Re: Rhythmbox Predictive Playback - Weekly Report

["Charlotte Curtis" <c f curtis gmail com>]

Weekly Report 9 to follow, but first a reply to last week's report
that I somehow missed:

On Tue, Jul 22, 2008 at 11:59 AM, Sebastian Pölsterl <marduk k-d-w org> wrote:
> You could also change the executable to read from stdin and provide the
> files that way.

I think this is the same as what I was trying to say with b) Change
the executable to one that reads in arguments while it's running and
then use subprocess.communicate to pass it the values.  Just not using
the right terminology ;)

Now, onto last week's activities...

I had a few brave souls willing to test out my plugin, and got plenty
of feedback, both feature suggestions and bugs that I never would have
imagined.  I fixed the easy ones and put the more difficult ones on my
issue tracker, and now I'm debating with myself whether I should begin
re-writing in C++ now (leaving the Python/C++ hybrid version in an
error-prone and semi-unstable state), or fix up the Python version for
the SoC deadline and do the C++ version when there isn't a deadline.

The other exciting development last week occurred late in the day on
Friday - I found out that one of Amarok's developers is creating a
music-matching feature for Amarok, also using Marsyas.  He is
interested in making a slimmed-down version of Marsyas that includes
only the feature extraction algorithms so that end users don't need to
download and build the whole 38 MB Marsyas installation (which
includes many music research tools).

As I said in last week's report, this week and next are going to be
pretty much write-offs as far as the project is concerned due to
traveling, first back to my parent's home (on the other side of the
country) and then to a conference in Ann Arbor.  I may send out a
quick report next Monday if I find time to do some work (on the plane
perhaps), but I'm not counting on it being productive.

Cheers,

Charlotte