[erlang-questions] Why Distributed Is So Hard

Michael Turner <>
Tue Aug 31 09:24:21 CEST 2010


Hey Jason, nice essay, but I just wanted to say: before DOS there was this
OS concept called "time sharing."  But don't worry, nothing ever came of it.
;-)

It would be interesting to extend Erlang to evaluate its current "boolean
expressions" in terms of Relevance Logic -- yielding something like "no
information" and "conflicting information" values alongside "true" and
"false".  There might be interesting implications here, for retrofitting
Erlang to selectively provide lazy evaluation, speculative execution and
certain pleasing variations on vanilla exception handling.  But perhaps the
most salutary effect could be "philosophical" in the best (i.e., most
practical) way.  In what ordinarily seems to be the very simplest form of
computation (e.g., "=:="), the problems and potential of distributed
computation would always be at least latent.

-michael turner


On Mon, Aug 30, 2010 at 7:01 PM, Jayson Vantuyl <> wrote:

> Hey everybody,
>
> While it's not Erlang-specific, I know a lot of people here care a lot
> about distributed programming.  I had a really good question from someone
> the other day and it spawned a blog post that might interest you.  In
> particular, most distributed programming in Erlang is usually based on the
> assumption of a highly-available network layer (Mnesia, I'm looking at you).
>  This post talks about why that isn't good enough for distributed, in
> general, and gives a way to think about the problem specifically.
>
> http://is.gd/eLhKF
>
> Thanks,
>
> --
> Jayson Vantuyl
> 417-207-6962 (mobile)
> 
>
>
> ________________________________________________________________
> erlang-questions (at) erlang.org mailing list.
> See http://www.erlang.org/faq.html
> To unsubscribe; mailto:
>
>


More information about the erlang-questions mailing list