[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

My report from Debian Med sprint 1.+2. February 2014 in Stonehaven



Hi,

this is the report of my work at the 4th Debian Med sprint in
Sponehaven[0].

On the first evening I gave a short presentation about the effect of our
sprints in the past and some introduction into Mentoring of the
Month[1].

The following packages were uploaded at the Sprint or in direct
consequence of the sprint after traveling home:

  gnuhealth: sponsored upload prepared by Emilien Klein
  gnumed-server: new upstream version
  gnumed-client:new upstream version
  flexbar: new upstream version + verbose testing including fixing
           the of flexbar, checking + fixing tests, contacted upstream
           about failing tests
  seqtk: created in a life packaging session based on some existing
         rough skeleton of packaging
  dnaclust: created from scratch in a life packaging session
  snp-sites: sponsored upload prepared by Jorge Soares
  python3-fitbitscraper: sponsored upload prepared by
         Iain R. Learmonth who just learned the packaging in the
         life packaging session - thanks to Iain for his engagement
         to finalise a package for the first time in just one day
  kissplice: new upstream version
  python-csb: sponsored upload prepared by Tomas Di Domenico

On Saturday (2014-02-01) afternoon I used for five hours the projector
to show live to all interested attendees how I would tackle "random" new
packages (see above seqtk + dnaclust).  The packages were suggested by
the audience and turned out to be simple without serious pitfalls.  Five
of the attendees followed closely my presentation on the screen and
confirmed that this kind of teaching was very efficient.  I admit that
while it was a bit exhausting for me it was fun anyway and sometimes
also some hints of the audience helped solving small stumbling stones.
I would recommend this kind of demonstration also for other events.

As a result of this session on the next day the attendees picked their
own prefered package.  Iain Learmonth was able to finalise
python3-fitbitscraper (see above), Brad Chapman picked freebayes and
reached some decent state and Detlef Wolf continued with the work on
libbioinfoc.  More packages were started by other attendees but nothing
commited until now.  ***  FIXME: Daniel ???

Also people who just had some packaging skills were able to learn some
tricks.  Jorge Soares started his second package (fastaq).  My main work
on Sunday was running around to help people gaining their first
packaging experience after the life packaging session yesterday or
answering other specific packaging questions.

Further work on packages:

  staden: while this was a major point on my initial agenda I only
          touched this a bit since other tasks became more urgent
  libbioinfoc: Detlef Wolf is upstream of this library and we realised
          that some proper build system for the upstream code is
          needed before packaging.  Detlef worked quite successfully
          with several attendees on this task
  vista-foia: via e-mail I worked with Luis Ibanez (who was not able
          to attend) on VistA 

An interesting non-packaging related work item was the discussion with
the EDAM people from Copenhagen.  They intend to create an creating an
ontology of bioinformatics software.  It has turned out that they can
base a large amount of data right upon the Ultimate Debian Database[2]
(actually debtags and package descriptions but also other metadata).  We
went together through their list of items they want to include in the
ontology and realised that for all packages inside Debian these data are
easily available.  Considering that a few missing software packages are
about to be included into Debian we can cover the waste amount of their
target packages which made them quite happy and they considered the
workshop a big success.

Since some time the commit information system CIA was disabled and
replaced by KGB[3].  I worked on the KGB configuration for our Git
repositories but it might need some further testing if it works as
expected and needs to be done for SVN.

Other more general work items:

I added some QA section about the relation to DebTags in the FAQ section
of the Blends documentation.  This was more or less in response to some
question of a member of the Debian Games team.  I also helped fixig some
bugs in the Blends tasks files of Debian Games which by chance
experienced a massive update in while we were on the Debian Med sprint.

I also updated the repack-compression patch for uscan to make it
applicable to the latest devscripts release state.

Since I'm continuosly offering Sponsering of Blends[4] and got a request
for sponsoring libcitygml from Bas Couwenberg and thus I uploaded this
on behalf of the Debian GIS Blend.

Finally I learned that exactly when the sprint started my new 4096Bit
GPG key was included into the Debian keyring and this I had frequent
chances to train my fingers with the new passphrase. :-)

I would like to thanks all attendees of the sprint who participated to
make this sprint another successful event in the row of Debian Med
sprints and I'm very optimistic for future sprints in the upcoming
years.

See you all next year (at latest)

      Andreas.

PS: I guess all attendees wrote their report on their way home as
    I did and will publish it here on the mailing list. :-)

[0] https://wiki.debian.org/DebianMed/Meeting/Aberdeen2014
[1] https://wiki.debian.org/DebianMed/MoM
[2] https://wiki.debian.org/UltimateDebianDatabase
[3] http://kgb.alioth.debian.org/alioth.html
[4] https://wiki.debian.org/DebianPureBlends/SoB

-- 
http://fam-tille.de


Reply to: