[IAEP] maintenance

[Martin Langhoff]

On Fri, Apr 30, 2010 at 6:38 AM, Tomeu Vizoso <tomeu at tomeuvizoso.net> wrote:
> follows a plan about how to improve the situation regarding
> maintenance of our software modules. If you care about it, please
> reply even if only to say so, or even better, comment on it and
> suggest improvements. I will assume that lack of replies mean people
> don't care about it and will stop caring about it myself.

I care deeply. I am working on several aspects of this -- trying to
bring OLPC closer to SL -- not alone of course, many hands are
involved in this.

Let me say -- strictly as my personal opinion -- that it is a good
idea to mimic successful long-lived self-sustaining FOSS projects.
Lots of tricky lessons need to be learned and figured out by all
parties. We reuse other people's code, reuse other people's experience
:-)

> The problem is that very few people in Sugar Labs are willing to do
> that maintenance work.

That is true. And it is a problem for downstreams as well. Following
other projects, I would say that maintenance is something that usually
gets done and funded (directly and indirectly) by those with needs
closer to the end user.

I've said this a few times in private, and I am happy to repeat here:
OLPC needs polish on what Sugar is, and maybe some very specific new
features. Deployments in general need a lot less whizbang -- focus is
on stability/maturity of implementation and APIs, and specific
features.

> I also want to make explicit that almost all maintenance effort has
> come from a few volunteers that are tired and disappointed about the
> little importance that has been given to this work.

I share the pain and frustration. To me, polish and maintenance are
the key ingredient when you deploy to tens of thousands of users.

> == Proposal A: Get downstreams working better inside Sugar Labs ==
> miscommunication. Downstreams don't know how Sugar is developed, who

I agree with your plan, but I disagree on your statement right above
these words. OLPC is your main downstream, and we know (though
sometimes we misunderstand/miscommunicate...).

Most deployments take a while to progress from the initial deployment
work, where logistics and figuring out how to run the show soaks up
all their time and energy, to planning for the next OS upgrade and
thinking "oh, we'd like feature X and bugfix Y in Sugar". And at that
stage, they need to get their team together, and they will -- in most
cases -- work based on OLPC's OS images.

So "get OLPC downstream working closer with SL" is a key step -- gets
almost all your downstreams closer to SL.

That's what I can help on, in any case :-)


m
-- 
 martin.langhoff at gmail.com
 martin at laptop.org -- School Server Architect
 - ask interesting questions
 - don't get distracted with shiny stuff  - working code first
 - http://wiki.laptop.org/go/User:Martinlanghoff