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

Summarising Debian Med sprint online (Was: Third day of online sprint ...)



Hi Debian Med sprinters and those who are interested in some summary

at first thanks to all who joined the sprint.  My personal impression
was:

  * I had the perfect sprint feeling since I took vacation as always
    and worked from before 8:00 than after 22:00 local time with only
    few breaks.  This was possible on one hand due to my very
    understanding and patient wife (thanks a lot to her) and my
    intention to focus on the sprint (like refusing invitations of
    friends for skiing etc.)

  * The impression to "share a common room" with other sprinters
    that would sit next to me should have been mimicked by a constantly
    open Jitsi meeting where it is possible to speak and get immediate
    attention.  Unfortunately not many people stayed constantly in
    that room (for whatever reason) but from my point of view that
    worked.  I would love if more people would join that "common table"
    and learn how this works.

  * After realising that only very few attendees joined the Jitsi
    room we started short meetings every second hour to provide
    some kind of face to face experience.  This seemed to work to
    some extend.  I've droped the chatlogs and some screenshots
    from the Jitsi room to our sprints repository[1].

  * Those who dedicated their time to the sprint created the usual
    productive and helpful environment that is typical for our
    sprints.

  * Its great that we established Matrix as a sensible means of
    communication allowing instant messaging with some modern
    interface.  Everybody is invited to join

        https://app.element.io/#/room/#debian-med:matrix.org

    since it seems sensible to continue using this channel.

  * All in all I'm happy we did that sprint and would love to
    repeat this kind of events in future.

Where we might do possibly better next time

  * My expectation to see those people who are kind of "regular"
    sprint attendees was only partly fulfilled.  I'm wondering how
    we can be more inviting next time.  Sure we were volunteers
    but the "no time" excuse was used by more people than I
    naively expected.  I assume this did not happen in past real
    life meetings after people had booked a flight and reserved
    a hotel room.  Once you are at a remote location the event
    becomes priority over other important things and finally
    "no time" is a synonym for "this event has not made it on top
    of my priority list".
    So how can we make the actually way more environment saving
    sprint format more "important" to behave the same as if the
    flight would have been booked and this appointment gets higher
    priority than other things of real life?

  * I'm constantly wondering how we could better mimik the face
    to face experience.  I admit before the sprint I developed the
    false hope that we could view the majority of the Debian Med
    team at least once on one screen - may be stress testing
    Jitsi by filling about 30 tiles on the screen.  IMHO this
    could help to identify better with the people behind the
    project who are spread widely all over the world.  Ideas how
    we could do this are very welcome.

  * Steffen requested more focussed tasks on sprints than just
    "normal" Debian Med work.  For this specific sprint I think
    the most urgent and burning task was to make sure all our
    packages will migrate to testing and are free of RC bugs.
    If we do not manage to do this outside a sprint (as it was
    the case obviously) so it needs to be done in a sprint.

    But in general I agree with Steffen.  However, we really need
    to clarify
      - Who will set these tasks?  We need an expert / users
        who set priorities what should we package.
      - Should we do some "Package of the week" (independently
        from the sprints?) to push all human-power to one
        specific and sufficiently important package?
      - How can we set the foxus more on full workflows a user
        can run.
    I do not think that simply maintaining a spreadsheet[2] and
    hoping that random developers will stumble upon it and
    expect them to pick the most urgent task is very fruitful.
    We need better coordination here (and this is another takeaway
    message of mine from our sprint.

I repeat my big thanks to all who joined, specific thanks goes
out to all newcomers and I would be very happy if you would
join future sprints.

Kind regards

    Andreas.

Links:
[0] https://salsa.debian.org/med-team/community/sprints/-/wikis/202102_debian%20med_sprint_online
[1] https://salsa.debian.org/med-team/community/sprints/-/tree/master/20210218
[2] https://docs.google.com/spreadsheets/d/1tApLhVqxRZ2VOuMH_aPUgFENQJfbLlB_PFH_Ah_q7hM/edit?ts=5eb957ba#gid=543782716

On Sun, Feb 21, 2021 at 11:06:38AM +0100, Andreas Tille wrote:
> Hi,
> 
> I'm happy to confirm that at the third day of our sprint more people
> joined.  Most possibly it is due to the weekend since people did not
> took vacation.  I admit I'm honestly wondering how we might turn an
> online sprint to the same importance as a real life sprint.  I mean, if
> there would be some real life sprint the usual contributors would have
> booked a flight and a hotel room and would be at the venue and have
> time.  In contrast to this in our online sprint many people who usually
> show up confirmed "can't join, no time".  So for the next online sprint
> we have the challenge to make people aware: "Hey it is sprint and it
> would be great if you would behave like beeing inside a hotel somewhere
> at the world.  By chance this hotel looks like home and you need to care
> for the food yourself."  (To repeat an old Debian quote: "Home is where
> you need to wash the dishes." ;-) )
> 
> Yesterday we established short Jitsi meetings every second hour at
> odd UTC hours.  Everybody is very welcome to join these meetings[1]
> at 11:00 / 13:00 / 15:00 / 17:00 / 19:00 (may be 21:00) UTC today
> just to say hello is perfectly welcome.
> 
> Those who took part actively recovered the usual sprint productivity
> which is really great.  Thanks a lot to everybody (specifically also
> to Tássia who does not consider herself a team member but wanted to
> support our sprint - I'd love to see more those contributions from
> other Debian people)
> 
>      Andreas.
> 
> 
> [1] https://jitsi.debian.social/DebianMedCovid19
> 
> On Fri, Feb 19, 2021 at 09:11:17PM +0100, Andreas Tille wrote:
> > Hi again,
> > 
> > as yesterday some people were hunting down bugs.  This morning we also
> > teached a newcomer quite a bit - my hope is Robbi will finalise his
> > first package with the end of the sprint.
> > 
> > I personally dived into RC bugs of other teams (Debian Science,
> > Debichem) affecting packages in Debian Med.  That's a general advise
> > also for the time after the sprint: If you want to help Debian Med
> > it makes sense to help hunting general RC bugs to keep the freeze
> > period as short as possible.  The earlier Debian 10 will be released
> > the sooner we can have new fun with new packages again. ;-)
> > 
> > Thanks to all who joined and see you tomorrow
> > 
> >       Andreas.
> > 
> > On Thu, Feb 18, 2021 at 09:31:35PM +0100, Andreas Tille wrote:
> > > Hi folks,
> > > 
> > > just a short notice about the sprint today.  We have established a
> > > new communitcation channel:
> > > 
> > >     https://app.element.io/#/room/#debian-med:matrix.org
> > > 
> > > Please join us on matrix if you consider instant messaging a sensible
> > > enhancement for your workflow.
> > > 
> > > Regarding the sprint itself:  I admit I had hoped to more attendees but
> > > it seems people do not take one or two days off if the sprint is "only"
> > > from home.  If you need to book a plane and travel to some place that is
> > > different.  Those who joined did some valuable contributoins reaching
> > > from very generic stuff (Étienne worked on libzstd and now has a
> > > changelog entry on every Debian 10 entry installation ;-)) via fixing
> > > several cross-build issues (Nilesh) and other things.  I for myself
> > > concentrated a bit on BioConductor issues and kept on hunting not yet
> > > migrated packages.
> > > 
> > > I'd be super happy to see more people tomorrow.
> > > 
> > > Kind regards
> > > 
> > >      Andreas.
> > 
> > -- 
> > http://fam-tille.de
> > 
> > 
> 
> -- 
> http://fam-tille.de
> 
> 

-- 
http://fam-tille.de


Reply to: