Blog: Archive for February, 2005

NPR had a story about the release of the final volume of Christopher Alexander's Nature of Order. There's also an audio feed of the interview available.

The AMS published “Foolproof: A Sampling of Mathematical Folk Humor” (PDF) by Paul Renteln and Alan Dundes, collecting and analyizing a body of jokes and other folklore associated with the mathematics profession. Some of these are way, way beyond me (to be fair, some of them are beyond some of the mathematicians in the department, too—some of them are quite discipline specific), but many of them are flat out hilarious, especially the longer stories featuring mathematicians, physicists, and engineers, one of which I will excerpt for you here:

A mathematician, a physicist, and an engineer were traveling through Scotland when they saw a black sheep through the window of the train.

“Aha,” says the engineer, “I see that Scottish sheep are black.”

“Hmm,” says the physicist, “You mean that some Scottish sheep are black.”

“No,” says the mathematician, “All we know is that there is at least one sheep in Scotland, and that at least one side of that one sheep is black!”

I spent most of Sunday cleaning up my Debian development machine. Most of that time was spent going through the 45,000 (seriously) messages in my spam folder to make sure that there wasn't any real mail that had gotten swept up by the filters.

There was, of course, most of it probably due to my having accidentally blacklisted a message or two when I was looking through my inbox a couple of hours after I should have been asleep.

I was impressed by how well “real mail” stands out from spam, especially if you sort the mail by subject. In the end, I used pick to pull out the messages I could identify as having been sent to a list (about 10%), picked out all the good mail I could still find by hand, then went through the 4500 list messages to get rid of the spam there, cleaned the legit mail of its SpamAssassin labelling, and ran it all back through procmail to get it into the folders it should have ended up in.

Overall, I think that the blacklisting functionality isn't very useful—most of the spam I get is presumably from random machines using other people's addresses (that's certainly true for most of the spam I get that purports to be from other Debian developers). I haven't taken a look at the code in mh-e to see how it's implemented and see if it's possible to turn off the blacklisting. In the meantime, I just go and manually delete the blacklisted addresses from my configuration file.

Spam sucks. It's times like last night that I think that disembowling spammers is maybe just a little too humane.

I'm struck by the fact that so many articles and books about free and open-source software waste several paragraphs or entire chapters on building the programs from source. Generally speaking, I'd say that if you can't figure out how to build a program from source, you shouldn't be building it—you should be installing a package from some reliable source (maybe even from upstream, as many upstream developers now provide packages or links to packages built by others).

Even if you grok the building-from-source thing, you should still probably be building a package rather than just compiling and installing, as packages are way easier to manage, and, perhaps more importantly, remove.

There are still a handful of things I build and install without packaging (for example, CVS GNU Emacs), but anything that I'm planning to use long-term, or deploying on more than one machine, I package.