Code Sprint or Code Klatsch

[John W. Eaton]

On 12-May-2011, Rik wrote:

| This is a somewhat different e-mail than usual.  I don't have a question
| about code, rather a question about coding.  It's quite nice that Octave
| has programmers from all over the globe contributing, but we all work alone
| (On mourra seul, aussi).  There is a social dimension to coding which I
| first got a chance to see while working on a code sprint with the gEDA
| project and I would like to replicate with Octave.
| 
| Basically, we would set a time and a place (#octave IRC channel) to gather
| for a coding session.  Having the necessary people together can really
| speed up development because the turn-around time for questions is nearly
| instantaneous.  For example, it seems like the FLTK backend is very,
| nearly, almost, 99% ready to become the default plotting engine.  Would it
| be useful for the graphics group in Octave to get together and work out
| amongst themselves the remaining issues?  I think of that sort of gathering
| as a Code Klatcsh since it is getting the right people together without
| definitive goals.
| 
| Alternatively, we could issue a challenge and have a code sprint.  This
| could be something like halving the number of outstanding bug reports, or
| writing tests to get 100% function coverage, or some other goal entirely.
| It can be exciting to see bugs being zapped right and left and finishing
| the day with 100 commits to the repository.  Depending on the challenge, we
| can also get other non-routine contributors to join the sprint.  More work
| is done, although luminaries like jwe probably get to do more advising on
| the IRC channel than coding.
| 
| Have I explained the concept well enough?  Does this seem like a good idea?
|  If so, I think it would be mostly workable.  JWE is at GMT-5 which means
| an early Saturday morning coffee and code session for him would be an
| afternoon coding session for Europe.

Sorry for the delay in responding to this message.

I don't normally use IRC/chat/instant message things but I would be
willing to give this a try if others are interested.

I've also been thinking about doing something like what Matt Mackall
has tried to allow him to spend more time working on Mercurial.  See
item 4 on the following page:

  http://mercurial.selenic.com/wiki/mpm

Unfortunately, I'm not sure how well that is working out for the
longer term since the status is that the funding apparently runs out
this month...  But I'd also be willing to have some consistent "office
hours" if there was some funding to pay me to do that.

jwe