 A big thank you to Peter Van Roy
started 11/1/2003; 9:36:53 AM - last post 11/2/2003; 12:59:51 PM
|
|
Ehud Lamm - A big thank you to Peter Van Roy 
11/1/2003; 9:36:53 AM (reads: 7624, responses: 2)
|
|
| A big thank you to Peter Van Roy |
|
Peter Van Roy concluded his guest blogging period on LtU. I want to thank Peter for agreeing to guest blog, and for his interesting posts.
Peter raised interesting issues, among them concurrency models and the notion of language expressiveness, and the resulting threads are well worth reading.
I would have been even happier if Peter would have chosen to talk some more about Prolog and Prolog implementation. Maybe next time?
Until then, I am happy to say that Peter's posts were a welcome change from are usual topics of discussion, and that I am sure quite a few LtU readers will now want to read CTM...
Posted to Guest Bloggers by Ehud Lamm on 11/1/03; 9:40:30 AM
|
|
|
|
Peter Van Roy - Re: A big thank you to Peter Van Roy
11/1/2003; 1:59:50 PM (reads: 389, responses: 0)
|
|
|
I would have been even happier if Peter would have chosen to talk some more about Prolog and Prolog implementation. Maybe next time?
Sure, I'd be happy to. Prolog implementation is actually quite simple
and well-understood--the insights are pretty easy to explain.
But in my view the real challenges
nowadays are how to implement more expressive languages like Oz
efficiently and how to do it in an unreliable environment (e.g., with geographic
distribution, partial failure, and security problems). These are worthy topics
for several Ph.D. theses.
|
|
Vesa Karvonen - Re: A big thank you to Peter Van Roy
11/2/2003; 12:59:51 PM (reads: 309, responses: 0)
|
|
|
Prolog implementation is actually quite simple and well-understood--the insights are pretty easy to explain.
You are obviously correct. At least when we limit the audience to those who already master Prolog implementation. ;)
Similarly, I think that basic behavioral OO design, like described in books such as Design Patterns and Designing Object-Oriented C++ Applications Using The Booch Method, is actually quite simple, quite effective and well-understood. Yet, I constantly bump into otherwise intelligent functional programmers that have never understood the striking similarity between behavioral OO design and functional programming. Those functional programmers have been exposed only to the representational school of "OO design" that is more related to database design than behavioral OO design.
|
|
|
|