random dreaming

[Raoul Duke]

i wish there were the resources available on earth to somehow get the
clojure stm working in/with/under cal (and scala ;-).

[Tom Davies]

On Nov 21, 10:58 am, Raoul Duke <rao...@gmail.com> wrote:
> i wish there were the resources available on earth to somehow get the
> clojure stm working in/with/under cal (and scala ;-).

Hi Raoul,

I don't think that's too hard -- I have written my own STM for CAL,
and while I suspect that mine was not nearly as performant as
Clojure's, it seemed to be correct -- I implemented the Santa problem
from Beautiful Code and a bunch of other small programs. I hadn't come
across Clojure at the time, so I didn't try that 'ants' example. I
even persisted committed changes to TVars to a BDB database.

My algorithm had a higher isolation level and locked a lot more (only
one commit in progress at a time, for instance).

I've had a look at Clojure's STM, and it's pretty simple (except for
the bits I can't understand :-). Clojure of course has a richer/
different STM interface than Haskell -- with commutative operations
and actors, for instance. I think you could plug a hacked version of
Clojure's STM into CAL in a few days, ending up with Haskell-style
operations but Clojure's MVCC.

Doesn't Scala already have STM? How good is the implementation, and
how does it compare to Clojure and Haskell?

Tom

[Raoul Duke]

> Doesn't Scala already have STM? How good is the implementation, and
> how does it compare to Clojure and Haskell?

i think there are a few "competing" stms for scala, and nothing that
is yet a de-facto standard there.

sincerely.

[Tom Davies]

This looks like an interesting STM implementation:
http://code.google.com/p/multiverse/wiki/Features and would be easier
to integrate with CAL than Clojure's STM would be.