learned how to do content creation in Plone with Archetypes today. holy crap! they make it really easy. i'm optimistic about the future of our web project with Plone; it actually kind of excites me about writing web software again.

research server needs super-new mysql. joy! oh wait, it needs perl too. oh wait, the mysql.com RPMs suck dirty ass. Unsuccessful stat on filename containing newline at /usr/lib/perl5/5.8.0/ExtUtils/Liblist/Kid.pm line 97. hmm.

installed/configured ldapv3 for some new servers, which was unbelievably time-consuming and painful. was bitten in the ass, as it turns out, by failing to stop/restart sshd when i changed pam stuff - apparently it doesn't pick up those changes on its own (thanks seth). anyway, now i can run away from both nis and hesiod as fast as my little legs will take me.

next project: determine how best to export home directories, since we've got all the centralized auth* going for them.

dan forwarded on this ala article about footers to me. i'm so excited to see standard not-ugly ways to do css page design emerging.

on a side note it's really discouraging to see that many of the freelance web designers around here are still stuck on table-based layouts. i had one (recommended to me by a friend) who was apparently totally uninterested in doing anything that she couldn't spew out of dreamweaver. now call me crazy, but when i do art things, music or whatever, i get really pissed off when i can't get down and create at a high level and then go in and tweak small pieces by hand easily, and you just can't do that with table-based layouts; i guess these people are either afraid of new tools or afraid of raw html, or both.

wxdu is now webcasting in ogg vorbis! this means i can record my show with a cron job :-D .

configured nagios for several of my systems this morning. it's very impressive; the setup is complicated enough, without being too complicated, and the functionality is fairly good. there are a lot of "sample" entries in the sample configuration files that are pretty much exactly the same, though, which made for a lot of irritated ^K-ing. also they should have at least provided some examples of configuration for groups of machines that have identical settings (like cluster nodes). still, the documentation is pretty good so far.

it's good to be snowed in sometimes.