- Haskell - Haskell.org

Postdoc and PhD positions at CWI, Amsterdam
by Farhad.Arbab＠cwi.nl 12 Nov '03

12 Nov '03

Position Description ==================== The Coordination and Component Based Software group in SEN3 at CWI has two open positions for: (1) a postdoc for a period of four years, and (2) a PhD student (OIO) for four years. Both positions are within the project number 600.065.120.03N02 Compositional Construction of Component Connectors (C-Quattro), recently funded by the NWO (the Dutch government's funding agency for academic scientific research), with F. Arbab (CWI) and J. Rutten (CWI and VUA) as Principal Investigators. The project C-Quattro is in fact a collaboration of SEN3 at CWI with the Section Theoretical Computer Science at the Free University of Amsterdam (VUA). The postdoc will be employed by CWI; the PhD student will be employed by the VUA but will spend part of his or her time at CWI. The main goal of the C-Quattro project is to develop an operational semantics based on a calculus of component connectors, as the foundation for further development of practical tools and programming environments for the actual deployment of Reo. Reo is a recently introduced channel-based coordination model wherein complex coordinators, called connectors, are compositionally built out of simpler ones. Reo is intended as a ``glue language'' for construction of connectors that orchestrate component instances in a component-based system. The emphasis in Reo is on connectors and their composition only, not on the components that are being connected. The activities under C-Quattro involve both system oriented and theoretical work. On the systems side, we develop tools supporting visual connector programming, as well as the automation of equivalence proofs, e.g., for optimization. On the theoretical side, the semantic modeling is based on the coalgebraic methodology, and coinduction, with its potential for automation, will be the main reasoning principle. The candidate for the postdoc position is expected to have a strong background in Software Engineering, maturity in formal methods and their practical applications, familiarity with component-based software, concurrency, and coordination. Teamwork and leadership, as well as the ability to work effectively with academic colleagues and PhD students, are all important qualifications for this position. The candidate for the PhD position should have a master degree in computer science or mathematics, with a clear interest in questions from theoretical computer science. Ideally, the candidate has a background both in mathematical disciplines such as algebra, analysis and discrete mathematics, and in such elements of theoretical computer science as automata theory and semantics. The Theme SEN3 (http://www.cwi.nl/sen3) at CWI is a dynamic group of internationally recognized researchers who work on Coordination Models and Languages and Component-Based Software Composition. The activity in SEN3 is a productive, healthy mix of theoretical, foundational, and experimental work in Computer Science, ranging in a spectrum covering mathematical foundations of models of computation, formal methods and semantics, implementation of advanced research software systems, as well as their real-life applications. General information =================== CWI is an internationally renowned research institute in mathematics and computer science, located in Amsterdam, The Netherlands. The focus is on fundamental research problems, derived from societal needs. Research is carried out in 15 research themes. More information about these themes can be found on the website www.cwi.nl where you can also take a look at our Annual Report. A substantial part of this research is carried out in the framework of national or international programs. CWI maintains excellent relations with industry and the academic world, at home as well as abroad. After their research careers at CWI, an increasing number of young staff members find employment in these sectors, for example in spin-off companies that are based on research results from CWI. Of course, library and computing facilities are first-rate. CWI's non-scientific services to its personnel include career planning, training & courses, assistance in finding housing, and tailor-made solutions to problems that may occasionally arise. Terms of employment =================== The salary is in accordance with the "CAO-onderzoekinstellingen" and is commensurate with experience. For instance, the postdoc base salary for a fresh PhD with no additional experience in scale 10 is around 2800 Euros/month, and for an experienced PhD in scale 12 it is around 4500 Euros/month. The current starting salary for a first year PhD student is around 1500 Euros/month with an incremental raise for each subsequent year. (Note: a considerable increase in the salaries of all PhD students is currently under negotiation). Besides the salary, CWI offers very attractive and flexible terms of employment, like a collective health insurance, pension-fund, etc. Application =========== For more information on these vacancies you can contact either of the PIs: F. Arbab, telephone +31-20-5924056, e-mail Farhad.Arbab(a)cwi.nl J. Rutten, telephone +31-20-592-4116, email Jan.Rutten(a)cwi.nl Official applications, together with curriculum vitae, letters of references, and lists of publications must be sent to Mrs. J. Koster, head of Personnel Department, P.O. box 94079, 1090 GB Amsterdam, The Netherlands.

1 0

Network.CGI, (embedded) graphics
by Johannes Waldmann 12 Nov '03

12 Nov '03

I'd like to embed pictures (.png) in a CGI-generated output. This doesn't seem to fit directly into ghc's Network.CGI setting. (Not entirely ghc's fault, of course :-) By the html definition, including a picture seems to require an indirection. But I don't want to expose a static address, so I am thinking of something like <IMG SRC="Get.cgi?identifier"> But this can't be done with ghc's CGI library (it assumes that text/html is output). So the library should export some more of its functions. (Yes, I can easily patch this in a copy of the source code http://cvs.haskell.org/cgi-bin/cvsweb.cgi/~checkout~/fptools/libraries/netw… Ideally, I am thinking of class Picture a where picture :: a -> IO [ Byte ] and I want to call Get.cgi with argument `urlencoded (show a)'. The underlying program should then emit the byte stream (from scratch, or from a cache). But then it would need to do `read (urldecode input)' and this would require to transmit a representation of the type as well. Any ideas? BTW, my application would be drawings of finite automata (which can be nicely done by dot/neato), in order to make this a bit more friendly to the eye: http://elearning.htwk-leipzig.de/~autotool/cgi-bin/Face.cgi (test account: matrikel 318, passwort guest, then check problem Determine-D-Quiz) -- -- Johannes Waldmann, Tel/Fax: (0341) 3076 6479 / 6480 -- ------ http://www.imn.htwk-leipzig.de/~waldmann/ ---------

1 0

RE: Question about how GUM works
by Simon Peyton-Jones 12 Nov '03

12 Nov '03

I think you'll maximise your chances by sending mail about GUM and Glasgow Parallel Haskell to the GpH mailing list http://www.macs.hw.ac.uk/~dsg/gph/ Simon | -----Original Message----- | From: haskell-bounces(a)haskell.org [mailto:haskell-bounces@haskell.org] On Behalf Of Yang, Qing | Sent: 12 November 2003 08:37 | To: haskell(a)haskell.org | Subject: Question about how GUM works | | Hi All, | | I have a question about GUM - the runtime system for Parallel Haskell. | There are below words in the paper addressing GUM architecture: | "The first action of GUM program is to create a PVM manager task, whose | job is to control startup and termination. The manager spawns the | required number of logical PEs as PVM tasks, which PVM maps to the | available processors." | | So my question is how many PEs the PVM mamager task will spawn? The | number is accordant with the processor number or determined by some | other things? And will the number of PEs be changed during the execution | of the program or not? | | Thanks, | | Qing Yang | qing.yang(a)intel.com | | _______________________________________________ | Haskell mailing list | Haskell(a)haskell.org | http://www.haskell.org/mailman/listinfo/haskell

1 0

Question about how GUM works
by Yang, Qing 12 Nov '03

12 Nov '03

Hi All, I have a question about GUM - the runtime system for Parallel Haskell. There are below words in the paper addressing GUM architecture: "The first action of GUM program is to create a PVM manager task, whose job is to control startup and termination. The manager spawns the required number of logical PEs as PVM tasks, which PVM maps to the available processors." So my question is how many PEs the PVM mamager task will spawn? The number is accordant with the processor number or determined by some other things? And will the number of PEs be changed during the execution of the program or not? Thanks, Qing Yang qing.yang(a)intel.com

1 0

RE: ANNOUNCE: HC&A Report (5th edition, November 2003)
by Simon Peyton-Jones 12 Nov '03

12 Nov '03

I think we all owe Claus a big "Thank You" for initiating and running the Haskell Communities Report. It's done a great job of telling us what's going on. Thank you Claus! I hope someone else will step in for May 2004... Simon | -----Original Message----- | From: haskell-bounces(a)haskell.org [mailto:haskell-bounces@haskell.org] On Behalf Of C.Reinke | Sent: 11 November 2003 21:16 | To: haskell(a)haskell.org | Subject: ANNOUNCE: HC&A Report (5th edition, November 2003) | | | On behalf of the many contributors, I am happy to announce that the | | ----------------------------------------- | Haskell Communities and Activities Report | (5th edition, November 2003) | | http://www.haskell.org/communities/ | ----------------------------------------- | | is now available from the Haskell Communities home

1 0

ANNOUNCE: HC&A Report (5th edition, November 2003)
by C.Reinke 11 Nov '03

11 Nov '03

On behalf of the many contributors, I am happy to announce that the ----------------------------------------- Haskell Communities and Activities Report (5th edition, November 2003) http://www.haskell.org/communities/ ----------------------------------------- is now available from the Haskell Communities home page in several formats: in PDF (for those who haven't noticed: that format is not just for printing, but also for online viewing, with working links and table of contents) or, for those who have problems with the PDF, in HTML (using John's secret weapon yet again) and Postscript. A big thanks here to everyone who contributed, be it by making sure that there are so many interesting activities to report on, or by sending in descriptions to make sure that we can read about these activities! I hope you will find it as interesting to read as we did. For those of you who haven't heard of these reports before (and have filtered all previous calls into the spam-folder..), the first edition of the HC&A Report was released in November 2001, with the goal of helping to improve the communication between the increasingly diverse groups, projects, and individuals working on, with or inspired by Haskell. The idea of these reports is simple: Every six months, a call goes out to all of you on the Haskell mailing list to contribute brief summaries of your own area of work. Many of you respond (eagerly, unprompted, and well in time for the deadline;-) to the call. The editor then collects all these into a single report and feeds it back to this very mailing list. And when we try for the next update in six months, you might want to add your own work, project, research area or group as well. So, please, put that item into your diary now! --------------------------------------- End of April 2004: target deadline for contributions to the May 2004 edition of the HC&A Report --------------------------------------- It has become clear that many Haskellers who work on interesting projects no longer have the time to follow the Haskell mailing list closely and may thus miss the calls for contribution. If you are a member, user or friend of such projects, please point them to the current edition, and invite them to "register" with me for a simple email-reminder in the middle of April (and no, you can't register anyone else:). Of course, they'll still have to act on that reminder, but perhaps we can extend our reach this way.. Enjoy (and communicate;-)! Claus Reinke -- Computing Laboratory University of Kent http://www.cs.kent.ac.uk/~cr3/ PS. Please note the preface: as announced at this year's Haskell workshop, there'll be a change of editor after this edition. Andres Loeh has kindly offered to take over starting with the November 2004 edition, but we are still looking for someone to take on the May 2004 edition. If you're interested, please get in touch with me! _______________________________________________ Haskell mailing list Haskell(a)haskell.org http://www.haskell.org/mailman/listinfo/haskell

1 0

ANNOUNCE: HC&A Report (5th edition, November 2003)
by C.Reinke 11 Nov '03

11 Nov '03

On behalf of the many contributors, I am happy to announce that the ----------------------------------------- Haskell Communities and Activities Report (5th edition, November 2003) http://www.haskell.org/communities/ ----------------------------------------- is now available from the Haskell Communities home page in several formats: in PDF (for those who haven't noticed: that format is not just for printing, but also for online viewing, with working links and table of contents) or, for those who have problems with the PDF, in HTML (using John's secret weapon yet again) and Postscript. A big thanks here to everyone who contributed, be it by making sure that there are so many interesting activities to report on, or by sending in descriptions to make sure that we can read about these activities! I hope you will find it as interesting to read as we did. For those of you who haven't heard of these reports before (and have filtered all previous calls into the spam-folder..), the first edition of the HC&A Report was released in November 2001, with the goal of helping to improve the communication between the increasingly diverse groups, projects, and individuals working on, with or inspired by Haskell. The idea of these reports is simple: Every six months, a call goes out to all of you on the Haskell mailing list to contribute brief summaries of your own area of work. Many of you respond (eagerly, unprompted, and well in time for the deadline;-) to the call. The editor then collects all these into a single report and feeds it back to this very mailing list. And when we try for the next update in six months, you might want to add your own work, project, research area or group as well. So, please, put that item into your diary now! --------------------------------------- End of April 2004: target deadline for contributions to the May 2004 edition of the HC&A Report --------------------------------------- It has become clear that many Haskellers who work on interesting projects no longer have the time to follow the Haskell mailing list closely and may thus miss the calls for contribution. If you are a member, user or friend of such projects, please point them to the current edition, and invite them to "register" with me for a simple email-reminder in the middle of April (and no, you can't register anyone else:). Of course, they'll still have to act on that reminder, but perhaps we can extend our reach this way.. Enjoy (and communicate;-)! Claus Reinke -- Computing Laboratory University of Kent http://www.cs.kent.ac.uk/~cr3/ PS. Please note the preface: as announced at this year's Haskell workshop, there'll be a change of editor after this edition. Andres Loeh has kindly offered to take over starting with the November 2004 edition, but we are still looking for someone to take on the May 2004 edition. If you're interested, please get in touch with me!

1 0

A question about monads
by Ryoko Nanase 11 Nov '03

11 Nov '03

2 1

neat trick: parallel show
by John Meacham 11 Nov '03

11 Nov '03

while working on a random project, I had opprotunity to write a 'parallel show' which concurrently evaluates everything needed to show a structure and lets you print succesivly refined versions as vaules are calculated and filled in. it is nice because a value that takes a long time (or perhaps forever) to evaluate at the front of your structure doesn't keep you from seeing what else is in it. I must say it has been a very useful little bit of code so thought I might share the concept and was curious if anyone else does things like this.. it works something like foo = [Right "foo", Left undefined, undefined, Right "here"] parShow (parPretty foo) >>= putStrLn => "(Right foo) (Left _|_) _|_ (Right here)" and succesive calls to parShow will fill in the _|_ values as they get evaluated. my attached implementation is rather weak (my real one is tied into a project and does wierd things with ANSI escape codes to redraw the screen as things get evaluated and handles exceptions differently) but the gist is the same. would people be interesting in me( or someone else :) ) turning this into a full fledged library? does one alreoady exist which does this? I was thinking integrating it into a real pretty printing system would be handy but might be too much work, if there were a way to reuse an existing pretty printing library without change and somehow add this capability then that would be cool. I couldn't think of a way to do it since the final 'render' routine obviously needs to be in the IO monad so the MVars can be checked again when rerendering.. John -- --------------------------------------------------------------------------- John Meacham - California Institute of Technology, Alum. - john(a)foo.net ---------------------------------------------------------------------------

1 0

CfP: ESSLLI'04 Student Session
by Paul Egré 10 Nov '03

10 Nov '03

CALL FOR PAPERS ESSLLI-2004 STUDENT SESSION 9-20 August, 2004 in Nancy, France Deadline: February 22, 2004 http://esslli2004.loria.fr/ We are pleased to announce the Student Session of the 16th European Summer School in Logic, Language and Information (ESSLLI-2004), which will be held in Nancy from August 9-20, 2004. We invite submission of papers for presentation at the ESSLLI-2004 Student Session and for appearance in the proceedings. PURPOSE: This ninth ESSLLI Student Session will provide, like the previous editions, an opportunity for ESSLLI participants who are students to present their own work in progress and get feedback from senior researchers and fellow-students. The ESSLLI Student Session encourages submissions from students at any level, undergraduates (before completion of the Master Thesis) as well as postgraduates (before completion of the PhD degree). Papers co-authored by non-students will not be accepted. Papers may be accepted for full presentation (30 minutes including 5 to 10 minutes of discussion) or for a poster presentation. All the accepted papers will be published in the ESSLLI-2004 Student Session proceedings, which will be made available during the summer school. REQUIREMENTS: The Student Session papers should describe original, unpublished work, completed or in progress, that demonstrates insight, creativity, and promise. No previously published papers should be submitted. We welcome submissions with topics within the areas of Logic, Language and Computation. SUBMISSION DETAILS: Student authors are invited to submit a full paper, not to exceed 7 pages of length exclusive of references. Note that the length of the final version of the accepted papers will not be allowed to exceed 10 pages. The submissions will be reviewed by the student session programme committee and additional reviewers. A plain ASCII text version of the identification page should be sent separately, using the following format: Title: title of the submission First author: firstname lastname Address: address of the first author ...... Last author: firstname lastname Address: address of the last author Short summary: abstract (5 lines) Subject area (one or two of): Logic | Language | Computation In case the paper is being submitted to another conference or workshop, this must be clearly indicated on the identification page. The preferred formats of submissions are PostScript, PDF, or plain text, although other formats will also be accepted. In case of acceptance, the final version of the paper will have to be submitted in LaTeX format. The papers must use single column A4 size pages, 11pt or 12pt fonts, and standard margins. Submissions not in accordance with the specified formatting and length requirements may be subject to rejection without review. The paper and separate identification page must be sent electronically to: paulegre(a)magic.fr by FEBRUARY 22, 2004. PROGRAMME COMMITTEE: Laura Alonso i Alemany (Universitat de Barcelona, Barcelona) Carlos Areces (LORIA, Nancy) Jaume Baixeries (Universitat Politècnica de Catalunya, Barcelona) Willem Comradie (Rand Afrikaans University, South Africa) Benoit Crabbé (LORIA, Nancy) Paul Egré (IHPST, Université Paris 1, IJN) Judit Gervain (SISSA, Trieste) Bart Geurts (University of Nijmegen, Nijmegen) Ivana Kruijff-Korbayova (Universität des Saarlandes, Saarbrücken) Marco Kuhlmann (Universität des Saarlandes, Saarbrücken) Mª Magdalena Ortiz de la Fuente (Universidad de las Americas, Puebla) Benjamin Spector (Laboratoire de Linguistique Formelle, Université Paris 7, ENS) IMPORTANT DATES: Deadline for submission of papers : February 22, 2004. Authors notifications : April 19, 2004. ESSLLI early registration : May 1, 2004. Final papers for proceedings : May 15, 2004. ESSLLI-2004 Student Session : August 9-20, 2004. ESSLLI-2004 INFORMATION: In order to present a paper at ESSLLI-2004 Student Session, at least one student author of each accepted paper has to register as a participant at ESSLLI-2004. The authors of accepted papers will be eligible for reduced registration fees even after the deadline for early registration. For all information concerning ESSLLI-2004, please consult the ESSLLI-2004 main website at http://esslli2004.loria.fr/ For more information about the ESSLLI'04 student session, consult the ESSLLI'04 student session website at: http://lingua.fil.ub.es/~lalonso/stusESSLLI04/. If you have any further question about the student session, do not hesitate to contact us: Laura Alonso i Alemany Paul Egré lalonso(a)fil.ub.es paulegre(a)magic.fr Departament de Lingüística General IHPST / Université Paris 1 / Institut Jean-Nicod Gran Via de les Corts Catalanes, 585 Institut d'Histoire et de Philosophie des Sciences et Techniques Universitat de Barcelona 13, rue du Four Barcelona, SPAIN 75006 Paris, FRANCE telephone number: +34 93 403 56 93 telephone number: + (33) (0)1 43 54 60 36 fax number: +34 93 318 98 22 fax number: + (33) (0)1 44 07 16 49

1 0