1. A good deal of the work on the new embedding API can be done
outside of the Mozilla code base (with an eye to eventually landing it
in the main mozilla tree once it's been reviewed and is stable.)
2. It has contributors who are new to the Mozilla process and don't
have commit access to the tree.
3. It has some contributors who are well-known to the Mozilla community.
4. It's a source code base that will undergo fast and rapid change
while during active development.
Here's Pelle's page on the API:
http://wiki.mozilla.org/Embedding/NewApi
Sponsors and Participants:
We have a few sponsors for this project: myself, Dave Camp, Mark
Finkle. I've been helping to drive where I can on the evangelism and
planning front, Dave has been doing a lot of the GTK+ work and Mark
has agreed to act as a review backstop so that we can land in the main
tree without too much pain.
Pelle Johnsen has been doing a lot of the heavy lifting on the
project. He's new to the Mozilla project and this is a good chance
for him to get more involved with our processes.
There are a number of other people on the .embedding newsgroup who
have expressed interest or otherwise might want to participate.
Giving them access to keep things moving quickly and drive to a better
API is something that an incubation repository might make easier. No
single person is identified yet, but there seems to be a lot of
interest and latent demand. Some subset of those people will likely
become active contributors.
Timeline:
The new embedding code should probably live outside of the main
Mozilla tree for 6-12 months at most. By then we should have a
relatively stable API with a set of consumers.
Results:
The output from this would hopefully be:
1. An API that people are happy with, is well documented, and makes it
easy to embed mozilla into another app.
2. Has at least two consumers of the API as in-the-wild examples of use.
3. A community of contributors that is larger than the original set of
people interested in starting the project.
4. Is well-aligned with the rest of the Mozilla project and the larger
Mozilla code base.
Let me know if you have thoughts! Assuming that no one thinks this is
a terrible idea, I'll open a bug and we can start the process.
Thanks!
--Chris
> Let me know if you have thoughts! Assuming that no one thinks this is a
> terrible idea, I'll open a bug and we can start the process.
Shaver and I both think this is a good idea. Other than getting IT to create
an embedding-api repository, is there any other overhead involved?
--BDS
Need to get pelle an account to start but there's a procedure for
that. Other than that, not really. Let's see if anyone else has
comments during the day.
--Chris
There's work for the sponsors; they need to remain involved and make
sure the right things happen, as the doc specifies. I'm not sure that's
"overhead" though :-)
it does make me wonder how we know which incubator branches exist at any
given time, although maybe some hg tool makes this easy? This last
question has nothing to do with whether the embedding incubator blizzard
proposes is a worthy start.
mitchell
There's work for the sponsors; they need to remain involved and make
I can't find the doc, beyond the two initial blog posts. If it's up,
could I get a link? Might be a good idea to follow up with that on the
incubator thread in .governance, too.
Thanks
Axel
> it does make me wonder how we know which incubator branches exist at any
> given time, although maybe some hg tool makes this easy? This last
> question has nothing to do with whether the embedding incubator blizzard
> proposes is a worthy start.
hg.mozilla.org has a list of all repositories. Is that sufficient?
--BDS
> I can't find the doc, beyond the two initial blog posts. If it's up,
> could I get a link? Might be a good idea to follow up with that on the
> incubator thread in .governance, too.
There's the original link that I posted that Pelle wrote up:
http://wiki.mozilla.org/Embedding/NewApi
Along with the (pretty exhaustive) blog posts, what else were you
looking for?
--Chris
As said on irc, I'm looking for something that publishes the process for
incubator reps in general, not a description of the embedding project in
particular.
Axel
I think that Mitchell's blog post is the best I've seen so far:
http://blog.lizardwrangler.com/2008/06/06/incubator-repositories-proposal/
She told me that they were planning on putting it up somewhere official.
--Chris