[erlang-questions] Erlang and Akka

[Ignas Vyšniauskas]

Funnily, I was just thinking about the semantics of Erlang and how
ad-hoc receive patterns are major blocker for formal reasoning and
also impedes performance / predictability. As a consequence, I wrote a
very simple model of async pi-calculus in Erlang, with similar
semantics to the one proposed in this paper (I also added half-baked
channel delegation support to allow dynamic communication patterns).

Here's the code:
[1]: https://github.com/yfyf/pierl

--
Ignas

On Mon, Sep 1, 2014 at 12:36 PM, Richard Carlsson
<carlsson.richard@REDACTED> wrote:
> Some of the ideas presented in that paper became the basis for a type
> analysis (without source level declarations) that I wrote with the intent of
> providing more information to the compiler. Then Kostis Sagonas and Tobias
> Lindahl took the ball and ran with it, and we ended up with Dialyzer.
>
> I remember that Sven-Olof Nyström also worked on typing of processes, but
> the main obstacles were (as Thomas Lindgren also mentioned) the
> overapproximation of which PIDs will receive which messages, the mixing of
> messages for different purposes in a single mailbox, and that the type for a
> process needs to describe what messages are acceptable at what times in
> order to be really useful. I can't find anything published about this
> research. He did however write up some of his ideas about how Erlang could
> be improved, partly in order to get more precise information about
> communication patterns:
> http://www.it.uu.se/research/publications/reports/2009-005/
>
>
>         /Richard
>
>
> On Mon, Sep 1, 2014 at 12:55 AM, Anthony Ramine <n.oxyde@REDACTED> wrote:
>>
>> Found this:
>>
>>         http://www.erlang.se/workshop/2003/paper/p56-nystrom.pdf
>>
>> --
>> Anthony Ramine
>>
>> Le 31 août 2014 à 23:47, Richard A. O'Keefe <ok@REDACTED> a écrit :
>>
>> > Is there a reference for that?
>>
>> _______________________________________________
>> erlang-questions mailing list
>> erlang-questions@REDACTED
>> http://erlang.org/mailman/listinfo/erlang-questions
>
>
>
> _______________________________________________
> erlang-questions mailing list
> erlang-questions@REDACTED
> http://erlang.org/mailman/listinfo/erlang-questions
>