gst-launch-0.10 \
v4l2src ! 'video/x-raw-yuv,width=800,height=600,framerate=20/1' \
! tee name=t_vid ! queue ! xvimagesink sync=false t_vid. ! queue \
! videorate ! 'video/x-raw-yuv,framerate=20/1' ! queue \
! mux. pulsesrc ! audio/x-raw-int,rate=48000,channels=2,depth=16 ! queue \
! audioconvert ! queue \
! mux. avimux name=mux ! filesink location=test6.avi

For some reason cheese becomes slow for me when I start recording. Probably the theora encoding? I need to investigate further.

I now have tickets for New York! And yay for Bosnia in 2011.

Next weekend I'll be teaching my brother how to triage Debian GNOME bugs. :)

This could mean that I can finally chat to non-Jabber using friends on a regular basis. It's been nearly four years since I switched from MSN to Jabber, and I think I lost contact with a lot of people. Facebook chat via AJAX is unreliable, and my visits to the site are quite irregular, so hopefully this will work around both problems.

I for one welcome our social network monopoly overlords…

This release was prompted by a patch sent to me by Ian Dash, implementing a filtering feature I was too lazy to write myself. If someone wants to put a fancy GNOME applet around it, I'd be grateful. ;) I think the next step is to add a nicer OO interface.

My original reason for writing the module was to advertise this SOAP API that ATOC publishes - it could easily be wrapped in languages other than Perl. That particular URL was found by inspecting the official widget for Windows Vista.

Actually, that's currently an exaggeration. All it does is add a checkbox in the configure page, and a link to Devel::Cover's reports on the build page when it's enabled. I spent the day remembering how to program in Java.

Tomorrow I might be in a position to extend it into something more attractive - I'll publish it very soon, but I need to run it past my employer. Watch this space.

In other news, I volunteered to package Hudson for Debian, and then discovered just how many dependencies it has. This should keep me going until squeeze+1, I think.

Of course, I wasn't entirely without visiting rights. It so happens that one corner of my new living room is just within range of a terribly expensive wireless access point. It was slow, and kept dropping out. But I managed at least one upload over it, and some email...

I have quite a lot of catching up to do.

Actually, I've been a DD since the weekend, but have been too absorbed to write about it.

In other news, I'm moving house today... I'm more or less packed now. My internet access may be intermittent at home for the next few weeks. (Home is now Southampton, UK.)

• 2009-11-25: #527710 in 'ming' - closed as no longer occurred.
• Prodded some bugs, especially #555036 in 'bash-completion-lib'; don't remember actually fixing any. :(
• 2009-11-29: #552680 in 'libtest-valgrind-perl' - investigated and closed. Was actually a previously-closed bug in valgrind.
• Then bringing up #551926 in pip and python-pip on debian-devel.

Less of the instant gratification of NMUs for me lately.

But; the Perl packaging team is down to a much lower number of RC bugs now. I'm one step closer to being a DD, apparently. And today I found somewhere to live for the next year; this is actually going to threaten my internet access in the near future, which is a little annoying. But once it's done, I'll actually be working on a desktop machine again, rather than a netbook.

While hunting for easy RC bug fixes yesterday, I stumbled across mit-scheme_7.7.90+20090107-1ubuntu1.patch [roughly 9MB]. It contains a Debian .deb to bootstrap the Ubuntu mit-scheme package (bug filed). Nice.

Now, to be fair, Debian has a bootstrapping problem for mit-scheme as well - it requires itself to build, but is currently uninstallable in unstable (although zack's on the case). Having looked at the problem, I think the best thing to do in the long term would be to package mit-scheme-c (which appears to be a superset of the upstream tarball for mit-scheme?) and then use that to build the other package. This would also provide a version of mit-scheme for arches other than i386. If maintainer-built binary packages are going to get thrown away at upload time, that would have been the only way to solve the bug, I think.

(I don't mean to fuel the Debian vs. Ubuntu flames; it would have been nice if the MOTUs could have fixed this in less of a hacky, license-violating way, but they did at least get the package to build, which is more than Debian can do at the moment. I intend to keep a closer eye on Ubuntu bugs and patches for the various packages I work on, because keeping divergence to a minimum should benefit both distributions.)

• 2009-11-17: #527838 in 'smart' - investigated, closed as fixed. (Blatant cheating.)
• 2009-11-18: #516338 in 'pornview' - debugged a segmentation fault on amd64
• 2009-11-19: #551251 in 'libjavascript-perl' - patch written, but need to finish off and release.
• On Friday I looked at postgresql-pllua and postgresql-plproxy, but they need a bit more work than I originally thought, so I joined the pkg-postgresql team. I'm amazed I wasn't a member before, in fact. :) Now just need to actually update the packages.
• Then I took the weekend off! Went out, had fun...
• Yesterday I followed up on bug #516338 above with a delayed NMU, but I believe we're close to a maintainer upload now.
• 2009-11-24: #553230 in 'libapache2-mod-macro' - NMU diff written. I think it's borderline removal material, since there's only been one maintainer upload ever, but it does have a few popcon votes.

The DELAYED queue is a wonderful mechanism for reducing the amount of busy waiting that needs to happen for NMUs; you don't have to remember to come back and upload several days later. For non-DDs, it takes a bit more co-ordination between you and your sponsor, unfortunately - do you send the NMUdiff to the bug before you get sponsorship, and risk having to change it? And then revisit the bug to give notice again once it does get sponsored? Or leave it until after upload, which might reduce the amount of notice you're actually giving the maintainer (if, say, your mentor is in a different timezone)?

I've been opting for the first approach, but it doesn't have the same fire-and-forget quality of the real thing.