November 2001 - Haskell

FORTE 2002, Call For Papers
by Doron Peled 19 Nov '01

19 Nov '01

Formal Techniques for Networked and Distributed Systems FORMAL METHODS FOR PROTOCOL ENGINEERING AND DISTRIBUTED SYSTEM FORTE 2002 November 11 -- 14 , 2002, Houston, Texas http://www.ece.utexas.edu/FORTE Program Chairs: Doron A. Peled Dept. of Electrical and Comp. Eng. The University of Texas at Austin Austin, TX 78712 Phone: +1-512-232-9371 doron(a)ece.utexas.edu Moshe Y. Vardi Department of Comp.Science Rice University 6100 S. Main St. Houston, TX 77005 Phone: +1-713-348-5977 vardi(a)cs.rice.edu Program Committee: R. Alur, UPenn D. Bjorner, IT University G. v. Bochmann, Univ.of Ottawa T. Bolognesi, IEI, Italy E. Brinksma, Univ of Twente A. Cavalli, INT, France S. T. Chanson, Hong Kong Univ P. Dembinski, IPI-PAN H. Garavel, Inria S. Gnesi, CNR - IEI, Italy G. J. Holzmann, Bell Labs, USA A. Hu, UBC, Canada C. Jard, IRISA - CNRS G. Leduc, Univ. of Liege D. Lee, Bell Labs, China I. Lee, UPenn S. Leue, Univ. of Freiburg L. Logrippo, Univ.of Ottawa S. Mauw, Eindhoven K. McMillan, Cadence M. Morley, Verisity A. Muscholl, Paris 7 E. Najm, ENST, France A. Petrenko, CRIM, Canada S. Smolka, Stony Brook R. Tenney, Univ.of Massachusetts K. Turner, Univ.of Stirling B. Roscoe, Univ.of Oxford S. T. Vuong, Univ.of BC M. Yannakakis, Avaya Labs Steering Committee: G. v. Bochmann, Univ.of Ottawa E. Brinksma, Univ.of Twente S. Budkowski, INT, France G. Leduc, Univ.of Liege E. Najm, ENST, France R. Tenney, Univ.of Massachusetts K. Turner, Univ.of Stirling The IFIP TC6 WG 6.1 Joint International Conference on Formal Techniques for Networked and Distributed Systems (FORTE) is focused on formal methods for communication protocols. FORTE is the new name of the joint FORTE/PSTV meeting, which has combined FORTE and PSTV into a single joint meeting since 1997. The conference will include invited talks and tutorials by Edmund M.Clarke, David Harel, Dave Johnson, Dan Wallach and Elaine J. Weyuker. The conference is a forum for presentation and discussion of the state of the art in theory, application, tools and industrialization of Formal Description Techniques (FDT's). Research papers and industrial usage reports are solicited, particularly in the following areas: >FDT-based system and protocol engineering. >Semantical foundations. >Extensions of FDT's. >Formal approaches to concurrent/distributed Object-Oriented systems. >Real-time and probability aspects. >Performance modeling and analysis. >Quality of service modeling and analysis. >Verification and validation. >Relations between informal and formal specification. >FDT based protocol implementation. >Software tools and support environments. >FDT application to distributed systems, high speed internet protocols, multimedia and multicast protocols and services. >FDT application to wireless and mobile communication, intelligent networks, network management, network security and voice services. >Protocol testing, including conformance testing, interoperability testing, and performance testing. >Test generation, selection and coverage. >Practical experience and case studies. >Corporate strategic and financial consequences of using formal methods. Paper submission: Submission instruction will be available at the conference web site. Submissions should be received by April 20, 2002. The deadline is firm; late submissions and papers departing from the submission guidelines will not be considered. Authors will be notified of acceptance or rejection by June 30, 2002. Accepted papers will be published by Springer Verlag in the Lecture Notes in Computer Science series, and are due by August 12, 2002. The first page of the extended abstract and the cover letter should include the title of the paper, names and affiliations of authors, a brief synopsis (at most 10 lines), keywords, and contact author's name, address, phone number, fax number, and email address. Full original research papers and industrial usage reports should be up to 16 pages, including abstract, names and affiliations of all authors, and a list of keywords facilitating the assignment of papers to referees. For industrial usage reports, short papers up to 8 pages are also welcome. The submission must be in English and provide sufficient detail to allow the program committee to assess the merits of the paper. It should begin with a succinct statement of the issues, a summary of the main results, and a brief explanation of their significance and relevance to the conference, all phrased for the non-specialist. References and comparisons with related work should be included. Technical development directed to the specialist should follow. Submissions departing significantly from these guidelines risk rejection. The results must be unpublished and not submitted for publication elsewhere, including the proceedings of other symposia or workshops.

1 0

seq / strictness and laziness
by Amanda Clare 19 Nov '01

19 Nov '01

I have some code which is being unnecessarily lazy (and occupying too much heap space). The code should read and process several files one by one. What's happening is that all files get read in but the processing is delayed by laziness, and the files are being retained. It looks something like this (after simplification): main = do result <- foldM countAFile initialcounts fileNameList prettyprint result countAFile oldcounts filename = do compiledFile <- readAndCompile filename let newcounts = countAssociations compiledFile oldcounts return newcounts I would love to have used CPS, but this seems to force me to lift a huge amount of code into both the IO monad and the CPS, as countAssociations just calls another function, which calls another, and so on. If I change countAFile to use a redundant test to force evaluation of newcounts: countAFile oldcounts filename = do compiledFile <- readAndCompile filename let newcounts = countAssociations compiledFile oldcounts if newcounts == newcounts then return newcounts else error "suitable error message" then the program runs beautifully (if a little slowly), occupying a constant amount of heap. If instead I use a strict data type to store my counts, or use seq, then I get no improvement, and it's just as if I hadn't used them. I understand why a strict data type won't help me, as it's the function calls that are lazy. But what about $! and seq? countAFile oldcounts filename = do compiledFile <- readAndCompile filename let newcounts = countAssociations compiledFile oldcounts return $! newcounts Does anyone understand why $! (or equivalently "seq") have no effect here? Is there a better way to do this? Amanda -- Amanda Clare http://users.aber.ac.uk/ajc99/ Tel: +44 (0)1970 621922 Fax: +44 (0)1970 622455 Dept. of Computer Science, University of Wales, Aberystwyth, SY23 3DB

4 4

30th Spring School Theorical Computer Science - Early Registration
by ecole＠pps.jussieu.fr 19 Nov '01

19 Nov '01

======================================================================= 30th SPRING SCHOOL THEORETICAL COMPUTER SCIENCE 24 - 29 March, 2002 in AGAY (VAR, FRANCE) EARLY REGISTRATION [Apologies for multiple copies] ======================================================================= This spring school aims at bringing together students and researchers eager to learn about the fundamental questions which language designers and implementors are facing those days, and on the most up-to-date tools that theoreticians have developed or are in the process of developing (such as games and ludics, or realisability for classical logic and set theory). Denotational semantics was born some thirty years ago from the encounter of computer scientists who aimed at implementation independent definitions of programming constructs on one hand, and logicians who provided mathematical tools to this aim (domain theory) on the other hand. Algebraic tools such as initial algebra semantics were also instrumental to the birth of this subject. Thirty years later, the subject has enriched considerably, both in tools (including connections with proof theory and category theory) and in coverage and applications (functional programming, state, control, objects, parallelism, mobility,...). Concepts and constructs that have arisen in the design of programming languages have found counterparts in logic, logical systems have found their way to applications in the form of proof assistants, etc... The rapid development of new programming paradigms, in which distributed computation takes a more and more important part, infers a strong demand on theory, more than ever. With the rise of quite specialized subcommunities (like type theory, linear logic, pi-calculus, functional programming, etc...), it is important to keep an eye on more comprehensive training events, that can address the interrelations between research areas in rapid growth, and between theory-oriented research work and more goal-oriented one, with the prospect of solving problems or of improving our understanding in one domain using tools of another domain. The diversity of the proposed lectures, combined with their conceptual overall unity (as witnessed by the very limited number of core formal systems: lambda-calculus, linear logic, pi-calculus) addresses this issue. ======================================================================= LECTURES ======================================================================= Denotational semantics and games semantics (5 hours) : Thomas Ehrhard , Guy McCusker Ludics (4 hours) : Pierre-Louis Curien, Jean-Yves Girard Realisability (4 hours) : Vincent Danos, Jean-Louis Krivine Continuations (4 hours) : Olivier Danvy Compilation, objects, modules (4 hours) : Xavier Leroy Mobility (5 hours) : Luca Cardelli, Cedric Fournet Security (5 hours) : Martin Abadi, Francois Pottier ======================================================================= CONTACT, FURTHER IMFORMATION & EARLY REGISTRATION ======================================================================= An early registration form can be filled at the following web address : http://www.pps.jussieu.fr/~ecole/ email: ecole(a)pps.jussieu.fr Equipe Preuves Programmes et Systèmes Université Denis Diderot Case 7014 2 Place Jussieu 75251 PARIS Cedex 05 FRANCE

1 0

not naming modules Main
by Hal Daume 19 Nov '01

19 Nov '01

I'm really frustrated that modules that you want to compile to executables have to be named Main. I often have a module with a main method that I use for testing or whatever (perhaps I want the gained speed of an executable) but is, for the most part, a module I import into others. I end up having to constantly change the module name whenever I want to compile it and I find this terribly frustrating. Is there any reason you can't just compile things that simply export a main method with the proper type? Is this a ghc specific thing or does nhc also have this restriction? Any chance ghc will change its policy on this? - Hal -- Hal Daume III "Computer science is no more about computers | hdaume(a)isi.edu than astronomy is about telescopes." -Dijkstra | www.isi.edu/~hdaume

5 4

Help
by Juan M. Duran 19 Nov '01

19 Nov '01

Hi, I am developing a software based on Genetic Programming and I am trying that this soft makes source code in Haskell. I've two problems, one is that I only count on The Haskell report and I need more information about haskell's available library. On the other hand, I need to read a file and (as in imperative programming) get each word the file has because they are parameters for future functions. Thanks Juan PS: Here is some code I made: import IO arch = "twoboxes.dat" mode = ReadMode main = do {handler <- openFile arch mode; contents <- hGetContents handler; putStr contents; hClose handler } putList :: a -> [a] -> [a] putList x [] = [x] putList x xs = x:xs

1 0

heap and walking through a very long list
by Lukasz Pankowski 19 Nov '01

19 Nov '01

Hi. I wonder if there are any methods of walking through a very long list without a huge heap. It is good that a laziness makes creation of large stuctures delayed, but it seems that destruction of never again used beginning of a list is also delayed (probably because of other reason). Consider the simple program: main = do putStrLn $ show $ hist_inc 100000 3 hist_inc n b = hist (0,b) $ take n $ repeat b hist bnds is = accumArray (+) 0 bnds [(i, 1) | i <- is, inRange bnds i] and it's output (compiled with GHC): $ ./hist_inc Stack space overflow: current size 1048576 bytes. Use `+RTS -Ksize' to increase it. I am using GHC 5.0, does it have any optimization to overcome it (`-O -fvia-C -O2-for-C' does not do this). It is obvious that the beginning of the list is no longer needed (does garbage collector now this?). If there is a way is it written somewhere in documentation? I feel it insane to have to have let say 256Mb of memory because I have 1 million measurements on a list. Currently I pipe the results to a C program, nice UN!X practice which I would like avoid. -- =*= Lukasz Pankowski =*=

1 0

ANNOUNCE: HOpenGL 1.01 released
by Sven Panne 17 Nov '01

17 Nov '01

HOpenGL, a Haskell binding for OpenGL and GLUT version 1.01 I am pleased to announce the twelfth release of the Haskell binding for GL 1.2.1 / GLU 1.3 / GLUT 3.7beta. It offers easy access to *the* industrial strength 3D graphics API and a GUI toolkit. More details about HOpenGL can be found at http://www.informatik.uni-muenchen.de/~Sven.Panne/haskell_libs/HOpenGL.html This is a compatibility release for >= GHC-5.02. With a little work, HOpenGL can probably tweaked to work with a recent Hugs and NHC98, feedback about this and/or patches are highly welcome. Have fun! Sven

1 0

announce: greencard, version 2.03
by Sigbjorn Finne 17 Nov '01

17 Nov '01

A new release of GreenCard is out, version 2.03. It is a refresh release, fixing a couple of minor bugs + updating the source code to make it work with current Haskell systems. Also new in this release is the inclusion of an example (contributed by Antony Courtney) which demonstrates how to use GreenCard together with Hugs and/or GHC on your favourite platform. More info, distributions etc. at: http://www.haskell.org/greencard/ --sigbjorn

1 0

Your New Account Info
by welcome 17 Nov '01

17 Nov '01

Your E-mail Address Removal/Deletion Instructions at the end of this email: We visited your site and thought the following might benefit you. Dear Advertiser: VnetFind.Com is currently achieving over 1,000,000 hits a month! We had been a meta search engine for quite sometime and just recently went to paid rankings (pay per click)! Most of our search terms are still available for only $.01! And with our Amazing Bonuses of up to 800% you could get highly targeted search engine traffic for as little as $.001 per unique visitor! You have nothing to lose and everything to gain! You pay only for traffic you receive NO OTHER FEES, NO MONTHLY MINIMUMS! Check out what VnetFind.Com has to offer! List your site now and VnetFind.Com will give you: Traffic - We deliver hundreds of thousands of searches every month. Targeting - Directly connect with specific audiences through custom titles and descriptions. Cost effectiveness - Pay when targeted customers click through to your site. Control - Choose how much to pay for each visitor. Easy account management - Get 24 hours a day, 7 days a week account access at Vnetfind.Com Free Money- VnetFind.Com will match your deposit with up to an 800% bonus! When you put in the promo code MPT1310 YOU WILL get $10 free just for signing up plus the following bonuses on deposits! Deposit $10 USD into your account and receive a 100% Bonus ($20 in your account) Deposit $25 USD into your account and receive a 500% Bonus ($150 in your account) Deposit $50 USD into your account and receive a 600% Bonus ($350 in your account) Deposit $100 USD into your account and receive a 700% Bonus ($800 in your account) Deposit $200 USD into your account and receive a 800% Bonus($1800 in your account) Within minutes you can place your site in our top rated search engine and be receiving highly targeted traffic for only pennies a click. Here are the details: We are a bid-for-placement search engine. You bid a certain number of cents you are willing to pay to have people come to your site. The more you pay, the higher you appear in our search. The best part is that you only pay for traffic you receive ... there are no other fees! Register for an account it's easy and only takes a few minutes! Deposit funds into your VnetFind account via check order or by credit card. Select the keywords you wish to place your site under. For example, if you run a site that sells rare coins, you might choose the keywords "rare coins" or "shillings". Create a description of your site for the keywords you chose. Choose how much you wish to pay per click for your sites link. The more you are willing to pay, the higher you will appear in our search engine. You will only be charged for traffic you receive. If you are paying .01 cents per click and receive 8 unique clicks, your account will be debited .08 cents. If a visitor clicks on your entry multiple times in the same day, you will only be charged once. We have special systems in place to make sure that you are only charged for unique clicks and not overcharged. You can request any remaining funds be returned to you at any time and there are no membership fees to use our service! That's it! You will begin receiving highly targeted search engine traffic for as little as $.001 per visitor in no time! Remember this is a special offer and you MUST enter the promo code MPT1310 to receive your bonuses! Thanks and Good Luck! Jennifer C Account Manager $$$$$$$$$$$$$$$$$$$-REMOVAL INSTRUCTION-$$$$$$$$$$ This message is sent in compliance of the new email Bill HR 1910. Under Bill HR 1910 passed by the 106th US Congress on May 24, 1999, this message cannot be considered SPAM as long as I include a valid return address and the way to be removed. ONCE AGAIN THIS IS NOT SPAM. PLEASE IF YOU WANT TO BE REMOVED SEND AN E-MAIL WITH "REMOVE" IN THE SUBJECT TO THE FOLLOWING ADDRESS: removeme(a)vnetdirect.com

1 0

How do I write the type of 'inverse'?
by marku＠cs.waikato.ac.nz 16 Nov '01

16 Nov '01

I have a class whose instances can extract a field of type a. Is there a way of referring to that type 'a' in the signatures of functions that use the class? Sigh! That wasn't very clear. I'll try explaining via example: I am trying to define a `Pairable' class, which is an abstraction of all kinds of pairs: (a,b), Pair a b, etc. > class Ord p => Pairable p where > is_pair :: p -> Bool > pair_fst :: Ord a => p -> a -- Precondition: is_pair > pair_snd :: Ord b => p -> b -- Precondition: is_pair > make_pair :: (Ord a, Ord b) => a -> b -> p > instance (Ord a, Ord b) => Pairable (a,b) where > is_pair (a,b) = True > pair_fst (a,b) = a > pair_snd (a,b) = b > make_pair a b = (a,b) I then want to define a `Relation' to be a set of Pairable values. > newtype Pairable p => Reln p = Reln [p] This works okay for many of my relation operations. But I run into an interesting problem with `inverse' (which swaps the components of each pair). inverse :: (Pairable p, Pairable p2) => Reln p -> Reln p2 inverse rs = Reln[make_pair (pair_snd p) (pair_fst p) | p <- reln2list rs] Hugs gives the error: ERROR /home/utting/jaza/tmp.hs:19 - Cannot justify constraints in explicitly typed binding *** Expression : inverse *** Type : (Pairable a, Pairable b) => Reln a -> Reln b *** Given context : (Pairable a, Pairable b) *** Constraints : (Ord c, Ord d) which I can understand. BUT, I cannot see any way of naming the `component types', c and d, in the signature of inverse! It seems like I want to pass the component types into the class definition somehow: class Pairable (p a b) How does one normally get around this kind of problem? Thanks. Mark Utting Professeur Invité Laboratoire d'Informatique de l'Université de Franch-Comté 16, route de Gray - 25000 Besançon, cedex, FRANCE Tel: (33) 3 81 66 20 69 Fax: (33) 3 81 66 64 50 Email: marku(a)cs.waikato.ac.nz

2 1