MSN SoC Project

[Elliott Sales de Andrade]

On Wed, Apr 28, 2010 at 12:36 AM, John Bailey <rekkanoryo at rekkanoryo.org> wrote:
> Sorry for this bit of top-posting here, but Mark was kind enough to forward this
> message on to me for my own reference, and I thought some of the points below
> were well worth discussing here.  I'd like to see more SoC discussions here on
> devel.
>
> On 04/27/2010 02:43 PM, Mark Doliner wrote:
> <snip>
> 2010/4/27 Jorge Villaseñor <salinasv at gmail.com>:
>>> Elliot, I think it worth to finish msnp16, merge it and then start
>>> over this code the refactor + rewrite, what do you think?
>>> Mark, this discussion must be private or is better to have it on devel?
>>> Do I need to create a new im.pidgin.gsoc.2010.msnp_rewrite or continue
>>
>> Probably better to have the discussion on devel, I guess.  I don't
>> think there are many people familiar enough with msnp16 to give much
>> useful feedback, but at least it makes it possible for people to
>> follow along.
>
> Having discussions on devel where relevant is a great idea.  It would probably
> be better for all our students and for our success in the SoC as a whole to have
> as many of our developers engaged in the discussions as possible, not just the
> specific mentor for a given project.  In this case, Ka-Hing and Stu may have
> useful feedback for you, and others may have useful insight for any design
> questions you may have.
>

It's probably good to keep everyone in the loop just in case I happen
to be away for
some long period (not that I plan to do so).

>> I don't see a need to create a new branch, but I don't know what
>> Elliott prefers.  Also, I think Google asks you to submit your changes
>> at the end of the summer?  It might be easier to produce one large
>> diff if you can just grab all changes from a given branch.  /me
>> shrugs.
>
> In this case, I have no preference whether the current msnp16 branch is used or
> whether a new SoC branch is used, since this work is already happening off
> im.pidgin.pidgin.  That said, Mark's point about producing the large diff from a
> branch would actually be an excellent argument for using the existing msnp16
> branch as the starting point, propagating im.pidgin.pidgin to it, and switching
> to something like im.pidgin.soc.2010.msn-tlc at the first commit after the
> propagate.
>

Officially, I think this is still the community bonding phase. So I'd
say you can
continue to work on the msnp16 branch until the official start of coding on May
24th. Around then, we can take a look again to see whether to start from there
or just from the main branch, depending on how well things work.

Google does seem to require code samples (of what sort, I didn't really look)
so a completely separate branch may be more useful. All that requires is a
diligent use of 'mtn propagate' anyway.

> Since this project touches core MSN functionality, I'd also say if you reach
> milestones during the summer where both you and Elliott feel the progress to
> that point is release-ready, it would probably be a good idea to merge back to
> im.pidgin.pidgin while continuing to work on the other branch.
>
> John
>
>