Varia:
Sometime ago I started learning Russian; installed Cyrillic fonts etc. Worked fine for a year (but messed up any accented characters on Portuguese web pages (my wife's portuguese)). Now I go to the Debian pages and find their webserver thinks my default language is Russian; amusing, but not always convenient (no, I haven't set Russian to 1st preference in the browser). Then I tried compiling the GTK tutorial file selector, and found it displays a broken mix of cyrillic and Latin chars. Somethings changing - in what direction? I can't find a good how-to on localization for multiple languages: if I get a Russian email, I want to see Cyrillic chars; if I go to a Portuguese or French web site I want to see the right accents on letters, etc. Lot's of how-tos on localization for ONE language around (mostly dated). Is there a group working on localization in general? Something that explains the 'grand plan' would be good. Or maybe there isn't one?

Started reading SICP. It's brilliant, I think it's the best programming book I ever read. Why did I try starting Scheme with the Little Schemer? You need to be brilliant to pick out all the implications in that. Only problem is I think its going to take a year or so to get through SICP. MIT use this with first year undergraduates!? I'm in total awe..

Still trying to sort out Savant. libstdc++2.10 has been removed from the Debian ftp site (they say they're sorting it)... I still seem to have g++ issues. Reminder to self: avoid anything dependant on unstable c++ libraries...