[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: CSP-OZ Backtracking.
My email program obscured what I was saying and what he said.
Sorry. Try again....
> The point is in the implementation (Java) version of the
>specification, we will
> need this kind of behaviour.
> My point was that even though CSP doesn't specifically include
> back-tracking as part of its model, you can get that effect with
> suitably contrived inter process events. Getting it right, without
> some sort of transformation tool might be tedious, but should be
> possible. You're going to have to determine exactly what sequences of
> events comprise a backtrack, and program it that way.
>This is the point but I am not tring to find or derive some new
> I suspect that if you articulate what you mean by backtracking,
> a set of tranformation rules will become apparent. Such
> transformation rules would guide you to the right set of "contrived"
> events that implement your back tracking in CSP. Thats all I
> meant by a new CSP theory. -jc