Re: [abc] abc release

On 25/07/07, Oege de Moor <Oege.de.Moor_at_comlab.ox.ac.uk> wrote:
> > I did not apply any AspectJ versions, just tracematches.
>
> I think AspectJ numbers would be very interesting, and easy to obtain?
> It's especially interesting for your work, since there is no chance
> of applying these static analyses to improve the AspectJ version.
> So you could make the point that a hand-coded aspect is sometimes
> (or even often?) slower than a mechanically optimised tracematch.
>
> The gold standard failsafe iter aspect is in cvs.

Yes, you are right (and I now remember you told me before). I should
definitely do that for our camera ready copy / next submission.

> Hm, the time for a journal paper is when the work has reached some
> stability. I don't see that yet, there's still tons to do.

Ok, fair enough.

> Thanks, we'll see what that throws up.

Great. Let me know how it does.

Eric

-- 
Eric Bodden
Sable Research Group
McGill University, Montréal, Canada