Who Is or Will Be the MusicBrainz Server Developer?

In the last weeks I have been talking to Yalaforge, Lukas, Robert and Keschte about integrating Keschte back into MusicBrainz server development. Now, one thing that I have learned from the GreatDispute is not to do mediating behind the scenes, but to MakeIssuesPublic. Therefore I will now tell you what came out of these talks: … Continue reading “Who Is or Will Be the MusicBrainz Server Developer?”

In the last weeks I have been talking to Yalaforge, Lukas, Robert and Keschte about integrating Keschte back into MusicBrainz server development. Now, one thing that I have learned from the GreatDispute is not to do mediating behind the scenes, but to MakeIssuesPublic. Therefore I will now tell you what came out of these talks:

Robert said that he owes Keschte a second chance — if only because the way he handled the conflict was shitty — but that he “does not really believe in it”.

Keschte told me that he wants to work on mbserver again, but on the other hand he will wait and see how the MB community evolves after the dispute. He also said that he probably does not want to start seriously getting involved before next year.

So that’s the situation. There is also a lot of hurt pride and lost trust in there. This does not make things easier. However that is Rob and Keschte’s problem, and nobody can help them with it.

What I can do, and kind of did, was to negotiate a kind of compromise — a situation in which both of them could work together again. This is what it could look like:

  • Keschte would get a kind of sovereign territory in which he can work his way, while Rob can be sure that whatever comes out of it has been thoroughly checked and is of good quality.
  • Keschte does not want to work alone and needs feedback from someone. Robert made very explicit that he cannot and will not fulfill this role. However co-hackers, interested users could do this. That means that Keschte’s development process would have to be opened up to the community. He would need communication tools like forums, a dev-blog, a test server of his own on which “Keschte’s Bleeding Edge” is always available to the public, etc.
  • On the other hand, Keschte would have to work more focused, in better delimited projects, one at a time, than he has on the XHTML server project. Initially this could mean to work with branches (it is clear that major refactoring cannot be done this way, but ArtistPageRedesign could).
  • I do not believe that bug triaging and buffers between Keschte and the users would do any good. Those were discussed in a chat session mediated by Lauri. As I wrote at the end of the Great Dispute, Keschte had been set up in a double bind. You cannot solve this kind of problems with padding, only by changing the setting.

So, this is what it could look like. I say “could”, because both Keschte and Robert stated that they will not put a lot of work into making this happen now. For me this means that I won’t do it either. While I would love to help both in setting up a “sovereign territory” which is helpful and acceptable to both, I will not try to do all the work that they are not willing to do.

The conclusion? There currently is no MusivBrainz server developer. Sad thing, but the truth.

5 thoughts on “Who Is or Will Be the MusicBrainz Server Developer?”

  1. When I read the initial bug triage proposal, I didn’t think it had anything to do with this – it was just a suggestion for improving processes in general, and is probably more applicable to Picard than it is to the server.

  2. There isn’t a software project in the world that doesn’t do triage, whether they call it that or not. MB already does, the developers decide which bugs they will fix, and which they won’t. Most developers hate having to do it by themselves. Giving it a name, and encouraging other people to help, has nothing in the world to do with people arguing, it’s simply about sharing the work.

    It’s something anyone who isn’t a developer could be useful doing, to a point, and the fact it does aid development process taking this task off the developers shoulders, is why everyone from Microsoft to IBM to your average “bedroom linux project” as someone put it, does it.

    Anything that saves developers stress, gives them a more pleasant environment to work in. Anything that gives a developer a more pleasant environment, makes them more productive.

    Trust me, wading through ill written and contentious bug reports from confused and frustrated users (which are generally the kind that make bug reports, happy folks aren’t the ones doing it) is tiring, boring, stressful and largely thankless work. The more shouldears bearing that load, the easier job it is all around.

    And yes, as a process, it’s certainly something easier to apply to Picard.

  3. Minor correction, MusicBrainz does have a server developer: Me.

    I’m making an effort to get 3-4 hours of coding done each day this fall. I expect to see some tangible results from these efforts on the staging server sometime next week.

    Phyltre: You can expect to see Picard development continue. It appears that Lukas is continuing the QT port of Picard as we speak. Fortunately him and Matthias have done wonders for our client projects, which frees me up to do server work.

    Things are not as dire as they appear to be. ๐Ÿ™‚

  4. Hi there,

    First I hope that this time my comment will make it through: I gave up entering a couple of comments in other posts because of some anti spam comment system that treats me as a spambot (mybe because I used the preview button, so I won’t this time, so bear with my typos).

    Anyway. I wanted to give the point of view of me, a casual – the kind that doesn’t keep up with the mailing-list (high volume !) nor #musicbrainz and only reads the blog -, yet very interested, MB user on what I just learned.

    First of all, praises to Don for setting up the GreatDispute wiki page. For weeks I could read between the lines that something was wrong without being able to tell what started it, nor the final results.

    But my point is to say “Cheer up” ! MB is not dead, very far from it. A little flu I’d say, but for the average user / editor, this flu can’t even be percieved. On the contrary, I was baffled to feel something was wrong somewhere, while the actual results are an MB better than ever:
    – the summer changes make importing / editing data so much easier I think this encourages editing a lot,
    – more and more accurate, coherent and exhaustive data, wile still avoiding bloat (and God knows introducing AR could have let the project there)
    – from what rob announced, MB getting a major actor in the music data planet, with a business model finding its way, and MB data / services / libraries being more in more frequently used.

    I guess you guys need to take a break, I’ll help you recognise the situation is not so messy. ๐Ÿ™‚


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.