geekin it up in steinbeck town

[substitute]

So, I'm almost done with USENIX 2002 here in lovely Monterey. Tonight was the dessert reception at the aquarium (yes you read right). At one point I was eating an eclair while one hirsute hacker near me disputed with another about the relative virtues of assembly programming on the VAX versus M68K as a large grouper and two small rockfish looked over their shoulders and moved their fishy mouths as if to disagree with both of them.

I took some pictures mostly of fish and one of some famous hackers, which are famous if you're a geek like me.

Tomorrow, after a couple of last sessions, it's off to San Francisco for a WEEK of UNINHIBITED HEDONISM. Woo-hoo!

Comments

Roy Fielding

[flipzagging.livejournal.com]

And still the man is not finished yet. You should check out Roy's PhD dissertation - now known as "the REST paper (http://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm)". In certain circles it's caused a minor earthquake.

The REST hypothesis is that URIs and HTTP are sufficient to encompass any desired computational communication pattern. Furthermore, this style has desirable characteristics: standards-based, can be made reasonably efficient, can be made secure, requires no new elaborate hardware or network building.

The immediate implication is that the technologies du jour -- SOAP, and complex RPC-like layers on top of the web -- are both Bad and Wrong. You already knew that they sucked, but this paper will show you why they inherently suck.

The paper is really long, so you might want to check out some of Paul Prescod (http://www.prescod.net/)'s popular articles on the topic.