[PRL] may be

[Karl Lieberherr]

Hi Matthias:

A system (with an efficient checking algorithm) that checks a class graph
where the nodes are labeled with "acquire x:X" or "[acquire x:X]" statements
([ ] means optional), for type correctness under a suitable default notion
of acquisition is not hard. We should probably discuss the options. I think
that the data acquisition case is particularly useful.

-- Karl

-----Original Message-----
From: Matthias Felleisen [mailto:matthias at ccs.neu.edu]
Sent: Wednesday, November 05, 2003 2:41 PM
To: lieber at ccs.neu.edu
Cc: PRL
Subject: Re: [PRL] may be

Karl, I didn't object to the idea of "fixing" the may-be formalization.
What I
heard you say was that we could easily come up with a type
inference/checking
system using this may-be system, i.e., a sound type system was
Mitchably fixable.
And that's what I objected to. -- Matthias


On Wednesday, November 5, 2003, at 02:32 PM, Karl Lieberherr wrote:

> Hi Matthias:
>
> I did not have a chance to respond in person. Here is my fix to the
> "may-be"
> formalization:
>
> A B-object may be contained in a C-object iff the traversal
> specification
> [C,B] defines a non-empty set.
>
> Traversal specifications are defined in:
>
> @ARTICLE{lieber-palsberg-xiao94,
> AUTHOR = "Jens Palsberg and Cun Xiao and Karl Lieberherr",
> TITLE = "Efficient Implementation of Adaptive Software",
> JOURNAL = toplas ,
> YEAR = 1995,
> PAGES = "264--292",
> MONTH = mar,
> VOLUME = 17,
> NUMBER = 2
> }
>
> The may-be relation is covered in my undergraduate class ((<=.C.=>)*.<=
> where . is relation composition). Maybe I misunderstood your statement
> that
> the may-be relation is NOT easy to formalize.
>
> -- Karl
>
> _______________________________________________
> PRL mailing list
> PRL at lists.ccs.neu.edu
> https://lists.ccs.neu.edu/bin/listinfo/prl