Limesurvey - Plowing ahead!
david at gnsa.us
Sun Mar 21 18:38:45 UTC 2010
-----BEGIN PGP SIGNED MESSAGE-----
On Sat, Mar 20, 2010 at 11:51 PM, Robyn Bergeron wrote:
> All -
> I'd really like to start nailing down some of our "next steps" for
> Limesurvey deployment. We seem to have a lot of bits of information
> floating around but nothing cohesive defining what the steps are to
> getting it out the door. Since we're getting really close to having
> Fedora Insight deployed (YAY!!) I think it's a good time to define
> what our plan will be for getting another important piece of marketing
> infrastructure up and running; once people have more time freed up, we
> can actually start plowing through the task list.
> I think it would be wise to have an hour-long IRC meeting to discuss
> the following:
> * A briefing on what our plan of action should be to get any of the
> limesurvey packages that have issues (for those not in the loop - some
> of the packages have been forked / modified from their originals, for
> various reason) into a state where they can be used. Some background
> here on "what has already happened and what we've already discovered"
> would be helpful, both in the meeting and then documented in wiki
> form; we don't want anyone wasting cycles on reinventing the wheel.
> * Figuring out some timelines - how long will packaging / fixing of
> packages realistically take, how long to develop the instructions for
> the Infrastructure team to get it into testing / deployment, etc. I've
> probably skipped over some things here, so I'm counting on those who
> are more Zen than I with infrastructure deployments to hit me with the
> cluebat here.
> * Getting owners for any tasks that we come up with.
> * Getting owners for documenting the actual plan of action /
> background information / etc. on the wiki in the near-term time frame.
> I have added some framework around "what the next steps are" on the
> wiki - please take a look.
> This is the wiki page where we had previously just been tracking the
> individual bz's for limesurvey packages - it makes the most sense to
> add in a timeline for packaging and deployment here (imho).
> There are probably task items I've missed - if so, please add to the wiki.
> If anyone is interested in working on any of the steps listed - or if
> anyone has thoughts, etc. - just respond to this mail, and I'll try
> and work out an agreeable time with those interested for an IRC
> meeting, which would probably be in the next 2 weeks. Most of our
> regular marketing meetings will be revolving around F13 marketing
> deliverables for the next few weeks and I don't want to interrupt
> taking care of those deliverables... but at the same time, I don't
> want to push limesurvey progress out to F28 time frame. :)
> Additionally - I've gone in and made a new custom report in the
> marketing trac instance for Limesurvey, and added Limesurvey as a
> component in trac. (This means that tickets / tasks can actually be
> assigned as Limesurvey tickets, rather than just as a general
> marketing ticket.) There are no tickets currently (save for the one I
> created to make sure the query I made was actually working) - but I
> would like to utilize trac for keeping track (har har) of all the
> associated Limesurvey tasks. The list of available reports for
> marketing's Trac instance is at
> https://fedorahosted.org/marketing-team/report - just click on
> Limesurvey to see where all the future tasks will be! (Sorry, I feel
> awesome whenever I figure out how to do something more technically
> advanced than writing wiki pages. :D)
> Also - David, I know you pointed me upwards in logs to a discussion
> you had with someone from limesurvey on IRC at some point in the past
> few days - it's long since scrolled off of my IRC backlog, as well as
> my memory. Can you bring everyone up to speed on what he had to say?
> I think it was mostly a background / why limesurvey forked things
> Thanks everyone!!
> marketing mailing list
> marketing at lists.fedoraproject.org
So from my perspective - there are really two things that come next,
and can run concurrently
1. We need to get the packaging done esp of the dependencies. There is
no way around this, a lot of these dependencies are in wait states.
2. We need to find out how badly some of the packages are forked. We
discovered that some are just forked for things like MS SQL support -
which we don't care about. If something is heavily modified we need to
get that list of libraries that can't be separated out and up to fesco
to ask for an exception.
Perhaps we need to ask the php-sig for help, or put some stuff in the
eng. svcs queue.
As for what upstream said
They indicated that they'd make the includes directory definition a
configuration option so that we could define that and greatly reduce
some of our work. That's a win for us and other distros IMO.
Timelines for this - well - it depends on quality of packaging - and
speed of reviewers, etc. We completely missed the boat with timelines
for Zikula - so I am hesitant to even venture a guess here.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Use GnuPG with Firefox : http://getfiregpg.org (Version: 0.7.10)
-----END PGP SIGNATURE-----
More information about the marketing