7/14/11
John,
Absolutely coincidentally, I took up the idea of a code sprint this
morning again and wrote an e-mail to Jordi about it. Here is the
message which lays out some ideas.
--Rik
-------- Original Message --------
7/14/11
Jordi,
So far, you were the only one who expressed interest in having a code
sprint. This may be because most of the other Octave coders have never
been through one. In any case, I'm going to propose that we two organize
one and hopefully it comes off well enough that people will want to do it
again.
Most of what we need to do is logistics and publicity.
On the logistics front, we need a time and a place. To avoid both work
days and religous days I would suggest a Saturday in July. I think the
16th or the 23rd would work well. I took a brief look at international
calendars and these days seem to be holiday free. Earlier in July will run
into the July 4th holiday in the U.S. and later in July gets close to the
big European holidays where countries like France take the whole month of
August off.
For time, I was thinking 10 a.m. EST (GMT-5). This is 7 a.m. for me, but
that's doable. I would say a minimum of 4 hours would be a good length and
it could possible extend longer if people are interested and still coding.
I propose just meeting on the Octave IRC channel.
For publicity, I was thinking of announcements on the Octave mailing lists
(maintainers, but also help). We can probably repeat the announcements as
the date approaches. I think jwe would also let us put something on the
main Octave.org website. Finally, what do you think about starting a Wiki
page for the code sprint on the Octave wiki? If you can think of other
ways of getting the word out I would be interested.
Finally, I thought it would be nice to have big goals for the sprint. I
thought of two, not exclusive, possibilities. 1) Halve the number of
outstanding bugs on the bug tracker and 2) Halve the number of functions
without built-in self tests. I like both because they are numeric, easily
trackable, and automatically split into bite-size chunks for handing off to
people. In particular, I thought the second goal might be good if we get
enthusiastic volunteers who haven't coded much for Octave. They could
certainly write tests by reading a particular .m file without having to
know all about Octave internals which some of the bugs on the tracker require.
What do you think?
Rik
|