[epo-core] Fw: Fwd: GSoC 2009 planning and U.R.

Matt S Trout mst at shadowcat.co.uk
Wed Sep 10 15:06:27 BST 2008 

----- Forwarded message from Eric Wilhelm <scratchcomputing at gmail.com> -----

To: directors at lists.parrot.org, Allison Randal <allison at parrot.org>,
	Matt Trout <mst at shadowcatsystems.co.uk>,
	David Wheeler <david at kineticode.com>
Subject: Fwd: GSoC 2009 planning and U.R.
Cc: Joshua McAdams <joshua.mcadams at gmail.com>,
	Karen Pauley <karen-perl at martian.org>,
	jose.castro at perlfoundation.org,
	Jim Brandt <cbrandt at perlfoundation.org>,
	Richard Dice <rdice at perlfoundation.org>
From: Eric Wilhelm <scratchcomputing at gmail.com>

Hi all,

I think the parrot foundation should plan to apply for Google's Summer 
of Code next year, and I would like EPO (and possibly others) to do the 
same.  Each org should really start planning this now and I would like 
to see TPF identify various segments of the community which might be 
able to apply as an org and/or participate as part of an org.  GSoC is 
getting rather crowded, but they seem to be refining their methods and 
gearing-up for next year, so it seems worth the effort.

And a very important note:  please do not assume that I am going to 
be "the guy".  This is where TPF dropped the ball in 2007 and I don't 
want anyone to assume that I'm going to do everything (or maybe 
anything -- I might be busy applying as VectorSection) next year.

So, some thoughts on what should be done, along with the forward 
below...

  A. Organizing:
    1. create a mailing list for perl/parrot GSoC/university relations
    2. find ~two volunteers per org to be the UR people
    3. identify potential GSoC admins, backups, advisors, and mentors

  B. Timeline:
    October:      Establish university contacts, concrete plans.
    November:     Activity on-campus: presentations, etc.
    December 15:  Reminders about GSoC for orgs, mentors, and students.
    February 10:  Reminders for mentor orgs: "prepare your application"
    March 1:      Ready to send applications, signup mentors, idea list.
    March 10-18:  Actively seeking students.

  (This assumes the 2008 timeline:
  http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline )

Elaborating:

1.  My vision of the mailing list is about sharing ideas and contacts, 
news, etc for all of the "pod people" who have an interest in 
GSoC/universities.  I think the UR effort has a lot of benefit to all 
of these communities and that success in GSoC follows from a good 
university presence.  So, I propose one common list as a bridge between 
the orgs - who wants to host it?

2.  I would recommend that the UR people and GSoC admins be different 
people.  The UR's could potentially act as GSoC backups, advisors, or 
mentors.  The spearhead of UR could also be the admin, but I would 
guess they would get overwhelmed trying to run an umbrella the size of 
TPF, especially if they've done a good job at UR ;-)

3.  Admins/Backups/Advisors - see below and/or ask me if it isn't clear 
how much effort I put in as admin.  I think the primary qualifications 
of an org admin are a very small and inefficient-for-me subset of my 
skills - namely:  editing wikis, reading google's trainwreck of 
documentation + mailing lists + arbitrarily changing decisions, and 
keeping track of non-technical issues (cats, deadlines.)  The other 
aspect is making executive decisions and assessing the technical 
details of a given proposal or student progress.  This is where I think 
technical advisors (who know the guts well, but still have some time to 
communicate with students and the admin) would be a better approach.  
The bigger the umbrella, the more diverse technology - I know a fair 
bit about perl, the CPAN, CPAN/CPANPLUS, Module::Build, TAP::Harness, 
swig, XS, wx, parrot, perl 6, catalyst, DBI, etc but I think the admin 
probably doesn't need to have all of that in one head so long as one or 
two trusted backups and advisors can be efficiently communicated with.

For smaller orgs, the slot allocation plays-out better than a large 
umbrella.  Python's ~22 slots was ~15% of their total student proposals 
(those numbers are only from memory), but each org is guaranteed one 
slot - so 4 proposals and one slot => 25%.  This means more "pod 
people" orgs is better, so umbrellas helping little orgs along is good.  
Did anyone look at how many orgs this year were for projects written in 
PHP or Python?

As for the timeline, I think November is the critical point.  More on 
that later.  Reminders to org admins would have helped Bricolage this 
year - David just forgot to apply.  Getting EPO on track earlier might 
have also helped their application and student turnout.

The Fwd below is just to bring the rest of you up to speed.

--Eric
----------  Forwarded Message:  ----------

Subject: GSoC publicity slingshot plan
Date: Sunday 31 August 2008 00:49
From: Eric Wilhelm <ewilhelm at cpan.org>
To: Richard Dice <rdice at perlfoundation.org>, Josh McAdams 
<joshua.mcadams at gmail.com>

Hi Josh and Richard,

I'm trying to gather more comments from the mentors regarding summer of
code and hoping that Perl can start to work toward SoC 2009 and a
general university presence by building on what we have done this year.

I have written a recap here with links to the various project blogs and
code.  I can hopefully add links to the recaps from the mentors within
a few days.  I would like to package everything up as an archive on
perlfoundation.org or something within the next couple of weeks.

  http://use.perl.org/~Eric+Wilhelm/journal/37299

I think it would be good to find somebody to spearhead a university
relations program for Perl.  This might take the form of organizing pm
groups to reach out to their local university or getting in touch with
department heads, professors, on-campus clubs, etc.  The semester is
just starting, and I think Perl's presence on campus in mid-November is
going to play a big part in determining what kind of student turnout we
get next year.

I think the weakness of our student turnout this year may have partly
due to my lack of confidence that we would get in (after doing the
historical research, it sounded doubtful), and consequently I didn't
really have time to focus on reaching out to students (possibly I
over-compensated in the mentor recruiting.)

But I also didn't really have a plan for reaching students and I think
the pmgroups may have failed there.

So, my conclusion is that TPF needs to start trying to reach students
now.  Regardless of whether we know there is a GSoC next year or
whether we'll be in it.  I think perhaps that we needed to do this a
couple of years ago anyway simply for the good of the community and the
future of Perl.  The more prominent communities within Perl (Catalyst,
DBI, etc) probably need to do the same and I hope to try to push EPO,
Parrot, etc to try to get a university presence and apply for GSoC as
separate entities next year.  EPO did apply this year, but I suspect
they got the "we'll just try one perl" treatment, and I hope that we've
now laid the foundation for them having a better chance next year.


Another thought:  if someone wants to go through the list of other SoC
organizations and look for projects that were done in Perl, that would
be good information to have for next year and possibly good fuel for
the PR engine.


So, that's things as I see them.  I would really like for TPF to
spearhead a university relations effort, but honestly I can't afford to
invest much more time this year.  It's been almost 250 hours now, which
is really difficult for a lone consultant.  I'll do what I can to help
whatever people you can get working on this, but there is a lot of work
to be done before next February and I think a lot of it is due before
Dec.

Ideas/thoughts?


Thanks,
Eric
-- 
perl -e 'srand; print join(" ",sort({rand() < 0.5}
  qw(sometimes it is important to be consistent)));'
---------------------------------------------------
    http://scratchcomputing.com
---------------------------------------------------

----- End forwarded message -----

-- 
      Matt S Trout       Need help with your Catalyst or DBIx::Class project?
   Technical Director                    http://www.shadowcat.co.uk/catalyst/
 Shadowcat Systems Ltd.  Want a managed development or deployment platform?
http://chainsawblues.vox.com/            http://www.shadowcat.co.uk/servers/

More information about the epo-core mailing list