how open is drizzle?
one piece of jay’s advice to mysql got me thinking about something that bugs me about drizzle development. jay said:
Make all decisions open and transparent: For the non-maintenance team, make a policy that all decisions about the kernel design be done in an open forum, with the community able to participate in the discussion. Have stewards that are willing to negotiate the design decisions with the community and do everything in a transparent manner.
since jay is one of the key lieutenants in the drizzle effort, it only seems to fair to put them up against that standard. one thing i have noticed is that there is relatively little discussion on the drizzle mailing list about all the coding that is going on. the latest discussions mostly seem to be flypaper for standards wonkery and taking shots at mysql.
meanwhile, you have big refactorings of code like mark atwood’s plugin-ization of several features that seem to just go into the code without any discussion of the design or implementation.
maybe all the discussion is happening on the irc channel, but that is not a great way to get much of the potential community involved. how much of it is happening off-line? maybe more than the core drizzle developers realize, or would care to admit.
Comments
This is not altogether surprising for a new project, prior to any kind of initial release, madly working on their first set of goals. It's hard to balance code-time and community-time, and I think it's entirely reasonable for the core group of Drizzle hackers (who appear know each other and their ideas pretty well) to be focused on code-time. Once they have their first release, a roadmap for their next steps, they'll have a basis for engaging with the community.
Code first. It's the golden rule. :-)
(No inside information here, just basing my thoughts on experience with FLOSS projects and watching the Drizzle hackers at the start of their adventure.)
Yeah. IRC is the place to be but I have seen more traffic on the -discuss mailing list in the past week.
if i look at the replication blueprint, i don’t see any mention of anything related to mark’s replication plugin action points.
if i look at the logging blueprint, there is at least a mention of making it a plugin, but none of the blueprints have any sort of information attached to them besides their name.
jeff, i’m also basing my thoughts on my experience with open source projects and the drizzle hackers — that’s why i suspect that they don’t really realize how non-transparent they are being.
i think jay’s point was a good one — a thriving open source project needs people acting to engage the community. mysql has always been terrible at this, and i’m not yet persuaded that drizzle is on the path to being substantially different in that respect.
i hope they become as good at purging the organizational dysfunctions they’ve inherited from mysql as well as the code dysfunctions.
Add a comment
Sorry, comments on this post are closed.
Hi!
Spend some time on IRC and you will find plenty of discussion. Also notice today on the mailing list the long thread on both NULL and storage for databases.
Mark has only put action points in the code, which are far from implementations, and if you look at the state of the blueprints you will see what the current milestones are.
If you care to question something, feel free to bring it up on the mailing list or take it to IRC. Get involved.
Cheers, -Brian