Another weird idea about intrusion detection crossed my mind.
Charmed is somehow full of flags these days. Not that it was worth anything to start with apart from a couple of attractive details.
Finished a few more books, replaced another fried power supply in a pc "server" and more life as usual.
I'm still reading The Art of Unix Programming. Funny I hadn't started before. It's an excellent book especially in contrast to all this hype and microsoftist creep of today.
Seems Firebird on Irix is quite bearable for a browser. And for that sacrifice, you get the usual benefits of these behemoths.
But the truly wonderful thing is, Mozex finally allows me to have view source and textareas editable in an external editor (gvim -f)! Maybe this would help rid me of my newfound hate toward wiki editing.
I made a little macro set that lets me write documents to be published in html with minimally annoying syntax.
So much else is going on. Back to work.
Got a new pet -- an Octane. Happy me.
Rewrote my Irix tardist building environment to use make. It doesn't seem to help too much. I should experiment a lot more, maybe publish something (if only sf wasn't so bad) and look at all the other make building systems like pkgsrc.
Toyed a bit with ocaml.
Someone got me looking at GStreamer. It seems very nice apart from linux-centrism in the docs and pages and some weird windos-envy. Then there's the state of the documentation and thread problems.. Still, looks like a great framework.
Some very nice and linkful web pages, especially Wiki's, can lead to me doing something like a very long depth-first search. When I hit the bottom of a branch, I should backtrack. The back button does something and maybe I even remember where I was in the previous page.
So I started wondering if it would be nice to have some functionality in a browser to help in this. First idea was to add an item to a link's contect menu that pushes it to a stack of "read this" pages (unless it's already there?) and I can then walk that stack when I have time. Well, it should be a bit more complex than that, but just accumulating bookmarks or windows/tabs is quite tedious.
Clearly that would require some more thought and time to try stuff like bookmarklets or hacking dillo to test approaches.
Besides diving into C2 again, I noticed one WikiClone doing a form of FacetedClassification. That spawned more ideas. Must try to write more stuff down outside my LogBooks.
Did a small scheme eval window type thing in FLTK to learn that and get some exercise. Great fun.
Quite a bit of downtime on the Elisa DSL once again. Nothing big, though.
Installed debian on a pair of Sun LX50 boxes and set up RAID1 in the process. Twice. The first try ended up totally dead and the second try sometimes hiccups at boot. There must be something I managed to mess up or something funny with the kernel that makes it sometimes refuse to start raid pairs.
Pretty nice machines all in all. SMP, serial redirected everything, very fast SCSI buses and quite nice construction in 1U. Heavy price tag though.
http://people.debian.org/~blade/install/ was a nice find.
A bona fide lucid dream entered from a repeated scene recognised as a dreamsign plus a false awakening! Finally I managed it. It's amazing how very few things surprise you when you're asleep.
And back to earth, this companion in life of mine has showed her scheming side (and I don't mean the language) again.
It's getting really hot in here. Better get back to reading Cryptonomicon outside.
Mr Herbert handed me a package of Dillo for Irix. Man, that thing *flies*. Shame about the messy macros that bother compiling it with mipspro, though.
Kolumbus/Elisa has surprisingly been giving me an internet connection without interruptions for a few days. Whee! :)
I've gotten more Mac stuff assigned to me the weird platform person and I'm getting more and more iffed looking at Apple. Systemwise OS X is missing everything that would be half-useful and the sites and docs are horrible and nothing seems to work in a sane way. And I don't mean differences, I know the system and I knew NeXT as well.
I got around to looking at FLTK again. There's many little crude bits and of course it doesn't use the system toolkits which has a few downsides (like no Indigo Magic), but it looks very interesting. There is some IM look in the clock and file requester, though. Which of course is nice.
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!