’s a leftover from when we used to have no code review system at all and people would push their code directly to the main branch. And it’s also a reasonable compromise because we don’t have enough time to review and merge all patches, it would be worse if we couldn’t take shortcuts sometimes.
9 Likes
Most of the movie is historically inaccurate, BTW. :D
I hope the discussion ends with that he will get a thread for posting progress, its one of the main reasons why I visiting the forums. I can understand that the mailing list were the way of communication back in 2001, but 2021? I agree that its very unconvenient to use mail when we have a forum where eveyone can join in easily, IMO.
4 Likes
He doesn’t need to get it, because he was not banned. It’s his choice that he is walking away.
They still are a way of communication in many open-source projects. I didn’t know how to use them before I joined the project, but decided to learn and it really is not as much hassle as people make it out to be. It’s actually easier, because I do not need to create an account, I already have one - my e-mail.
1 Like
No one complained about the posts for posting progress on various things. The initial message was about a very specific incident where the discussion should have happened on the code review tool and possibly on the mailing lists. You can easily find that specific thread as it is linked in the first post of this one. I suggest you read it and see if that particular thread was a valuable contribution or not.