Why are you abandoning everything?

Because I am fed up with Solus and needing to beg for attention. Back when I joined I loved Solus for its ideas and spirit, but since then it became harder and harder to actually change something. Just to name a few points:

• A few weeks back I asked Josh if we could include screenshots in translations and he wanted to give me the rights to upload them. Did he? No. Did I get any response at all afterwards? Also no. Apart from that some Swiss guy decided to cut all ß from the German translation ?
• Back on 24. October Josh discussed new messaging platforms in a blog post. 5 months later we're still talking criss-cross in IRC, missing messages and what not.
• Enabling WPA3 in wpa_supplicant because it's a cool feature to have, isn't it? Apparently not, no response for 1,5 months.
• Updating openjdk because security and stuff. Or maybe actually think about an upgrade plan for JDK 11? Naaa, we don't need that.
• Pushing an iso-codes update because in a year many things can change. Too much work I guess.
• Finally include fwupd to give people firmware updates. Again security. But no.

Solus is still an amazing project built by many volunteers, many of which aren't or weren't experts when they started. But damn, the communication is garbage. See my PXE enablement task because we were doing this stuff at work. I wanted to do it, contribute to the project. I didn't say "hey core team, do this for me!". Even the commits were already finished. Josh's response? "No because f you". No explanation, no reasons, we just don't have any plans right now. I talked about these problems before, but there seems to be no ambition to conquer these.
IMHO the main issue is missing organization. You guys are working on the important stuff like Budgie and other things, I get that. Also the roadmap you presented is cool, although it doesn't involve the community, it's just what the core team is doing. Maybe this will change as soon as there is a new comms platform, where people can create channels and teams like virtualization, desktop environments, network and start working on them instead of updating everything they can find. For me the weekly task list was a really cool idea, but that stopped back in September. I'm also pretty sure that the ongoing popularity of Solus is too much too handle for you few guys. 400 open tickets and counting, 3 unbreak now tickets from last year. You probably have a different view on these points, but that won't change mine. I have a feeling that we are working next to each other, not with each other.

TL;DR: Will start building packages for my own devices on my server or use another distro - maintainers for docker stack, php, dbeaver, Hashicorp stuff, kubectl, seafile and remmina wanted.

In D5809#92470, @der_eismann wrote:

Because I am fed up with Solus and needing to beg for attention. Back when I joined I loved Solus for its ideas and spirit, but since then it became harder and harder to actually change something. Just to name a few points:

• A few weeks back I asked Josh if we could include screenshots in translations and he wanted to give me the rights to upload them. Did he? No. Did I get any response at all afterwards? Also no. Apart from that some Swiss guy decided to cut all ß from the German translation ?
• Back on 24. October Josh discussed new messaging platforms in a blog post. 5 months later we're still talking criss-cross in IRC, missing messages and what not.
• Enabling WPA3 in wpa_supplicant because it's a cool feature to have, isn't it? Apparently not, no response for 1,5 months.
• Updating openjdk because security and stuff. Or maybe actually think about an upgrade plan for JDK 11? Naaa, we don't need that.
• Pushing an iso-codes update because in a year many things can change. Too much work I guess.
• Finally include fwupd to give people firmware updates. Again security. But no.

I can answer some of the points.

• We had a very busy time preparing the release of Solus 4 and now Josh is on holiday (and even then he still find time to handle some of the most urgent things). Regarding weblate he posted this
• As far as I know, there hasn't been any commitment regarding a new messaging platform - I don't see any in the blog post - It's still an open point. And the committed points are being delivered (Weblate is there, Flarum with 0Auth module is there, a few things have been transfered from GH to phab, ...)
• WPA3 is certainly a cool feature and will have to come, but today how many people have WP3 compliant equiment (if there are nobody mentioned it yet) so what's the problem if it sits in the backlog ?
• For JDK 11, I think it's way more complex than just updating a package and your diff doesn't even link to T6658. Your test plan just mention minecraft. How could you seriously expect this to be landed like this ?
• fwupd is currently on hold because pre-requisites are not met (T2532). So what do you expect by submitting patches for tasks that haven't been approved ?

You're not the only one who has patches waiting in the queue sometimes for a long time. I regularely go a IRC few days before a Sync to ask to review a submitted patch that is important to me or to ask a status of a patch that remains in the queue.
We all would like to make more things to get them faster but there are only volunteers humans here who all have a life next to Solus. And landing things in a hurry without reviewing them correctly some stuffs are more critical than others. If JDK 11 was landed, manything would for sure be broken now. fwupd wouldn't work. Are you 100% sure that the WPA3 update would not cause any trouble to exsiting users? I was tested just on 1 host... so perhaps this will be tested deeper by core team before it gets landed... I would be sad if it would break things for people to add a feature than nobody (or very few people) really needs yet.

In D5809#92470, @der_eismann wrote:

Because I am fed up with Solus and needing to beg for attention. Back when I joined I loved Solus for its ideas and spirit, but since then it became harder and harder to actually change something. Just to name a few points:

• A few weeks back I asked Josh if we could include screenshots in translations and he wanted to give me the rights to upload them. Did he? No. Did I get any response at all afterwards? Also no. Apart from that some Swiss guy decided to cut all ß from the German translation ?
• Back on 24. October Josh discussed new messaging platforms in a blog post. 5 months later we're still talking criss-cross in IRC, missing messages and what not.
• Enabling WPA3 in wpa_supplicant because it's a cool feature to have, isn't it? Apparently not, no response for 1,5 months.
• Updating openjdk because security and stuff. Or maybe actually think about an upgrade plan for JDK 11? Naaa, we don't need that.
• Pushing an iso-codes update because in a year many things can change. Too much work I guess.
• Finally include fwupd to give people firmware updates. Again security. But no.

Solus is still an amazing project built by many volunteers, many of which aren't or weren't experts when they started. But damn, the communication is garbage. See my PXE enablement task because we were doing this stuff at work. I wanted to do it, contribute to the project. I didn't say "hey core team, do this for me!". Even the commits were already finished. Josh's response? "No because f you". No explanation, no reasons, we just don't have any plans right now. I talked about these problems before, but there seems to be no ambition to conquer these.
IMHO the main issue is missing organization. You guys are working on the important stuff like Budgie and other things, I get that. Also the roadmap you presented is cool, although it doesn't involve the community, it's just what the core team is doing. Maybe this will change as soon as there is a new comms platform, where people can create channels and teams like virtualization, desktop environments, network and start working on them instead of updating everything they can find. For me the weekly task list was a really cool idea, but that stopped back in September. I'm also pretty sure that the ongoing popularity of Solus is too much too handle for you few guys. 400 open tickets and counting, 3 unbreak now tickets from last year. You probably have a different view on these points, but that won't change mine. I have a feeling that we are working next to each other, not with each other.

TL;DR: Will start building packages for my own devices on my server or use another distro - maintainers for docker stack, php, dbeaver, Hashicorp stuff, kubectl, seafile and remmina wanted.

I get your frustrations. We see them and aren't deaf to them. There are limits to what we are capable of doing right now, and unfortunately for you, most of that stuff has to wait for the Core Team to bring things up to speed. I'm not going to try to convince you to stay or of how best to spend your time. That's your decision and I will respect it.

I'll quickly respond to your points and then I'd like to share my view of the situation so that you have a better idea of where I'm coming from:

1. 400 open tasks is small potatoes. We have done a tremendous job in reducing open tasks over the last year and the numbers speak for themselves: https://dev.getsol.us/maniphest/report/burn/ . Of the remaining over tasks, a significant number of them are pending on the next versions of the Software Center and Installer which will happen in due time.
2. We are still looking for an actual usable alternative to IRC for the kind of stuff we need to communicate for development and as Josh has pointed out several times now, nothing ticks enough of the boxes to meet our needs. I get that IRC is frustrating to you, but we also have Phabricator for development discussions and people aren't taking advantage of that. I don't mind having tasks open to try and figure out the best way of working on different projects and for coordinating efforts. The weekly Task Lists stopped because it has gotten to the point where Josh and I are too busy outside of Solus to maintain a meaningful list for each week. No one ever said that maintainers couldn't keep doing that on their own.
3. PXE booting has been off the table for a long time. I get your excitement and wanting to contribute what you know, but it's just not something we think belongs in a Home computing environment. Solus isn't meant to be used as an enterprise OS and PXE is an enterprise feature, whether you agree with that sentiment or not. It's also hardly the first time it was brought up on the bug tracker. And you're also forgetting the fact that Solus isn't a democracy. We have to say no to some things in the interest of not dividing our focus.
4. IDK what's going on the translation stuff because I don't have time to even look at it.
5. OpenJDK is a shit-show and no one has been able to tell me how they propose that we support multiple Javas out-of-the-box apart from awful shell scripts that don't solve the problem. A ton of our software only works on OpenJDK 8 and we have to stick with that until the situation changes. Give me a solution and we'll talk.
6. I've intentionally held off on fwupd because it has as much potential to do harm as it does to fix things. I need a concrete plan for integrating it and no one has given me one beyond putting the package in the repo.
7. The rest of your patches have been waiting because Josh and I are exceptionally busy and had to focus on getting the milestones for Solus 4 finished. I fully intend on going through the patch backlog this weekend, but you abandoning your patches means I now have to waste time going through the backlog to find them again so that I can land them. These "rage quit" moments of yours aren't helping your case.

Now that most of your concerns are at least partly addressed, I implore you to listen to why this is all taking so long.

Personally:

People forget sometimes that I'm not paid to do this. I'm still in school working hard towards earning a PhD. I get paid a stipend for living expenses. This takes an inordinate amount of time and has gotten steadily worse over the last two years. I typically work 60-80 hours a week on average. Couple that with Q1/Q2 being the busiest time of year for me and of course things are going to take longer. I also had an unfortunate accident in February that left we with a pretty badly sprained ankle which has made everything more difficult, adding an hour to my daily commute and making it extremely difficult to sit upright in a chair for long periods of time.

Professionally:

You say that our roadmap has nothing to do with community engagement and that couldn't be farther from the truth. Nearly everything we have on the docket for this year is a roadblock for contributing to Solus. We needed Weblate to even start handling translations properly and we are still learning the best ways to use it and integrate it into our projects. GNOME 3.32 is a huge blocker for updates and will take some time to build and test before it hits the repo. ypkg needs significant upgrades to help speed up patch reviews (linting especially) and deal with long-standing issues that lead to lots of time on workarounds for things that ypkg should already handle. Not to mention new versions of yauto and yupdate that don't mangle package.yml files. eopkg needs replacing because it's handing of things like file conflicts and circular dependencies have been a huge stumbling block for the last year. Just ask Girt how much time he's wasted fighting with file conflicts this year for Plasma. sold is necessary so that the Software Center can finally have proper integration with the package manager and so that we can get rid of our vulnerability to Python breakage. The new Software Center will fix way more of those open Phab tasks than you think. And then there's all of the problems we've had with ferryd in the past two months. You may not realize this, but I have been replacing large chunks of ferryd with every spare moment of coding time I have. It's become a huge bottleneck for weekly syncs and is at a point now that it is actually slowing down our ability to land patches quickly. And then there's the ongoing issues with the network for the package mirror that I've also been working to resolve.

Every spare moment of mine is going towards keeping the critical items up to date and getting all of this behind the scenes stuff working so that you can get things done faster and so that the Core Team isn't your road block. We are in a period of tremendous renewal of core Solus codebases that are essential to smooth daily operation. This stuff takes time and can't be rushed. Patience is a virtue.

Dictatorally:

The Core Team is responsible for what gets added to Solus and deciding what we are willing to support. As a maintainer, you have some control over the things that you are responsible for, but ultimately we get the final say. You do not get to decide that you are going to add something completely new to Solus without us signing off on it. If we say "no", we expect you to respect our wishes and not pursue it further. If our explanation is lacking, ask for more information. We are generally very reasonable people and more often than not a "no" is more of a "not right now" than a "never".

In D5809#92470, @der_eismann wrote:

Because I am fed up with Solus and needing to beg for attention. Back when I joined I loved Solus for its ideas and spirit, but since then it became harder and harder to actually change something. Just to name a few points:

• A few weeks back I asked Josh if we could include screenshots in translations and he wanted to give me the rights to upload them. Did he? No.

I was preparing for the release of Solus 4 as well as doing my own work outside of Solus, which will always take precedent. Solus 4 was originally slated for January of 2017, back when Ikey was part of the project, but was pushed back for a Software Center that was never completed. Then, we had to sort out infrastructure issues and get back on our feet after losing access to a significant portion of the infrastructure. The only reason Solus 3.9999 ISO Refresh went out was because of the repo URL change, having an ISO where people couldn't even upgrade (as would be the case for Solus 3) wasn't acceptable. So to say I was pre-occupied with the importance of ensuring a good Solus 4 release would be an understatement. I did not forget about including screenshots in translations, it was not a priority. Prioritization is something which occurs no matter the company, organization, or project. It can be a single-individual open source project and that individual would still prioritize items on importance. Screenshots in translations were not a blocker for the release of Solus 4 and I did not deem it a priority to resolve it while on vacation.

Did I get any response at all afterwards? Also no.

See above explanation. Surely you knew we released Solus 4, some understanding about prioritization would've been appreciated. Furthermore, you never asked me personally for a follow-up on it either.

Apart from that some Swiss guy decided to cut all ß from the German translation ?

If you have an issue with a specific translator, you should bring it up to them.

• Back on 24. October Josh discussed new messaging platforms in a blog post. 5 months later we're still talking criss-cross in IRC, missing messages and what not.

I indicated in the Improving Community Engagement blog post that there was multiple criteria which needed to be addressed. I then followed-up in Shiny Delights with a comprehensive list of messaging platforms and my evaluation of them. The reason we are still on IRC is because issues I highlighted have not been resolved yet. For Riot, it was laid out in:

• "Riot does not support voice channels. This is a known issue that I am subscribed to, so I will keep informed about any resolution to it. Additionally it doesn’t support push-to-talk in conferences. The latter is not a requirement however, just something I noticed in my research."

This has not yet been resolved. The task is still in progress.

• "There are still various issues around message deletion and roles that should ideally be resolved though, such as sending server ACLs and support for redacting multiple messages."

Neither of those two mentioned issues have been resolved either.

• Enabling WPA3 in wpa_supplicant because it's a cool feature to have, isn't it? Apparently not, no response for 1,5 months.

I have stated on multiple instances to you that such patches, such as to network-manager as well, are intentionally held off until the GNOME Stack upgrade. I'm not really sure why you expected a different response again, that wouldn't be consistent.

• Updating openjdk because security and stuff. Or maybe actually think about an upgrade plan for JDK 11? Naaa, we don't need that.

@kyrios123 already laid out why this would be an issue. So has @DataDrake in the past. You already know that.

• Finally include fwupd to give people firmware updates. Again security. But no.

We have stated on multiple occasions we have no desire for fwupd support until it is integrated into the Software Center and items around clr-boot-manager have been validated to be resolved. Any miniscule amount of research on our tracker into the matter would show our consistent answers on the matter. You can't really fault us for providing a patch for an item which wasn't going to be accepted in the first place. It'd be like providing a patch for a package request that was marked as WONTFIX, rejected for inclusion.

Solus is still an amazing project built by many volunteers, many of which aren't or weren't experts when they started. But damn, the communication is garbage. See my PXE enablement task because we were doing this stuff at work. I wanted to do it, contribute to the project. I didn't say "hey core team, do this for me!". Even the commits were already finished. Josh's response? "No because f you". No explanation, no reasons, we just don't have any plans right now. I talked about these problems before, but there seems to be no ambition to conquer these.

Your remarks in your task were:

• "Ideally you could name more maintainers that can commit themselves, but that needs a lot of trust (and knowledge to fix broken updates)."
  • This is something we have done.
• "I was thinking about some categories like "just a minor patch, please commit asap" (e.g. PHP 7.2.x), "big update, maybe with breaking changes, request for comments and tests" (e.g. PHP 7.1 -> PHP 7.2) and "initial inclusion, please check for mistakes"
  • We have had the "Requires Rebuilds" tag to indicate when large ABI changes warrant rebuilds since Mar 24th 2017. This was already addressed by the time you provided that item.
• "An alternative could be a current roadmap to see what's being worked on and what's important in the near future."
  • Which we have. But apparently resolving the issue doesn't actually mean resolving the issue? Because according to you "there seems to be no ambition to conquer these"
• "Haskell stack and kernel upgrade? Okay, I won't bother you. Software xyz needs an update? I'll be happy to help. Software xyz is already maintained by Person xyz? Sorry, that wasn't noted anywhere and I just wanted to help."
  • Part of our work on ypkg 3 is to lay the groundwork for Solhub and be able to start enforcing MAINTAINERS file to make it clearer what is in the domain of a specific person. But to do that, we need ypkg 3, and as you can see from our Roadmap (which you've acknowledged exists), ypkg 3 does not exist yet.

As stated by @kyrios123 in his remarks to your task: "It is not really a problem to have to wait for a patch to be committed. I think IRC is a mandatory place for packagers."

Reality is you rarely engage with us on IRC. You upload patches every Thursday or Friday (which I'm happy for, to be clear) but do not communicate with us any intentions around those. We can't engage with you if you don't engage with us. It's not a one-way street.

IMHO the main issue is missing organization. You guys are working on the important stuff like Budgie and other things, I get that. Also the roadmap you presented is cool, although it doesn't involve the community, it's just what the core team is doing.

The community has gotten involved in the development of Budgie and helped reduce the amount of milestone issues that needed to be addressed for Budgie 10.5. Translations were done entirely by the community. Maintaining of packages reduces Core Team workload so we can focus on other items, or you know, take a vacation for once in two years.

I'm also pretty sure that the ongoing popularity of Solus is too much too handle for you few guys. 400 open tickets and counting, 3 unbreak now tickets from last year.

And yet that number is 1/3 what it was less than a year ago. You'd see that by looking at our burnup rate. We went through and resolved a huge amount of issues, closed ones which had long been addressed by not triaged, etc. It's been holding steady around 400 for months now, because we've been mostly keeping up with tasks.

TL;DR: Will start building packages for my own devices on my server or use another distro - maintainers for docker stack, php, dbeaver, Hashicorp stuff, kubectl, seafile and remmina wanted.

I have no issues taking over Docker. I was the original maintainer of it =)