[PPL-devel] Re: PPL library

[Roberto Bagnara]

Sriram Sankaranarayanan wrote:
 > Dear Prof. Bagnara,
 >
 > I am a student of computer science at Stanford University and
 > have been working on static analysis with Prof. Zohar Manna.
 >
 > We just released the binaries and examples for our prototype analyzer,
 > which we describe in our upcoming SAS 2004 paper.  The analyzer works
 > over the polyhedral library PPL developed by your group. A *draft* of
 > the project's page  can be found at
 >
 >   http://www.stanford.edu/~srirams/Software/sting.html
 >
 > Your comments on the draft are welcome!
 >
 > First of all, we would like to thank you and your team
 > for developing the PPL library.  It is very fast, very
 > easy to use, and superbly documented!

Dear Sriram,

thank you very much for your nice words.  We are very glad that
you found our library useful: this gives us incentive to continue
this line of work.

 > we would like also to know if you object to our bundling of
 > PPL with our binaries.

Provided you comply with the license under which you received the PPL
(Version 2 of the GNU General Public License) I see no problem at all.

 > If there is anything else that you
 > would like to see cited on our project page, please let us know.

Oh yes, thanks.  The main paper about the Parma Polyhedra Library is

@Inproceedings{BagnaraRZH02,
   Author = "R. Bagnara and E. Ricci and E. Zaffanella and P. M. Hill",
   Title = "Possibly Not Closed Convex Polyhedra
            and the {Parma Polyhedra Library}",
   Booktitle = "Static Analysis:
                Proceedings of the 9th International Symposium",
   Address = "Madrid, Spain",
   Editor = "M. V. Hermenegildo and G. Puebla",
   Publisher = "Springer-Verlag, Berlin",
   Series = "Lecture Notes in Computer Science",
   Volume = 2477,
   ISBN = "3-540-44235-9",
   Pages = "213--229",
   Year = 2002,
}

By running `nm --demangle lsting' I have got the impression that you
are using the class NNC_Polyhedron implementing not necessarily closed
convex polyhedra and never the class C_Polyhedron for topologically
closed convex polyhedra.  If that is the case, then you may want to
cite

@Inproceedings{BagnaraHZ03a,
   Author = "R. Bagnara and P. M. Hill and E. Zaffanella",
   Title = "A New Encoding and Implementation
            of Not Necessarily Closed Convex Polyhedra",
   Booktitle = "Proceedings of the 3rd Workshop on Automated Verification
                of Critical Systems",
   Address = "Southampton, UK",
   Editor = "M. Leuschel and S. Gruner and S. {Lo Presti}",
   Year = 2003,
   Pages = "161--176",
   Note = "Published as TR Number DSSE-TR-2003-2, University of Southampton",
}

as this paper details the unique features of the PPL as far as the
handling of NNC polyhedra is concerned (if you have tried another
implementation of NNC polyhedra you certainly know how these features
are crucial also from the user's perspective).

We have added a link to your web pages to our credit's page:
http://www.cs.unipr.it/ppl/Credits/.
By the way, the name of our library is "Parma Polyhedra Library",
not "Parma Polyhedral Library".

 > Thanks!
 > Sriram Sankaranarayanan
 > -------------------------------
 >
 > PhD student,
 > Theoretical Computer science,
 > Stanford University,
 > Stanford, CA 94305.
 >
 > http://www.stanford.edu/~srirams

Thanks to you for using our library.  Actually this is very good
for our project as it gives us a new application to play with,
new ways of performing regression testing, and an opportunity
to see the user's needs in a concrete way.

Please send us the sources of StInG so that we can try it with the new
versions of the PPL (version 0.6.1 is being released right now) and
see the effects of our experimental development branches on your
application.  We will keep you informed of any new feature/improvement
that may be of interest to you.  By the way, let me suggest you to
subscribe the PPL-announce mailing list
(http://www.cs.unipr.it/mailman/listinfo/ppl-announce).
It is a very low-volume, zero-spam (see the archives to check by yourself),
strictly moderated list with which we announce new releases.
All the best,

    Roberto, on behalf of the PPL team

-- 
Prof. Roberto Bagnara
Computer Science Group
Department of Mathematics, University of Parma, Italy
http://www.cs.unipr.it/~bagnara/
mailto:bagnara at cs.unipr.it