August 1998 - Categories - categories.org.au

Call for Papers
by Robert A.G. Seely 28 Aug '98

28 Aug '98

LAMBEK FESTSCHRIFT - CALL FOR PAPERS The (electronic) journal "Theory and Applications of Categories" has agreed to publish a special volume in honour of the work of our colleague Joachim Lambek, in celebration of his 75th birthday, which was marked by a symposium at McGill on his actual birthday last December 5th, 1997. We welcome submissions to this volume from anyone interested. All papers submitted will undergo the usual TAC referee process, and we are hopeful that the final volume will be ready before summer 1999. (For the TAC home page, see <http://www.tac.mta.ca/tac/>). Topics suitable for the volume include any of the subjects to which Jim Lambek has contributed which fall roughly within the scope of TAC. For example, categorical algebra, categorical logic and proof theory, mathematical linguistics, algebra and ring theory (preferably with some categorical application or methodology), categorical computer science, etc. We would like to note that a companion volume is also being prepared by the journal "Mathematical Structures in Computer Science" - there may still be room for a very small number of short-to-moderate-length papers to be included in that volume. If you wish your submission to be considered for that, please indicate this when you submit your paper to us. (Of course, MSCS's emphasis is more towards computer science). The submission deadline for the Lambek Fest volume is 30 January, 1999. Please send a postscript file (uuencoded and compressed, if possible) to rags(a)math.mcgill.ca, or 3 paper copies to R.A.G. Seely Department of Mathematics McGill University 805 Sherbrooke St W Montreal, Quebec Canada, H3A 2K6 (If possible) please send an email message to rags(a)math.mcgill.ca indicating the title (and authors, if other than the sender of the email) of the paper, as well as an abstract (in standard ascii format, maximum 1 page) of the paper. A technical point: TAC can only accept files prepared with some "flavour" of TeX or LaTeX - most authors ought to be able to arrange for such a file to be prepared from their paper, but if you have problems, let us know and we shall see if we can provide helpful advice. General advice for preparation of papers, including suitable macros, may be found at the TAC home page, <http://www.tac.mta.ca/tac/>. The Editors: Michael Barr (e-mail: barr(a)math.mcgill.ca) Philip Scott (e-mail: phil(a)csi.uottawa.ca) Robert Seely (e-mail: rags(a)math.mcgill.ca) A copy of this CFP may be found on the CTRC (Montreal category group) home page <http://triples.math.mcgill.ca>.

1 0

Looking for a post-doc
by Michael Barr 24 Aug '98

24 Aug '98

Postdoc at McGill Centre de Recherche en Theorie des Categories -- Montreal -- Category Theory Research Center Postdoctoral Fellowship in Category Theory The Categories Team at McGill and at the University of Ottawa has had an unexpected resignation and invites applications for Postdoctoral Fellowship. The salary is CA$30,000, and it might possbily be supplemented by a part-time teaching post at McGill. It is a one year limited term appointment, with a possible second year. There is in principle, no nationality restriction, but the successful candidate will have to go through Canadian and Quebec immigration procedures. In the case of a non-North-American resident, this will entail difficulties that would probably delay taking up the fellowship till the beginning of next (calendar) year. Interested candidates should apply, with a CV and a statement of plans, as soon as possible and should ask for letters from three referees. Candidates interested in teaching should also provide evidence of teaching experience as well as English proficiency if that is not their native language. Owing to the lateness, electronic applications and references are much preferred, but if an application comes in by paper in the next couple of weeks it will be accepted. The members of the Categories Team are: M. Barr (McGill), R. Blute (Ottawa), M. Bunge (McGill), T. Fox (Vanier Col.), J. Lambek (McGill), M. Makkai (McGill), A. Sangalli (Champlain Col.), P. Scott (Ottawa), R. Seely (John Abbott Col.). Michael Barr barr(a)math.mcgill.ca Department of Mathematics and Statistics McGill University 805 Sherbrooke St. W Montreal, QC Canada H3A 2K6

1 0

MFPS Call for Papers
by Andre Scedrov 18 Aug '98

18 Aug '98

Announcement and Call for Papers MFPS XV Fifteenth Conference on the Mathematical Foundations of Programming Semantics Tulane University New Orleans, LA USA April 28 - May 1, 1999 Partially Supported by the US Office of Naval Research The Fifteenth Conference on the Mathematical Foundations of Programming Semantics will take place on the campus of Tulane University in New Orleans, LA from April 28 to May 1, 1999. The MFPS conferences are devoted to those areas of mathematics, logic and computer science which are related to the semantics of programming languages. The series particularly has stressed providing a forum where both mathematicians and computer scientists can meet and exchange ideas about problems of common interest. We also encourage participation by researchers in neighboring areas, since we strive to maintain breadth in the scope of the series. The invited speakers for MFPS XV include: Martin Abadi (DEC Research Center) Matthew Hennessy (Sussex) Tom Henzinger (Cal Berkeley) Uday Reddy* (Illinois) Peter Selinger (Michigan) *: To be confirmed In addition to the invited talks, there will be two special sessions. One will be on security, and is being organized by Martin Abadi, Catherine Meadows (NRL) and Dennis Volpano (Naval Postgraduate School). The second session will be on Object-Oriented Programming. The remainder of the program will consist of papers selected from submissions we receive in response to this Call for Papers. The Organizing Committee for MFPS consists of Stephen Brookes (CMU), Michael Main (Colorado), Austin Melton (Kent State University), Michael Mislove (Tulane) and David Schmidt (Kansas State). The Co-chairs for MFPS XV are Stephen Brookes and Michael Mislove. The Program Committee Co-chairs for the meeting are Andre Scedrov (Penn) and Achim Jung (Birmingham). The Program Committee also includes: Stephen Brookes (CMU) Adriana Compagnoni (Stevens Institute) Kathleen Fisher (AT&T) Paul Gastin (LIAFA) Andrew Gordon (Microsoft Research) Reinhold Heckmann (Saarbruecken) Catherine Meadows (NRL) Michael Mislove (Tulane) Peter O'Hearn (Queen Mary and Westfield) Catuscia Palamidessi (Penn State) Prakash Panangaden (McGill) Dusko Pavlovic (Kestrel) Andrew Pitts (Cambridge) A. W. Roscoe (Oxford) Giuseppe Rosolini (Genoa) Jan Rutten (CWI) Eugene Stark (Stony Brook) Submissions must be extended abstracts of 12 pages or less. They should be in the form of a PostScript file that can print on any PostScript printer. In particular, submissions should use the US letter size format, as opposed to the European A4 size. Submissions either can be sent by email to mfps(a)math.tulane.edu or they can be deposited by anonymous ftp in the directory pub/incoming on the machine 129.81.96.30. In the latter case, a message also should be sent to mfps(a)math.tulane.edu advising us of the submission. The Deadline for Submissions is November 3, 1998. Information about decisions will be available shortly after January 1, 1999, at which time a list of accepted papers will be posted. As with MFPS XI and MFPS XIII, the Proceedings of the conference will be published as a volume of the Electronic Notes in Theoretical Computer Science. For information about this series, access the URL http://www.elsevier.nl/locate/entcs. General inquiries about MFPS XV can be addressed to mfps(a)math.tulane.edu. In addition to supporting the conference overall, the support provided by the Office of Naval Research makes funds available to help offset expenses of graduate students. Women and minorities also are encouraged to inquire about possible support to attend the meeting. Registration Information Detailed information about registration and accommodations will be available shortly after January 1, 1999. Participants should be prepared to make travel arrangements for the meeting as soon as information is available; the timing of the meeting will overlap the New Orleans Jazz and Heritage Festival, and it is anticipated that securing travel into and out of New Orleans at this particular time will be difficult if arrangements are not made as early as possible. More Information Updates to the information about MFPS XV can be accessed at the home page for the meeting, http://www.math.tulane.edu/mfps15.html. Inquiries about the meeting can be addressed to mfps(a)math.tulane.edu.

1 0

Re: the lanl e-print server
by James Stasheff 10 Aug '98

10 Aug '98

One correspondent writes: for example, there's only one paper per year classified as "Category Theory" (currently there's only math.CT/9805102), then we conclude that either a) no papers suitable for that classification are being written, b) the people writing those papers are unaware of the archive, c) they're aware of it but prefer not or are too busy to use it, or d) they're actively boycotting it. I'm hoping that c) is the reason and not d). Listing with xxx.lanl.gov in noway precludes or detracts from the use of e.g. hypatia comments? jim ************************************************************ Until August 10, 1998, I am on leave from UNC and am at the University of Pennsylvania Jim Stasheff jds(a)math.upenn.edu 146 Woodland Dr Lansdale PA 19446 (215)822-6707 Jim Stasheff jds(a)math.unc.edu Math-UNC (919)-962-9607 Chapel Hill NC FAX:(919)-962-2568 27599-3250

1 0

New Preprint (revised)
by Ronnie Brown 07 Aug '98

07 Aug '98

The url for the following was not correct and is revised below: New preprint R. Brown and I. Icen. "Lie local subgroupoids and their monodromy", UWB Math Preprint 98.15, 12pp. ABSTRACT:The notion of local equivalence relation on a topological space is generalised to that of local subgroupoid. Properties of coherence are considered. The main result is notions of holonomy and monodromy groupoid for certain Lie local subgroupoids. http://www.bangor.ac.uk/~mas010/brownpr.html#monodromy Prof R. Brown, School of Mathematics, University of Wales, Bangor Dean St., Bangor, Gwynedd LL57 1UT, United Kingdom Tel. direct:+44 1248 382474|office: 382475 fax: +44 1248 383663 World Wide Web: home page: http://www.bangor.ac.uk/~mas010/ Symbolic Sculpture and Mathematics: http://www.bangor.ac.uk/SculMath/ Mathematics and Knots: http://www.bangor.ac.uk/ma/CPM/exhibit/welcome.htm

1 0

Injectives via KZ-monads
by Martin Escardo 04 Aug '98

04 Aug '98

The following short paper is available from my home page: http://www.dcs.ed.ac.uk/home/mhe/pub/papers/top97.ps.gz or http://www.dcs.ed.ac.uk/home/mhe/papers.html (It is an updated version of a paper which was previously circulated in other lists.) ===================================== Injective spaces via the filter monad ====================================================================== An injective space is a topological space with a strong extension property for continuous maps with values on it. A certain filter space construction embeds every T_0 topological space into an injective space. The construction gives rise to a monad. We show that the monad is of the Kock-Zoberlein type and apply this to obtain a simple proof of the fact that the algebras are the continuous lattices (Alan Day, 1975, Oswald Wyler, 1976). In previous work we established an injectivity theorem for monads of this type, which characterizes the injective objects over a certain class of embeddings as the algebras. For the filter monad, the class turns out to consist precisely of the subspace embeddings. We thus obtain as a corollary that the injective spaces over subspace embeddings are the continuous lattices endowed with the Scott topology (Dana Scott, 1972). Similar results are obtained for continuous Scott domains, which are characterized as the injective spaces over dense subspace embeddings, via the proper filter monad. ====================================================================== Two notes (and some questions concerning credit) ========= (i) Bob Flagg and I have also considered the following variations on the filter monad (a report is being written) (a) Category: T_0 exponentiable spaces (= core-compact = open sets form a continuous lattice) Restriction on filters: Scott open. => Associated maps: "semi-proper" embeddings (= right adjoint of the frame maps preserve directed joins) => Algebras (and hence injectives over semi-proper): continuous meet-semilattices with Scott topology. (Corollary: continuous meet-semilattices and Scott continuous functions form a CCC. Was this known before?) This characterization of the algebras was previously known (Andrea Shalk--anyone else?), but the "KZ-method" outlined in the above abstract gives a much shorter proof. (b) Category: T_0 spaces Restriction on filters: prime. => Associated maps: flat embeddings (= right adjoint of the frame maps preserve finite joins) => Algebras (and hence injectives over flat): compact, stably locally compact spaces. (A localic version is given via the ideal monad. What Johnstone refers to as Joyal's Lemma appears as a special case of this.) (I don't know what was previously known about this.) (A result by Isbell (in his paper "Flat = prosupersplit") implies that the flat embeddings form the largest class of embeddings over which the CSLCSs are injective, because finite spaces are (trivially) CSLCSs.) (c) Category: T_0 spaces Restriction on filters: completely prime. => Associated maps: "completely flat" embeddings (= right adjoint of the frame maps preserve all joins) => Algebras (and hence injectives over completely flat): sober spaces. (d) Category: T_0 locally connected spaces Restriction on filters: filters of connected open sets. => Associated maps: "locally dense" embeddings (= frame maps preserve connectedness (and hence right adjoints preserve disjoint unions)) => Algebras (and hence injectives over locally dense): L-domains. (This was obtained by Bob, based on some previous work by Paul Taylor (and Andrea Shalk) on the algebras. Again, the KZ-method gives a simpler proof of the characterization.) (ii) The filter monad is formally analogous to the so-called continuation monad, as it is observed (with the formal details of the analogy) in the paper being advertised. I would like to also mention that the general injectivity result for KZ-monads referred to in the above abstract was established in the paper http://www.dcs.ed.ac.uk/home/mhe/pub/papers/injective.ps.gz which is (mainly) about continuity of the extension process (answering a question by Scott in his 1972 paper on continuous lattices). Comments are wellcome. Martin ================================================================= Martin H. Escardo, Department of Computer Science, LFCS King's Buildings, Mayfield Road, Edinburgh EH9 3JZ, Scotland office: 2606 (JMCB) fax: +44 131 667 7209 phone: +44 131 650 5135 mailto:mhe@dcs.ed.ac.uk http://www.dcs.ed.ac.uk/home/mhe =================================================================

1 0

New Preprint
by Ronnie Brown 03 Aug '98

03 Aug '98

New preprint R. Brown and I. Icen. "Lie local subgroupoids and their monodromy", UWB Math Preprint 98.15, 12pp. ABSTRACT:The notion of local equivalence relation on a topological space is generalised to that of local subgroupoid. Properties of coherence are considered. The main result is notions of holonomy and monodromy groupoid for certain Lie local subgroupoids. http://www.bangor.ac.uk/~mas010/papers/sub6.ps,.dvi Prof R. Brown, School of Mathematics, University of Wales, Bangor Dean St., Bangor, Gwynedd LL57 1UT, United Kingdom Tel. direct:+44 1248 382474|office: 382475 fax: +44 1248 383663 World Wide Web: home page: http://www.bangor.ac.uk/~mas010/ Symbolic Sculpture and Mathematics: http://www.bangor.ac.uk/SculMath/ Mathematics and Knots: http://www.bangor.ac.uk/ma/CPM/exhibit/welcome.htm

1 0

The Logic of Bunched Implications.
by David J. Pym 03 Aug '98

03 Aug '98

We, Peter O'Hearn and David Pym, are pleased to announce our new paper, ``The Logic of Bunched Implications''. We hope it will be of interest to readers of `categories'. BI is a relevant logic which extends both linear and intuitionistic logic. It has a semantics of proofs based on `doubly closed categories', which carry two monoidal closed structures, one of which is cartesian in models of BI. A rich class of models is provided by Day's tensor product construction on the category of presheaves over a small monoidal category. It also comes with a lambda calculus, a truth semantics and a computational interpretation as a logic of resources, quite different from that of linear logic. The paper is available at: http://www.dcs.qmw.ac.uk/~pym and http://www.dcs.qmw.ac.uk/~ohearn where drafts of various companion and related papers can/will be found. P.W.O'Hearn and D.J. Pym Queen Mary & Wesfield College, University of London Abstract. Introduce a logic BI in which a multiplicative (or linear) and an additive (or intuitionistic) implication live side by side. The propositional version of BI arises from an analysis of the proof-theoretic relationship between conjunction and implication, and may be viewed as a merging of intuitionistic logic and multiplicative, intuitionistic linear logic. The predicate version of BI includes, in addition to standard additive quantifiers, multiplicative (or intensional) quantifiers ``forall-new'' and ``exist-new'' which arise from observing restrictions on structural rules on the level of terms as well as propositions. We discuss computational interpretations, based on sharing, at both the propositional and predicate levels.

1 0