Doug's musings

Computing

[2003-06-25] The insomniac and the PowerBook that wouldn't wake up ::

I was awakened at 2 am by some loud and apparently drunk young women talking loudly outside my apartment windows. I couldn’t get back to sleep, so I went to the computer. Unfortunately it didn’t want to wake up, or even boot after what should have been a hard reset. I grumbled to myself that the computer was pulling a Rip Van Winkle on me, as if to make up for my not getting any sleep. I somehow stayed awake, played keyboards, did some reading... At 6 am, I called a hardware geek friend on the East Coast. He told me it sounded like the Power Management Unit (PMU) needed to be reset. I agreed, remembering that a couple of my previous PowerBooks had such a button and had needed it pushed on occasion. I couldn’t find any such button on my PB G4, however. I went into work early. Eventually 9 am rolled around and the repair department at work opened. I presented my baby to a woman who seemed to be restraining a smile for some reason. She disappeared around a corner for 30 seconds and returned to show me my PowerBook booting. I felt really glad that at 6 am, I hadn’t gone digging around for a Torx screwdriver with which to perform exploratory surgery. ...

(268 words)

[2003-05-15] SpamBayes and Mail's Junk filter ::

Via Jon Udell, Bayesian Nets, Latent Semantics, Despamming and other speculations: a fascinating (well, at least for programmers and mathematicians ;-) comparison of how Bayesian nets differ from latent semantic analysis. While I don’t believe Apple has said anything about how Mail’s junk filter works, it does seem to use some form of latent semantic analysis. I’ve noticed that it doesn’t catch spam which doesn’t contain many recognizable words (e.g. base64-encoded and all-HTML messages). I do seem to have trained SpamBayes enough for it to be catching such spams, however. The combination of Mail with only a minimally-trained SpamBayes seems to be catching almost all of my spam now, and no false positives yet...

[2003-05-13] SpamBayes and resonances ::

SpamBayes: Bayesian anti-spam classifier written in Python. Apple’s Mail is doing a pretty good job at detecting my spam, but a lot of offensive stuff is still getting through. So I’m giving SpamBayes a whirl. It sits as a proxy between your mail client and server, and just adds notes in the headers of incoming mail. Let’s let it train for awhile... ...

(555 words)