mike’s wondering how i got my rss feed. well, it’s very much homegrown, but, inspired by blogify. basically, i’m using dublin core metadata in the pages for the channel description and marking up the text with divs and spans. xpath and a little perl do the rest. “conflux style” posts are marked-up differently than “machina style” posts and, as jon udell noticed, i rather unelegantly truncate the descriptions for scannability. when i wrote the script, i think i remember trying to get Lingua::EN::Sentence to elegantly pick off sentences, but my horrid punctuation and general disregard for capitalization made it, um, less than reliable. i should probably look back into it, since i agree that’s it’s a bit jarring the way it currently is implemented.
so, here’s xhtml2rss with the usual, “you break it you get to keep the pieces” caveat. and i know it ain’t exactly the purtiest thing in the world [ why exactly do i clean the text twice? ], but i don’t want to hear from some snot-nosed kid who can compact the whole thing into a one-liner.