Re: Dan Sugalski's Parrot Post Mortem

[pg]

After I read that post mortem, a clear picture was painted for me. From my point of view, the real problem is the lack of project management. The role of design architect and project manager were totally mixed up. Dan is not only doing the technical stuffs that he was good at (I assume this, but I am an outsider), but also doing things that he was not good at, such as dealing with people and the management side of the project.

I also can feel the tension within the project team(s), not just about personality, more importantly different people were(are) pulling the project to different directions, both Parrot and Perl 6.

[Ovid]

You'll be happy to know that Jesse Vincent has now stepped in as the project manager for Parrot and I'm the temporary grant manager for the project. Having a technical lead doubling as the project lead can create tremendous difficulties for large projects. Fortunately, Parrot is back on track and while there are certainly obstacles to overcome, we now have a set of milestones for the project and funding from NLnet.

I've also read Dan's comments and agreed with some, disagreed with others. I know that many are frustrated by the pace of Perl 6 development and I am too. However, in building and designing a project this large by using primarily volunteer labor, it's still amazing how far the project has come.

[dpuu]

I had a quick look at the "set of milestones" referenced by Ovid. It has the appearence of a dead document: the status of Milestone 1 ends with: "Patrick's expecting a vastly updated engine incorporating subrules and these new features within March 2005." Given today's date is October 2005, this suggests that the status hasn't been updated for at least 6 months.

Later, we see threads: "Design worked out, but still in Dan's head". Maybe this is still true but, if so, it needs to be extracted.

[Anonymous Monk]

To be frank, it will not be a surprise if the project is actually cold dead now. To pick up something years later, most likely throw some old idea around, in a industry famous for its changing, that itself is a bad sign instead of a good one. Not to say, as you have pointed out, there is sign that they didn't really pick up.

The fate of Perl 6 and Parrot could be quite miserable, and is not the first tme you see in this industry.

[Elian]

Oddly, I've been working on documenting the threading plan for Cola. Maybe someone can steal it when I'm done :)

[pg]

This is really good to know!

[Elian]

Y'know, you hit it right on the head. Project management was one of the big missing things in Parrot's run, and I'm glad someone's been gotten to do it. Having one would've made things go better, and realizing I was doing project management would've potentially made things go better as well.

Not that I necessarily would've been great as a project manager -- that's definitely not one of my strengths, and I've some people issues as well -- but at least acknowledging that it was one of the duties would've helped.