Welcome, a heap of gratitude and update!

[elbenfreund]

To everyone involved, be it on the sidelines or right down in the thick of it for the past weeks:
Thank you all for your continued interest and investments.

After two years of work related neglect, I am starting to slowly claw my way back towards hamster.
As this is just the start, I want to focus on one big message right here:

THANK YOU ALL I understand there has been friction among some of our recent contributors and I would like to offer the following solution: Just dump your frustrations on me. Whilst there clearly are differences in approach the sad truth is the main reason things escalated, more than they probably needed to, is because I failed to provide the required guidance and structure.

If you can find it in your heart, I would like to invite all of you, @ederag , @jtojnar , @hedayat @DirkHoffmann , @mwilck and everyone I failed to mention here (I am still catching up, so pretty please do mind me not listing you here...) to work with me towards turning hamster into a well maintained and productive tool. Lets try to forget about the frictions of the past and try to establish a productive and friendly future community spirit!

Immediate actions taken

For now, in order to unblock some of the outstanding contributions so far, I have done the following:

1. Invited @ederag to the main 'legacy-hamster' team, granting full (provisional) write permissions.
2. Invited @hedayat , @mwilck and @DirkHoffmann to join the hamster-shell-extenion team, granting (provisional) write access to all of its members.
3. I made sure master and develop branches in hamster and hamster-shell-extension repositories are "protected", so we keep merges clean and tidy using PRs.

Near future topics

Whilst the recent developments have succeeded in getting my head out of my ass, I would be lying if I would claim that I do have the resources to start proper full time work on hamster in the immediate future. My current work and private live does not permit that. What I can and will do is commit 4-6 hours of "catching up and maintenance" contributions for now.
I hope that helps elevating some of the current social friction and provides a overarching perspective into the future.

• I will post a much needed update on the current situation on projecthamster.org.
• I will upload any build artifacts we generate to extentions.gnome.org. I will also investigate if we can streamline this process.

Questions

That being said, I can not address all outstanding issues right away, so I will be needing your help :)

• Where are we on your preferred communication channel? Do we want to keep the ML or focus on Issues? Or "github team discussions"?
• In order to deal with low hanging fruits and getting things done, I would like to focus on those development efforts / PRs that deal with uncontroversial maintenance work. I would kindly ask of you to give me a bit more time to read up on any actual design decisions that are currently disputed (links welcome) before we merge. Ultimately, some of the (currently discussed) problems stem from legacy-hamster's monolithic design, which makes it sometimes hard to accomodate different user expectations. Overcoming those limitations is one of the main goals of the (now stalled) rewrite.
• What are your most urgent pain points right now?

[ederag]

@elbenfreund Glad to see you back !
Thanks for the invitation, although that did not grant any further rights AFAICS.

Friction on this repo is very localized IMO, and had been sorted out, from my point of view.
Friction on the whole project is mainly due to the fact that I did not have the necessary rights.

From the mails that I have seen recently,
IMO your comprehension of the current status is tainted by @mwilck's known concerns
and by @DirkHoffmann subterranean fighty messages.
It's a bit sad that you seem to have mainly listened to them up to now.
But that's it, some people make a lot of noise,
and can easily give the impression that everything is going to pieces, including this repo.

To catch up, I'd advise to read #433, #493 before going any further.
I do hope you'll revise your judgment and we'll be able to work hand in hand eventually.

---

Edit here, since I have been denied the right to conclude

It's a long noisy thread. Here are important posts IMO:

• Whether the project was well maintained or not: remove ederag from maintainers #575 (comment), Welcome, a heap of gratitude and update! #574 (comment) and Welcome, a heap of gratitude and update! #574 (comment)
• Context: Welcome, a heap of gratitude and update! #574 (comment)
• Disrespect: Welcome, a heap of gratitude and update! #574 (comment)
• Why such a clash: Welcome, a heap of gratitude and update! #574 (comment)
• Progressive withdrawal: remove ederag from maintainers #575 (comment)
• Innocents ? Better document project state #611 (comment)

[elbenfreund]

I do hope you'll revise your judgment and we'll be able to work hand in hand eventually.

We will, absolutely. I hoped my original message made a big point in "not taking sides", as I fell the best way forward is for all of you to just blame me ;) and get on with it.

[DirkHoffmann]

Thanks, @elbenfreund. Good to hear and see that the boat still has a captain.
As requested, here is one hint to steer in the right direction:

Invited @hedayat , @mwilck and @DirkHoffmann to join the hamster-shell-extension team, granting (provisional) write access to all of its members.

Please add @benjaoming to make the Three Musketeers complete.

[ederag]

@elbenfreund Your message was very negative,
and you did that, not in the extension you are the sole owner of,
but here, in the repo I am responsible for.

That might be a mistake due to writing on sunday after a busy week,
and the last days have been draining,
so don't blame you too much 🙂

Besides, you did not take into account my explicit recommendation (by mail)
of giving access to @hedayat only, following the community consensus summarized here.
I expressed strong support in favor of @benjaoming as well (after taking community consensus),
and he's not in the extended list.
Edit: the last one was based on superficial knowledge.

That does not bode well, that's not how a team should start.
I hate forks, but that's it, life is too short, and I'm tired of fighting.
So I'll just continue in my fork, hoping you'll integrate as much as possible
in the main repo and in the rewrite.
Please do ask if certain design choices are unclear, or if I can help with technical issues.

---

Edit:

repositories are "protected", so we keep merges clean and tidy using PRs.

Wow, that means that I actually lost rights. That only confirms my impression.

Edit 2: (3 days later)
about the fork idea: #574 (comment)

Edit 3: Why the OP was "playing the savior" and negative, from the active maintainer perspective.

I failed to provide the required guidance and structure.

towards turning hamster into a well maintained and productive tool.

so we keep merges clean and tidy using PRs.

current social friction

[elbenfreund]

@elbenfreund Your message was very negative,

As I said via mail, I didnt mean to and quite frankly do not see where I was. I feel very positive we can overcome this misunderstanding :)

and you did that, not in the extension you are the sole owner of,
but here, in the repo I am responsible for.

I did that in the place that I thought would get the most attention in order to reach the biggest audience with my "news". To be honest, I would have preferred using the mailinglist (which is in fact intended for exactly this kind of stuff), but was unsure if everyone was reading there.

That might be a mistake due to writing on sunday after a busy week,
and the last days have been draining,
so don't blame you too much slightly_smiling_face

I suspect the last days were busy and straining for all of us, so I am confident all of this will appear less conflict prone once we get some rest and a fresh week :)

Besides, you did not take into account my explicit recommendation (by mail)
of giving access to @hedayat only, following the community consensus summarized here.
I expressed strong support in favor of @benjaoming as well (after taking community consensus),
and he's not in the extended list.

Yes, that is correct. Quite frankly: I am in no position to judge who is best suited to take the lead, and so far all members of the extension team seem to be on generally good terms as far as I can see in the issues/PRs I've read so far. The extension's scope is rather limited and I trust that all contributers involved will work together to move things forward.

I hate forks, but that's it, life is too short, and I'm tired of fighting.

I am very sad to hear that, and I honestly do not understand the strength of your reaction.
I would love to keep the lones of communication open and hope that some rest and distance will bring you back as the valued contributor that you are.

[ederag]

the valued contributor that you are

Contributor, but not maintainer, yeah, again you marked well your territory.

You have no legitimacy to do so because

1. You did well earlier, but then have left the project for years;
   I'm the one who brought it back to life recently,
   together with many people, notably @GeraldJansen
   (by the way, he's great, and should be invited to be part of the organization)
2. Most of your perception of my maintainer skills is based on unverified ill hear-say.
   You should have contacted me to have a full understanding before posting.

I'll leave you play the savior here.

To the many kind users who trusted me, most of the experience was a nice one,
and I liked the constructive discussions,
so it's not my intention to let you down, don't worry.
Edit: the "don't worry" link means there's a backup. A hard fork would just harm the project.

[GeraldJansen]

@elbenfreund Actions speak louder than words, and much louder than loud announcements and promises. How can you be trusted this time around given that as project "captain" you practically shipwrecked the project with the abandoned rewrite initiative?

On the other hand, @ederag has put in two years of solid actions and demonstrated his strong commitment to this project. He has been the de facto project leader during this this time, much more so than simple "contributor" or "maintainer". It's great to see him finally added to the list of project members. I do trust that he will be granted full control of the "legacy" Hamster sub-project as an encouragement to stay with the project.

[matthijskooijman]

As a user, occasional contributor and planning-to-be-Debian-maintainer for hamster, let me also chip in.

I'm a bit surprised by the intensity of @ederag's comments here. I did not read anything particularly negative about @ederag or anyone else in @elbenfreund's comments and posts. There are a few things that, now that they're pointed out, could be read like that (protecting branches to take away rights, not following some advice given, saying "contributor" rather than "maintainer"), but I also tend to believe @elbenfreund when they explicitly say things like "I didnt mean to and quite frankly do not see where I was. I feel very positive we can overcome this misunderstanding :)" (about the negativity of his post).

Admittedly, I have not read all previous discussions here (only recently subscribed to the repo as a whole) and of course have not seen any private e-mails, so maybe that would put things in another light. I did read some of the linked discussions just now, and it looks to me, @ederag, that you have a tendency to quickly read negativity and attacks in comments. Personally, I'm usually on the other side of that (I often assume good intentions even they are not there), but in my experience it seems better to not feel attacked when you are, than to fell attacked when you are not :-) And to be explicit: Please do not take this as an attack or judgement, but just as a (subjective) observation and suggestion.

All in all, @ederag, I really hope you will reconsider staying on-board with the project, rather than doing your own fork. If there is anything that needs to happen (e.g. in terms of repository access, or stated responsibilities, etc.) for that, please also make that explicit.

@GeraldJansen said:

@elbenfreund Actions speak louder than words, and much louder than loud announcements and promises. How can you be trusted this time around given that as project "captain" you practically shipwrecked the project with the abandoned rewrite initiative?

This is also a concern of mine. It seems that some of this is mitigated by sharing at least access to some repositories, and I think sharing access to things like the projecthamster.org and extensions.gnome.org might also be useful here (but maybe that's already the plan?).

However, I'm not so sure if @elbenfreund is intending to the "captain" that some people now name him. He writes to have limited time and wants to focus on making things move again, which for a most part, I think, is forming a team and making collaboration work even without his direct involvement (but I am filling in some blanks, so maybe @elbenfreund can confirm).

@elbenfreund said:

In order to deal with low hanging fruits and getting things done, I would like to focus on those development efforts / PRs that deal with uncontroversial maintenance work. I would kindly ask of you to give me a bit more time to read up on any actual design decisions that are currently disputed (links welcome) before we merge. Ultimately, some of the (currently discussed) problems stem from legacy-hamster's monolithic design, which makes it sometimes hard to accomodate different user expectations. Overcoming those limitations is one of the main goals of the (now stalled) rewrite.

I think that the most effort is needed in the shell extension department, where no development (at least no merges) have been happening for quite some time, and some forks have been uploaded to extensions.gnome.org that really need to be unified.

As for the main hamster repository (referred to as hamster-legacy, but given the state of the rewrite, I do not think the "legacy" label is applicable already), I am under the impression that that is actually in a good state, thanks to @ederag's efforts. There is probably a lot of development work that can happen to further improve things, but I think there's not much lacking in project maintainance or stalled pull requests (but I might be wrong - I have not been too closely involved).

Where are we on your preferred communication channel? Do we want to keep the ML or focus on Issues? Or "github team discussions"?

Personally, I like the github issue structure, since that really allows in-depth discussion about specific issues. OTOH, a mailing list might be useful (in addition to github issues) for over-arching discussions and project directions, or to get help with specific problem (though I only just discovered the mailing list actually existed, so was not subscribed before).

As for github team discussions, I think those are access-limited to organization members only, which would make development discussions non-public, which is a bad thing IMHO.

I also like a chat channel for quick questions and discussions, but that mostly works if sufficient people are on-board. I have been one of the few people that hung around in the IRC channel (#projecthamster on Freenode) for the past few years, but maybe something more modern (Riot/matrix is nice and open and can even bridge to Freenode easily) could be better, if we want to use this at all.

[ederag]

@matthijskooijman
Please read again #574 (comment) and #574 (comment) and
try to walk in my shoes, you might see better. Or not, I'm done fighting.

And to be clear, I did not feel attacked by the philosophical part of your post 😉,
but it does not help at all. It is just common sense.
It's clear that you don't know me,
and did not read carefully the links.
I never felt personaly attacked by @mwilck,
IMO a good will, trying too hard to help the project and hurting it instead, but there's hope.

Thanks for the support expressed in the remainder.

---

OTOH, a mailing list might be useful

Splitting channels does not work, not enough critical mass. It was already tried by @elbenfreund.
Discussions can happen in dedicated issues (see #433), no big deal.

---

I really hope you will reconsider staying on-board with the project, rather than doing your own fork. If there is anything that needs to happen (e.g. in terms of repository access, or stated responsibilities, etc.) for that, please also make that explicit.

Copied from #575 (comment):

This was a cold and well thought decision.
(I saw something coming, not that nasty, but enough to be prepared)
I will not fight with @elbenfreund and those who do not see the wrong in his acts.
Life is too short.

And how could I continue in an organization owned by someone
who proved yesterday disrespect to me and the community
(as explained in #574 (comment)).

No, the only way to continue is where I have full control.
[meaning last resort control;
I'd rather resign than go against the community will]

[mwilck]

I think that the most effort is needed in the shell extension department, where no development (at least no merges) have been happening for quite some time, and some forks have been uploaded to extensions.gnome.org that really need to be unified.

All that's needed to keep the extension going again is a merge and an upload to https://extensions.gnome.org. Branches for different GNOME versions might be in order, too, but that shouldn't be too hard. The commit history of the 3.34 PR mentions every stable GNOME release since then. Out of these, 3.32 with the JS6 classes was the most intrusive one. I believe it should be possible to maintain just two versions of the extension, one for GNOME 3.22 - 3.30 and one for 3.32 and newer. The "official" extension, as currently uploaded to e.g.o. by @elbenfreund, supports GNOME up to 3.20, and could be left as-is for now.

On e.g.o there's currently @jasuarez version with GNOME 3.32 and 3.34 support. He did what I didn't dare / care to do (upload his stuff to e.g.o. without @elbenfreund's blessing). He has merged a previous version of my 3.34 branch, and added features from his PR projecthamster/hamster-shell-extension#305. I think @jasuarez should also be invited, or otherwise encouraged to participate in maintaining the official GNOME extension.

What the extension still needs most is testing. As of today, I test with 3.34 and 3.26, but I can't promise much for other GNOME versions. As we have so few active people, I guess the "dirty" approach would work best: fork at 66a717d, claim support for all GNOME versions from 3.22 to 3.30, and see if any bug reports come in. After all, for those using 3.22 - 3.28, things can't get worse as they're now.

Of course, these are only the commits necessary to make the extension work again for users, with basic functionality. I don't deny that there are other areas where the extension might be improved, but all that can be done on a relaxed time scale.