Technology

Note on the below — “NetNewsWire”: already produces OPML that tags feeds as type “rss”. Reverse compatibility nearly requires others to follow suit; I’m sure NetNewsWire is not the only application that does this. However, the more I think about it, the more I think it’s a good idea to distinguish between multiple feed types. I would like to know if a feed is Atom or RSS before I grab it. Saves time, saves CPU on both ends of the transaction, saves network, and so forth. ...

Obscure technical quibble of the day follows. Warning: technogeeking ahead. Dave Winer’s OPML sharing guidelines (original) are a little wonky. Point 1 says: If an element is pointing to a feed, set its type to “rss”. Do this even if it’s not an RSS feed. Nope. Set the type to “rss” if it’s an RSS feed. Set it to “atom” if it’s an Atom feed. If you want a generic type name for a feed, I’d suggest “feed”.

L’il Bolt’s training is coming along nicely, but I fear he lacks the fire necessary to — Oh, sorry, wrong kind of sidekick. I’m still pretty happy with this Sidekick, with some caveats. The software bugs which prevent it from ringing when Keyguard is enabled are annoying. The battery life is short enough so that I have to recharge it every night without fail. Also, my thumb wheel broke and I had to get the entire unit replaced. All in all, I’d say it’s about 85% of the way there and I still recommend it for people like me who don’t mind being early adopters. Danger is currently claiming that a software update to address known bugs will show up sometime in March, but that’s not a firm date.

Coming this fall: TiVo to computer functionality (original). It’s not what I’d like, since you don’t get full functionality — it’s some encrypted video format. Mind you, DRM is generally broken… but in any case, it’s more than we had before. I would very much like to be able to easily copy programs from my TiVo to a DVD. Cory Doctorow, bless him, is up in arms (original). However, I think his analogies suck. While TiVo is a disruptive technology business, it is not much like steam engines. It’s a different delivery mechanism rather than a new media form. ...

Well, mt-blacklist lasted a good month or so. I’m now getting hit by a clever guy who figured out that by varying the capitalization of his URLs, he can get past the simple filters. Course, I can complicate the regexp some, but it’s still fragile. Second order solution is the distributed spam database, which works pretty well as these things go.

Funky stuff. Nick Olejniczak is teaching a class on weblogs at the University of Wisconsin next spring, and he’s put the syllabus up on a wiki (original). He’s using the wiki to gather community thoughts on the syllabus. Darned cool.

Did you ever read Usenet? Miss those newsreader days? Population: One is now available via NNTP. It is perhaps a bit quixotic of Dan to scrape weblogs rather than slurping up RSS feeds, but it’s a kind of quix I have to admire. The old diesel engine of blog syndication, as it were.

1&1 is offering free Web hosting for three years, which certainly seems too good to be true. But hey — Metafilter confirms that it’s a valid offer. Apparently these guys are a huge European hosting company, and they want to make a splash in the US. So it seems safe to go wild.

RFID just succeeded (original). When Wal-Mart wants manufacturers to use a technology, they will use it — Wal-Mart’s leverage is just that big. The privacy implications are horrendous (original), but the applications are fairly neat.

CafePress whiffed badly on book pricing (original). The base fee is five bucks per book for saddle stitch or wirebound, and six bucks per book for perfect bound. On top of that, they’re charging six cents per page, and that’s printed page rather than physical page. A 250 page paperback would have a base cost of $21. A 32 page comic book would have a base cost of $6.60. (Er. $6.92. I dunno where I went wrong.) Profit margins are optional. I don’t think this’ll take off. ...