read-only vars and speculative bindings

[Raphael Collet]

> The original motivation for futures was to have "safe placeholders".
> I had not realized this before but to me "read-only" and "safe
> placeholder" carry different contractual implications.  I have
> different intuitions about them.

I have never seen a clear and precise definition of futures, BTW.
It looks like everyone has its own definition and motivations.

> For "read-only", my intuition is that allowing speculative bindings
> would be a double contract violation:
>   - it's read-only, therefore you are not allowed to bind
>   - you are supposed to suspend when synchronizing on a read-only, but
>     speculative bindings would allow you to continue making progress

I agree.

> I had no such intuition about "safe placeholders".  Somehow the
> concept seemed reasonably friendly to the constraint-based view and
> the notion of speculative bindings.
> 
> I'll admit that the one jarring point was that the synchronization
> behaviour was inconsistent between home and sub spaces.  So, perhaps
> that is good enough a reason to adopt the stricter interpretation of
> "read-only".

I am in favor to this interpretation, too.

Cheers,

raph
-
Please send submissions to hackers at mozart-oz.org
and administriva mail to hackers-request at mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.