Leadership, an apology, and important questions

[Evan Schoenberg]

I am now aware, after receiving messages from Gary Kramlich (grim) in private communication, that I may have overstepped or sidestepped the wishes of the current Adium community.  It also possible that this is a misunderstanding and the community is not, in fact, up in arms at me. This certainly wasn’t my goal. I will not be responding to any further private emails about the current roadmap, individuals’ involvement, etc in the immediate future.  I will respond to any messages sent to adium-devl or to the private leadership-only channels, though. This is because I would prefer to keep the whole conversation fully in the public to avoid any further misunderstandings.

I understood Robbie’s unanswered request for help moving forward with a Catalina release, in the context of zero Adium commits since 4/23/2017 (with the exception of merging in the fix-sFlags and fix-autoscroll branches from @trolan who I believe is new to the scene), to mean that nobody was involved from a programming and codebase perspective at this point.  This opinion was furthered when I downloaded the current release branch and found that it did not compile successfully with current version of macOS and Xcode.  

This is, of course, separate and distinct from documentation, AdiumXtras, user support, and website infrastructure, the status of which I did not assess in detail (but recall that the last message about it that I was privy to was on 4/13/19, when Colin wrote that:

A few of us are working to get the behind-the-scenes infrastructure repaired and restored so we can proceed with development and releases

I stepped up to try to resume a leadership role to bring this back to a level that someone without nearly 20 years of experience in the Adium codebase could work on it.

With extremely limited time resources, I felt the best move for Adium was to provide Asher, who offered to help and who I trust as much as I trust myself, with commit access to help directly after he offered to do so.    I felt that “unilateral decisions” were the only decisions likely to move forward in a timely fashion given the apparent radio silence.  We’re all the heroes of our own stories, right?

It is clear that regardless of the radio silence at the code level, Robbie and Mathew should have been involved in this discussion.  That was an error, and it was my error.  Gentlemen, please accept my apologies.  The most I can protest is that I haven’t done this in a long time. Small excuse.

Nothing destructive and nothing permanent has been changed through any action I have taken or Asher has taken.  Asher has proposed (in his characteristic direct way) that Adium’s code base should move to git rather than mercurial, because he is the only person stating a plan to work with the code in the immediate future and is much more comfortable with git. If the community of those who are working with the Adium program code disagree with his proposal, let’s have a discussion about that.  Worst case he takes whatever patches he’s made and rolls them back into the hg repository, and if desired the new branch in the GitHub mirror is kept as a mirror (via git-hg mirror or some other mechanism) rather than becoming canonical.

What was perhaps destructive, at least to espirit d’corps, was taking a call for help to be an invitation to do more than submit a patch or pull request. The Adium team page lists me as ‘lead developer emeritus.’ Honorary titles don’t necessarily confer any control.  

I don’t have the time-bandwidth to get into long debates, so I have several simple questions:  

1. Thijs, if you’re still monitoring this list, I handed the reigns over to you quite a few years ago. I have no intention of staging a coup.  If you intend to return to programming Adium, and you would like to take a lead role in welcoming Asher as a potential new developer for the project, and you’re willing to take responsibility for reviewing and accepting his patches in a timely & collaborative fashion, please speak up.  If you wish it, we can together ask Asher to follow a more traditional path to having direct commit access in the official repository.  With the advent of distributed version control, this isn’t the hardship it once was, since he can work exactly as he currently is, within his own fork of the project, until changes are pulled in.

- On the other hand, if I’m to act as lead developer for a time, I deputize Asher directly to commit access, and I embrace simplifying our codebase architecture by moving away from less-known tools and toward widely-embraced ones.  There is no reason that doing so has to interfere with our documentation on Trac (and anywhere else it may be).  We can even keep the ticket system on Trac, either as the canonical bug system or as the historical one with a transition to GitHub’s issue tracker. 

- I poured my soul into the project for a decade. Please consider that I wouldn’t hand the keys to someone lightly.

2. Robbie and Matthew, the two of you have in various ways served as Adium’s stewards and project managers for many years.  Asher doesn’t want your jobs.  I saw a #adium log where the phrase ‘project manager’ was used, but I am 100% confident from talking with Asher that this doesn’t mean to him what it had traditionally to Adium. He just wants as frictionless an environment as possible to bring Adium’s code into 2019.  He is going to need your help and your enthusiasm to do so.

3. I have reviewed the Adium archives mailing list since early 2016. Thank you to everyone from the ‘old guard’ who has remained on the list, evidence that you still care about Adium. I don’t know who has been in #adium on freenode so this is my place holder. I call out Chris and Colin, though I may have missed others.  Question 2 applies to you as well.

4. Gary, I seem to have mis-stepped with you. I sincerely hope you will rescind your apparent decision to offer no further help to the project, which I conclude from your statement that you’ve removed your ssh keys from the servers.  Perhaps I just misunderstand what that meant, but it sounded to me like you’re washing your hands of it.  Wearing your IMFreedom hat, I hope you would support Adium the project — GPL’d, free, open source instant messaging client based upon the amazing libpurple (née libgaim) library — regardless of who is contributing and what infrastructure the project chooses to use.  Let’s engage the community of active contributors - programmers and non-programmers alike - and figure out a professional context in which your valued contributions can continue.

Best regards,

Evan

[Robert Vehse]

Thanks for this, Evan.

> This is, of course, separate and distinct from documentation, AdiumXtras, user support, and website infrastructure, the status of which I did not assess in detail (but recall that the last message about it that I was privy to was on 4/13/19, when Colin wrote that:
>
>> A few of us are working to get the behind-the-scenes infrastructure repaired and restored so we can proceed with development and releases

I'd like to take this opportunity to (belatedly) acknowledge that in recent months Gary has been putting a huge amount of work into restoring and improving Adium's broken infrastructure. The items listed at https://trello.com/b/nhz2JwDw/adium only show some of the work done but not all of it.

> I stepped up to try to resume a leadership role to bring this back to a level that someone without nearly 20 years of experience in the Adium codebase could work on it.

Thank you! :-)

> With extremely limited time resources, I felt the best move for Adium was to provide Asher, who offered to help and who I trust as much as I trust myself, with commit access to help directly after he offered to do so. I felt that “unilateral decisions” were the only decisions likely to move forward in a timely fashion given the apparent radio silence. We’re all the heroes of our own stories, right?
>
> It is clear that regardless of the radio silence at the code level, Robbie and Mathew should have been involved in this discussion. That was an error, and it was my error. Gentlemen, please accept my apologies. The most I can protest is that I haven’t done this in a long time. Small excuse.

Apology accepted. I know you meant well, Evan.

Asher's "i just moved adium off hg" on IRC sounded very definitive and I can understand how that would have irked Gary, who, as part of his work on our infrastructure, has been busy researching Mercurial alternatives to Bitbucket.

I regret not having informed the community on the mailing list of the work Gary has been doing and in this sense I feel partly responsible for the conflict that occurred between Asher and Gary. If Evan and Asher had known that there was work going on, they might have felt more confident about getting a response on the mailing list.

I am fine with the move to Github as long as we make sure the migration works well and we don't lose any important information in the process. Asher and I have been working on that. I think there are a lot of benefits to Github compared to what we have.

Asher has already done a lot of work towards bringing "Adium's code into 2019" and that's great. I'm glad to contribute my help and enthusiasm. :-)

-Robbie

[Mathuaerknedam !]

Sorry for the delay in replying. Life's been a wee bit more complicated than usual over the last week.

I agree with your conclusion that the apparent suddenness of the announcement was a big factor, combined with the fact that there was no prior discussion in leadership channels. I accept your apology, and I certainly hope we can all work together to move Adium forward. A lot of the enthusiasm around Adium has been the community involvement, and I hope we can rebuild that as well.

I don't have an opinion on Git vs Mercurial, but I believe that there were real practical reasons for choosing hg over git. I have a vague recollection that the ability to import svn history worked better? Whatever the reasons, it's possible that they are no longer relevant, or that there are still good reasons to use it. I don't know, but it would be great to hear from anyone that recalls the history or has a pointer to it.

In hope that the discussion will continue, I'll refrain from trying to say everything possible in one email. I think this is probably enough for now. :)

I look forward to hearing more from others.

 

Matthew

--

Matthew