Karsten Bräckelmann’s Open Thoughts

Yay, mc-fast is back. And what a nice feedback it sent me…

Freqs for those rules in 'mc-fast' mass-check:
  MSECS    SPAM%     HAM%     S/O    RANK   SCORE  NAME
      0     1996     1999    0.500   0.00    0.00  (all messages)
0.00000  49.9625  50.0375    0.500   0.00    0.00  (all messages as %)
0.00000  41.1323   0.0000    1.000   1.00    0.01  T_PQRTW_4

Could it possibly be I just created a killer rule to identify > 40% spam with no false positives? Seriously low scoring spam. Using a single, really short RE? Could it possibly be there is one major spammer out there, that uses this easy to catch finger print on all his spam? And that no one spotted it before…?

Granted, that’s just a tiny pre-flight corpus used for some very basic, fast evaluation. Eagerly awaiting the real mass-check results tomorrow…

Some of the stuff I have been hacking on in the background recently finally has been rolled and pushed on bugs.maemo.org — now that the infrastructure issues have been resolved. Thanks, Ferenc.

Besides some nicening up (A bunch of delimiting dashes with HTML and CSS? Yeah, right…), the reporter and some roles are now clearly stated for every comment. No more scrolling and jumping through the page, just to find out if the commenter is the original reporter. No more mouse twiddling, to see if a commenter is a Nokia employee. You now have that invaluable information right at your fingertips.

Yes, I’ve been spoiled by GNOME bugzilla. Probably the best bugzilla out there. There are a lot of features that just make a triagers life much easier. Supporting the workflow rather than being interruptive. And now we got the first set of bugzilla enhancements for maemo.org.

Hope you will enjoy these first custom features. Got some more tweaks like this and larger features on my list, and started working on that already…

More backlogged news. I’ve been invited and accepted to become a SpamAssassin committer a couple weeks ago. Quite an honor, given there are not many hackers with this status. Even more so, because I did contribute to SA before in various ways, but didn’t yet submit rules or code for upstream inclusion.

I guess, coming up with a few rules that seriously hit low-scoring spam, while resulting in a (less than) negligible amount of false positive hits triggers some attention. The style of arguing and dealing with the developers may have helped, too. Anyway, I am now one of these developers myself. Yay me.

Unfortunately, ever since I’ve been granted this status, I didn’t have the time nor sufficient sneaky spam to develop new rules. Until now. Spotted a pattern yesterday night, and hacked up some rules — turns out, this is a new ratware variant, no hits older than a week. I expect more of these, soon.

Thanks guys (you know who you are). Access to the nightly mass-checks is just awesome to check a rules performance in development.

Consequently, I shouldn’t have mentioned that in public either. Well, I’m not going to talk about fight club (even though it is a great movie, and I finally got the book)…

But myself, my involvement in Free Software and maybe some random thoughts and rants. I’ll try to keep rants low, though. Re-introducing myself, since I have neglected my blog for way too long, here are some quick facts in a nutshell:

I’ve been an active Community guy for a long time, dating back to the old Helix-Code / Ximian days. Bugsquad volunteer, lurker and occasional triager, have been a heavy load triager in the past. Maintainer of GARNOME, the build utility for the GNOME Desktop, member of the GNOME Release Team, and of course GNOME Foundation member.

And recently hired by Openismus as maemo.org bugmaster, together with Andre, to support the community around Maemo. An awesome opportunity, and working with the friendly maemo.org community already has been quite some fun. I’ll add more colorful bits soon while I go along. I promise.

GNOME 2.20.0 has been released. See it. Build it.

Thanks Diego and crew for the Pisco Sour, awesome drink! And a great stunt in the Etap lobby. If you are nearby, you should hurry to get here and join the party. (Also, thanks to the hotel stuff for the support.)

My pleasure, Srini, you’re most welcome.

We are particularly proud of all the hacking and smoke-testing that has been going on during the GNOME 2.19.3 and .4¹ releases². New tarballs have been built and tested by various GARNOMEies almost in real time, as fast as they have been uploaded and as fast as we could update SVN.

Once again, this early testing revealed a number of serious issues with some of the GNOME applications. A bunch of bug reports where filed, resulting in new, fixed tarballs being rolled as quickly as possible — before the official release deadline. Of course, thanks to the developers too, who have been highly responsive.

Our contribution to make even unstable development releases a somewhat sane place to live. Thank you, #garnome!

1. Which, coincidentally, still has not been released due to one particular severe issue, spotted by the GARNOMEies. Bug 449318, immediately worked around by the GARNOME Team, so we could roll 2.19.4 in time. [↩]
2. Yes, Luis, I too love footnotes. [↩]

Managed to negotiate an oral exam (for a later date), instead of the written exam, which was scheduled for the week of GUADEC. Yay.

Hasn’t been tough convincing the prof. Mentioning it’s a developer conference almost was sufficient, he supports things like this. While I didn’t really expect him to know about GUADEC… having to explain GNOME, however, was a little bit embarassing.

Just a friendly reminder – 2.19.3 is due to be released before midnight UTC, today.   (Come on, this is supposed to be fun for volunteers, isn’t it? You know I love you guys.)

On a related note, GARNOME 2.19.3 has just been rolled and is available at a mirror near you.