zaitcev.mee.nu

We had 3 meetings of CS 591 with Prof. Dorian Arnold so far. The first two were rather dull, the third more fun. At first I pined for Prof. Bridges' class, but now I expect the class to get better once we're done with the basics.

One notable thing about the class discussion is that Prof. Arnold has a habit of explaining things in extremely general manner, covering all eventualities and using the broad language only. He says all the right things, but I have a strong suspicion that my classmates may have trouble with mapping the concepts that he discusses to the reality. Granted, if he simply says that non-consistent cache is just like DNS records (governed with TTL), students will think of that example whenever caching comes up, and perhaps miss some other ideas. But this may be an acceptable price for knowing at least one example.

The whole session reminds me about the story of Richard Feynman in Brazil, a bit. Light reflected from the surface of the bay, triboluminiscence, etc.

Another downside of abstract conversation is that I am not sure I understand it all myself. For example, when Prof. Arnold says "overlay network", I think "a bunch of IP-IP tunnels between some daemons" even before he ends the sentence. But is that right? Maybe he means something entirely different!

Papers weren't difficult so far, but I'm already slacking: made a note to read up on 2-phase commit protocol and didn't do it. Tsk, tsk. The next one is already Lamport's thing about timing (that guy didn't just invent LaTeX, you know). This is going to be fun.

Posted by: Pete Zaitcev at 10:53 PM | No Comments | Add Comment
Post contains 278 words, total size 2 kb.