[e-lang] Proposal: E / Erlang integration

[Mark S. Miller]

Ulf Wiger wrote:
> You could read and perhaps comment on my presentation at EUC 05:
> 
> Structured Network Programming 
> (http://www.erlang.se/euc/05/1500Wiger.ppt), where I take a basic 
> telephony problem ("complex concurrency") and implement it both using 
> selective receive and in an event-driven, non-blocking fashion. One of 
> [...] Mike Williams, Bjarne Däcker and others 
> wrote about this already in the late 70's, I believe (at least in 
> internal Ericsson documents.) However, it's not that well covered in 
> teaching material.


I am overjoyed to encounter a realistic challenge problem. Thank you! However, 
I am not able to understand it well enough from these slides to respond. I 
have googled "Mike Williams, Bjarne Däcker" and found many hits, but it's not 
clear which if any of these are relevant. Do you know of any documents 
explaining this or a related problem in enough detail that I might try 
expressing a solution in E?

Btw, I am not claiming that non-monotonic[*] selective receive is never 
useful. In my experience so far, I have found it useful only once, but that 
one time I did find it essential. My claim is only that it is the wrong 
default. Unfortunately, this weaker claim is also much harder to evaluate. 
We'll see if this challenge problem helps us do so.

[*] E's promises provide a limited monotonic form of selective receive, which 
often provides the synchronization that one needs from selective receive, but 
with reduced danger of deadlock. If this turns out to be relevant, I will 
explain further.

-- 
Text by me above is hereby placed in the public domain

     Cheers,
     --MarkM