Hi,

thanks for your remarks.

@Timothy: Yes, looks like I mis-understand the main motivation for 
core.async or I'm just not used to see my programs as an assembly like 
sending data boxes around ;-)

On Tuesday, September 3, 2013 4:20:19 PM UTC+2, David Nolen wrote:
>
> Which also creates a resource problem as a result of the subscription side 
> effect. You can build subscription based on Rx over core.async if you like, 
> but I think it's more of a tradeoff than people realize. 
>

Correct, the producer can easily outrun the consumer in Rx ... appears to 
be somewhat similar to holding the head space-leaks in lazy sequences, but 
probably much more subtle to detect and fix. 

>  
> Problematic in what real sense? Haskell provides similar facilities over 
> the Chan data type. Clearly they also believe it's useful to model problems 
> in this way.
>   
>
As far as I know, Haskell has Chan data types in its concurrency 
extensions, but I have never seen them in FRP. Maybe this means that FRP is 
addressing a different problem. On the other hand, the Automaton Arrow can 
be used to implement state machines that transition on receiving inputs in 
a composable and referentially transparent way ... but they still have some 
problems regarding efficiency and space-leaks.
I was just wondering, how core.async relates to FRP and like that the 
latter allows for equational reasoning about your data flow. In the end, 
side-effects have to occur somewhere since callbacks/event handlers/etc. 
can never return values, so FRP has to deal with similar problems 
(currently it looks like the higher abstraction provides less control about 
resources).

Best,

    Nils
 

> David 
>

-- 
-- 
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your 
first post.
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to