graydon linked me to this note about distributed computing. And he oughta know, 'cause he spent a lot of time on Berlin (aka Fresco, which was actually referenced in this paper).The whole thing feels relevant to a lot of stuff I've written about here, even beyond the obvious RPC comparisons. I kept thinking of all of the arguments about the wire format for these blogging protocols, and especially that article that criticized the simple-text LJ protocol for "escaping issues". There's been talk on the Atom lists for sending deltas efficiently, as if figuring out what to rename all the RSS nodes to* was the hard problem and tacking on distribution will be easy.
The hard problems in distributed computing are not the problems of how to get things on and off the wire. The hard problems in distributed computing concern dealing with partial failure and the lack of a central resource manager. The hard problems in distributed computing concern insuring adequate performance and dealing with problems of concurrency. The hard problems have to do with differences in memory access paradigms between local and distributed entites. People attempting to write distributed applications quickly discover that they are spending all of their efforts in these areas and not on the communications protocol programming interface.
* I'm kidding (on the square).