Denys Duchier writes:
> Isn't the claim to fame of AOP that it permits a _modular_ yet
> cross-cutting way of addressing aspects? We don't particulary address
> this issue of modularity. There are many applications of AOP about
> which we have nothing to offer. What we have is probably a better
> foundation for e.g. adding fault-tolerance to distributed
> applications. I don't think we solved AOP. We just have better
> solutions to offer for certain problems that AOP attempts to solve.
>
When I first saw this thread I was somewhat sceptical,
but that was largely due to the justification
that was given.
However, I am now starting to think
that concurrent constraint programming,
particularly the Oz model, can be used to do AOP in the above sense.
One can implement "aspects" as constraints with their own threads,
and the Oz model gives you the "weaving" for free.
The AOP folks do not have a definition of "aspect" and "weaving"
as far as I know. Their idea is just that all the aspects should be
written separately from each other and from the weaver,
which should ideally be fairly generic.
Concurrent constraint programming seems to provide for that.
Daniel
-
Please send submissions to hackers@mozart-oz.org
and administriva mail to hackers-request@mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.