Sounds like a fancy shell program.  I have considered temporal state
machines which recognize temporal state machines?  I talked with one person
who had researched them in europe.  I think they may be handled in
augmented transition machines.

You may want to look at promises as well.  I think this is handled in
languages like e and maybe erlang.  Promises are in the grammar of e, which
I think is what you are suggesting.  Other languages implement promises as
a library--this may not be as clean.

I've also seen people add differential equations to state transition
machines.

You're probably looking for links, and I just provided keywords.
On Mar 14, 2013 8:55 PM, "David Goehrig" <[email protected]> wrote:

> I was wondering if anyone on the list is aware of research of adding
> temporal conditionals (pre/post) to PEGs to generate context sentiments
> grammars capable of modeling transitions of a state machine.  Has anyone
> explored using a set of preconditional or postconditional predicates to
> modify a given pattern?  Also, has any looked at using multiple matches to
> mimic concurrency or parallelism?
>
> I was thinking of it in terms of a washing machine example, say a washing
> machine operation grammar defined a language as:
>
> Precondition <- Predicate.     # defines a match only when the left hand
> side has been matched, <- is  read precedes
>
> wash -> fill with water.
> wash -> release soap.
> wash -> agitate drum.
>
> Insert quarter {18} <- wash.      # need $4.50 in quarters to start
> machine {18} repeats
> Select cycle <- wash.
> Lock door <- wash.
>
> wash 20 minutes <- spin.     # can not enter spin till we see wash 20
> minutes on input stream
>
> If you think of a program as a sequence of messages passing by the PEG
> producing matches over time, you could see how nicely this can map to
> certain classes of difficult distributed programming problems.   Some tasks
> which can happen in parallel like filling the drum with water, releasing
> the soap, and spinning the drum, are naturally parallel if no order is
> imposed by the rule engine.  Preconditions provide a way to ensure
> dependencies are met prior to a match being found.  I would love to see if
> anyone did any research into this sort of application.
>
> Dave
>
>
> --
> -=-=-=-=-=-=-=-=-=-=- http://blog.dloh.org/
>
> _______________________________________________
> fonc mailing list
> [email protected]
> http://vpri.org/mailman/listinfo/fonc
>
>
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to