|
⇤ ← Revision 1 as of 2005-08-21 15:35:12
Size: 923
Comment:
|
Size: 929
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 9: | Line 9: |
| - Threaded programming -- still hard. Why? - Deferreds -- Twisted, what benefits have been gained thru this experience? - Generators, Stackless, Greenlets -- new control flow structures - Multi-process -- So many methods to do IPC... |
* Threaded programming -- still hard. Why? * Deferreds -- Twisted, what benefits have been gained thru this experience? * Generators, Stackless, Greenlets -- new control flow structures * Multi-process -- So many methods to do IPC... |
| Line 14: | Line 14: |
| Could be interesting. Just a suggestion. --blais | Could be interesting. Just a suggestion. --blais |
Ideas for Panels.
Panel on Concurrency
Given the amount of debates surrounding concurrency models, apparent misunderstanding of threads as a necessary model for concurrency, and the fact that language-level features may provide new opportunities for concurrent programming (i.e. generators), why not setup a Panel with some key opiniated people, to discuss concurrency models tradeoffs and heat up a bit of discussion?
I'm thinking about the following approaches, and selecting a person to defend/represent viewpoints w.r.t. to those, with maybe a 10 min presentation each, and then some discussion and questions:
- Threaded programming -- still hard. Why?
- Deferreds -- Twisted, what benefits have been gained thru this experience?
- Generators, Stackless, Greenlets -- new control flow structures
- Multi-process -- So many methods to do IPC...
Could be interesting. Just a suggestion. --blais