[erlang-questions] Teaching Erlang as part of a paper -- advice sought

Michael Turner <>
Tue Feb 9 10:13:49 CET 2010



On 2/9/2010, "Ulf Wiger" <> wrote:

>I solved the race problem on registration by only allowing a
>process to register itself. There is still a race in the sense
>that two processes may contend for the same name, or that a
>registered process may die before a "registered send" is delivered,
>but this is, I think, an unavoidable part of the semantics of
>referring to processes by name, combined with dynamic typing
>and asynchronous message passing.

*Sigh*.  As a hack, I've been relying on the process registry for
something -- luckily in a way that's not going to mean ongoing
performance bottlenecks if I serialize the name registrations, since the
registrations are restricted to a brief initialization phase.  However,
the realization that I was using well-known race-condition code sent me
on a literature search that serendipitously turned up this:

  "Fallacies of Distributed Computing Explained"
  http://www.rgoarchitects.com/Files/fallacies.pdf

To the extent that the course is about *distributed* computing, not just
concurrent computing, structuring the discussion around the now-classic
Eight Fallacies could be useful.

-michael turner


More information about the erlang-questions mailing list