Older blog entries for cjwatson (starting at number 20)

Whee

Digital camera goodness: a Canon PowerShot A40. And it even works perfectly with the Linux USB stack and gphoto2 (at least once I remembered to turn the mode switch on the camera to the "playback" position rather than the "take pictures" position ...).

Debian/BSD

ncm: Actually, the Debian/BSD people have known about the glibc-on-BSD effort for a while (in case you didn't know - I read your diary entry as if you didn't), and I believe the FreeBSD part of that group is already using it. I remember the news came out just before the Debian conference last year, which would have been July or so.

A parallel effort based on BSD libc is useful too, I think, despite being a lot of work, since it makes our code more portable and thus more useful to the wider community.

Cognitive dissonance

It's really weird to see this kind of thing. I look at the top of the page and automatically think "oh, another corporate web site". Then I scroll down and, hey presto, it's a Debian swirl and a link to www.debian.org.

Happy new thingy

To celebrate, I've been hacking down the Debian OpenSSH bug list with vigour. It's slightly depressing that I'm pleased about getting it below 220, but hey. The single most useful thing anyone familiar with OpenSSH can do there is to look for bugs they believe are solved and dig out the actual changes in CVS that fixed them, so that the bugs can be closed with confidence. As such, I've been making a lot of use of 'cvs annotate', 'cvs log', and 'cvs diff' recently; does anyone know of a tool that combines the three so that I don't have to keep switching terminal windows?

Discovered just how large SuSE's patch to man-db is. Eep. Now I understand how some upstreams feel when they see Debian patches - although I do at least try to talk to upstreams before doing significant work. I've started going through the patch and trying to merge as much as possible, but it'll take a while.

I wish I saw this sentence (from a summary of yet another XML-based language) more often:

"Therefore, merely stating that we are going to use XML is not sufficient, we must also explain HOW we will use XML."

Oh joy. Insomnia. I hate it when I can't shut my brain down for long enough to actually get some sleep. I don't know whether it's due to spending several hours triaging openssh bugs or to listening to Sarah McLachlan's "Full Of Grace" too many times.

mulix: If it's been published where you are, try the book of "Things my girlfriend and I have argued about". It's great fun.

MichaelCrawford: If I had received your bug report as a Debian bug report, I would have marked it invalid in just the same way that the Mozilla guys did. Here's why: the purpose of bug reports is to improve the software. Your bug report certainly does a lot of complaining, but there is nothing in it that comes anywhere close to helping the developers improve the software. Instead, it tells them how terrible they are, demoralizes them, and gives them no useful information that might help them resolve your complaint.

When you are submitting a bug report, no matter what the project, please follow these simple guidelines:

  • One bug report per problem.
  • Each bug report should contain enough information on how to reproduce the problem, or you should be willing to work with the developer to find a way to reproduce the problem. (Not every bug has a simple recipe, and that's OK.)
  • Each bug report should be resolvable: there should be something the developer can do to satisfy you that the problem is fixed. (In other words, "there is a problem somewhere, you have to fix it" is not an adequate bug report.)

Each project has its own guidelines, but I doubt there are many who would disagree with these. Michael, please understand that by submitting vague and unresolvable bug reports, no matter what their emotional content, you are not helping the community: you are hindering it by making its members waste time cleaning up bug reports that they have no realistic hope of ever resolving to your satisfaction. In this case, if you truly want to help, what you need to do instead is submit bug reports that say "Mozilla crashes on this page (URL) under these circumstances". Help the developers, otherwise there is nothing they can do but dismiss you as noise. And that helps no-one: not the community, not your wife, not you.

A lazy weekend; I watched videos and did a fair bit of hacking, including tagging debbugs 2.4 (only three years or so after 2.3 ...). Plus I finally wrote up the report for my new-maintainer applicant, which had gone too long without attention. My head is above water for a bit again. :)

I am so much looking forward to taking a week's holiday in a week's time. No particular plans, just a much-needed break.

Mmm. Good party at Vicky's last night, lots of beer, fun people. Took advantage of the end of daylight savings time to party until 2am and sleep from shortly after 1am.

OpenSSH 3.5p1 seems to more or less work. I should be able to upload the Debian package shortly. Then I need to get back and sort out the various people waiting on me for new-maintainer/sponsorship work.

Simon is a genius, as usual, and pterm is a sweet piece of work. xterm takes 9.5 seconds to run 'ls -l /dev' here; pterm clocks under half a second, and with pretty much all the features I want. It's rare that a piece of software is so good that I switch to it from the alternatives practically the moment I see it.

11 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!