[Server-devel] MoodleMoot A/NZ report

[Martin Langhoff]

The moodlemoots are fantastic gatherings with lots of teachers with
good experience and self-confidence using moodle as a teaching tool,
and many of them using social constructivist approaches. I've
encouraged teachers and educators to visit our wiki and consider
joining our mailing lists, some were very keen on volunteering in
regional efforts. I ran out of Sugar LiveCDs in about 30 seconds.

What it looks like:

  Driving around Napier with Martin Dougiamas
  http://flickr.com/photos/tabitharoder/2928136438/

  NZ moot pictures
  http://flickr.com/photos/tags/mmnz08/

  AU moot pictures
  http://www.flickr.com/search/?q=moodlemootau08

Not many developers at the moot in terms of numbers, but the bulk of
development is done by the team that gathered here. Martin Dougiamas
-- founder and project lead -- has a fantastic eye on UI-and-education
so had some interesting chats about Moodle and the sugar UI, and he'll
be exploring it in his own time too. Also talked with other developers
about fixing (minor) bugs that affect moodle on Browse.xo, and they
were very enthusiastic.

Moodle as a project is at an interesting time. v1.9 has been a very
solid release, closing a long track of incremental releases done every
~6 months. Now the core team is working on a 2.0 that has some
significant improvements... and breaks APIs and changes some coding
conventions. They've spent 6 months on it already, and it's planned
for mid-2009, which is a bit of a long release cycle.

This is a chance to make a huge leap forward -- some of the v2.0
features already in-tree are great, and have been extremely popular
wishlist items -- but also entails the risk of making 2.0 a bit too
hard to upgrade to and in the shadow of the popularity of 1.9 and its
many contributed modules and patches. A case study in this is the
Apache 1.3/2.0 history. It is definitely not the end of the world -
but it is something to keep an eye out for.

Some strategic notes from the POV of the XS,

 - We have a very stable base for our current work (very good news for
the short term). After that, the gap to 2.0 will be large, so when we
want to port to 2.x we will have to rework things a bit. Any new
feature work that is mergeable upstream I'll be trying to merge into
HEAD immediately rather than waiting.

- Upstream is generally very keen to merge our work if it's well
thought out and doesn't break stuff. I think I have a good record in
that regard --  :-) -- and a good sense of what things will be
rejected or cause discomfort with others in the core team.

 - Any new features we develop on top of 1.9 we should publish as
"unofficial" feature patches to get more eyeballs and co-maintainers,
as the community will be starved of new toys to play with in the long
haul to 2.0. I've already started this with a feature backport that
the Edublog team needs. See
http://moodle.org/mod/forum/discuss.php?d=107550

 - Moodle v2.0 has some very good work on content repositories,
inspired by design discussions back in the 2007 moots between Petr
Skodak, MartinD and me. It is strongly aligned with what we (OLPC)
want for the "sharing Library resources across school" goals
(discussed at length earlier this year with sj). The good news is that
it's being developed almost to our specs and needs by the core team;
the bad news is that we have to wait for 2.0 - it's an intrusive patch
that we cannot really backport.

 - Offline Moodle - sj and jg stressed the importance of an
offline-capable moodle. Bryan Berry (Nepal) is also advocating for it.
We now have a reasonable implementation strategy and I talked a lot
with teachers and developers at the moot about it. There is interest
so we may get help from third parties who want it for their own
purposes. I'll be happy to provide architecture, some code and
guidance to anyone ready to help implement it... as long as it covers
our needs :-) A university in Barcelona (UOC) is already rallying
resources to see if they can help. See this forum thread for more info
http://moodle.org/mod/forum/discuss.php?d=107920

The moodle community is broad and some large institutions are pining
for an offline moodle. They can get it early (on 1.9 even) if they
work with me on it :-)

The work pattern ahead of us with Moodle is that we'll develop on 1.9
and port things to HEAD for merge at an appropriate time (immediately
in some cases), rather than develop on HEAD and hope that upstream's
release cycle meets ours. This is how I have been working with Moodle
for ~4 years to great results. This also means that you rarely see me
working in HEAD directly, except when landing patchseries :-) (and
it's the lack of good SCMs in this space that dragged me into git
development).

cheers,



m
-- 
 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