Hi Fellow Stackless Pythonistas:
Pycon Canada 2012 is winding down. Sunday afternoon, I gave the talk "How to
Solve a Problem Like Santa: Prototyping Join Patterns with stackless.py for
Stackless Python." The talk and the corrected slides will soon be online. At
the sprint, I am fixing up the code base so I can put it online (yes, I know I
know). I am also trying to create a more end-user friendly API.
My talk was modestly attended. After all, it is a strange talk in that not only
do I have to bring one up to speed about Stackless Python, but talk about how
it can be extended. During the Q/A, only one person asked a question:
essentially have I used Go? What do I think of it? And why don't I just use
that? It was a rhetorical question. My answer: "I'm interested in making Python
(especially Stackless) better." My approach is to develop the feature as if it
were to be included - this provides a technological constraint. More
importantly, it is good to rise to challenges. After all one may learn
something new along the way. In my case, I saw a lot of literature concerning
STM and lock-free concurrency strategies. This helps in following Armin Rigo's
amazing STM work.
Funny, the join pattern model with some changes ought to work in Go too ....
Off-line, folks liked the talk. One person got a kick seeing Limbo code.
Another person thought the talk was going to be another using concurrency for
distributed load-balancing sort of thing.
There was a lot of buzz around gevent. And a desire to understand
the underpins.
Once again, it is a pleasure to talk about Stackless Python and do cool things
with concurrency. This may be the best job I have ever had! With stackless.py,
The JIT, continuelets and STM, I really believe wonderful things are in store
for 2013.
Cheers,
Andrew
_______________________________________________
Stackless mailing list
[email protected]
http://www.stackless.com/mailman/listinfo/stackless
