'any' and 'all' compared with the rest of the Report
Thu, 25 Jan 2001 10:40:49 +0100 (MET)
>There is also the problem of error behavior; if we unfold an "or"
>eagerly to exploit its data parallelism we may uncover errors, even if
>> or [True, error "This should never be evaluated"]
>My current work includes [among other things] ways to eliminate this
>problem---that is, we may do a computation eagerly and defer or
>discard any errors.
What you basically have to do is to treat purely data-dependent errors (like
division by zero, or indexing an array out of bounds) as values rather than
events. Then you can decide whether to raise the error or discard it,
depending on whether the error value turned out to be needed or not.
You will have to extend the operators of the language to deal also with
error values. Basically, the error values should have algebraic properties
similar to bottom (so strict functions return error given an error as
argument). Beware that some decisions have to be taken regarding how error
values should interact with bottom. (For instance, should we have
error + bottom = error or error + bottom = bottom?) The choice affects which
evaluation strategies will be possible.