- Haskell - Haskell.org

Research Associate position available
by Olaf Chitil 06 Dec '04

06 Dec '04

Research Associate Computing Laboratory, University of Kent at Canterbury, UK £19,460 - £21,640 pa Full-time and fixed term for 30 months, to start early 2005 The post is in association with the EPSRC project "A Theory of Tracing Pure Functional Programs". The overall aim of the project is to develop a semantical theory of tracing pure functional languages, eager and lazy, including: tractable formal definitions of traces, views of traces and fault location methods; theorems proving the correctness of the methods. The theory will allow existing systems to be improved and novel systems to be built. Candidates should have a good working knowledge of programming language semantics, in particular operational semantics for functional languages. A PhD degree in Computer Science (or related) is highly desirable. Enquiries about the project should be addressed to Dr Olaf Chitil, O.Chitil(a)kent.ac.uk <mailto:O.Chitil@kent.ac.uk?subject=Reply%20sent%20viajobs.ac.uk%20ref:R05/11> and further information can be found at http://www.cs.kent.ac.uk/people/staff/oc/. *Closing date for receipt of completed applications is: 12 noon Friday, 7 January 2005. Note that this project grows out of my previous work with Colin Runciman and Malcolm Wallace at York on the Haskell tracer Hat (http://www.haskell.org/hat) While the project itself is theoretical and more general, knowledge of Haskell would be beneficial. Olaf *

1 0

ANN: MissingH 0.7.2 with GZip
by John Goerzen 04 Dec '04

04 Dec '04

Hello, I'm pleased to announce the release of MissingH 0.7.2, available from http://quux.org/devel/missingh. This release incorporates Ian Lynagh's pure-Haskell Inflate algorithm, and CRC-32 and GZip file parsers of my own design, to make a pure-Haskell solution[1] to decompressing .gz files. At present, it is rather slow, as there has been little effort to optimize any of the three components. Patches to address the speed will be happily applied. The Inflate and CRC32 algorithms also provide the necessary support to be able to handle ZIP files. I plan to introduce ZIP and tar file support in upcoming releases of MissingH. Also, MissingH 0.7.0 was not announced on the library list. It introduced the ConfigParser[2] module. This module can parse the flat or hierarchical configuration files made popular by Python's ConfigParser module. The files are simple for both humans and programmers <grin>. If care is taken, it is possible to craft files that are valid both to ConfigParser and to other systems such as the Unix shell or Make. ConfigParser can also completely parse various other formats "out of the box", including SMTP headers, Cabal Setup.description files, and, to a limited extent, /etc/passwd files. All code in MissingH is pure Haskell. There is no C required. [1] http://quux.org/devel/missingh/html/MissingH.FileArchive.GZip.html [2] http://quux.org/devel/missingh/html/MissingH.ConfigParser.html -- John

1 0

more haskell in the news (for nerds)
by isaac jones 04 Dec '04

04 Dec '04

(please followup-to haskell-cafe) This slashdot story mentions Haskell being used to solve puzzles: http://developers.slashdot.org/developers/04/12/04/0116231.shtml?tid=159&ti… I did a quick search on slashdot to discover that Haskell has been in the topic of a slashdot story 4 times in the last 10 months (since march 2003). 5 times if you count a review of "Purely Functional Data Structures". In the year before that, it was mentioned only once in passing. In 2001, it was the topic once, and mentioned once in passing. In 2000 there were several stories on functional programming that mentioned Haskell. Any way you look at it, Haskell has gotten a lot more attention on Slashdot lately than it has since the beginning of the archives. Here's a link to the stories: http://slashdot.org/search.pl?tid=&query=haskell&author=&sort=1&op=stories peace, isaac

1 0

ANN: MissingH 0.7.0 with new ConfigParser
by John Goerzen 02 Dec '04

02 Dec '04

MissingH 0.7.0 is now available from http://quux.org/devel/missingh MissingH is a collection of Haskell-related utilities, all written in pure Haskell with an eye towards portability. The major new feature in this version is the ConfigParser module, which is a clean implementation of the Python ConfigParser module for Haskell. Its format is such that it can parse, out of the box, INI-style files, passwd, Cabal description files, etc. Docs are at http://gopher.quux.org:70/devel/missingh/html/ -- John

1 0

CFP: FOAL 2005, Foundations Of Aspect-Oriented Languages
by Ralf Laemmel 02 Dec '04

02 Dec '04

FOAL: Foundations of Aspect-Oriented Languages A one day workshop affiliated with AOSD 2005 in Chicago, USA, on Monday, 14 March 2005. Themes and Goals FOAL is a forum for research in foundations of aspect-oriented programming languages. Areas of interest include but are not limited to: * Semantics of aspect-oriented languages * Specification and verification for such languages * Type systems * Static analysis * Theory of testing * Theory of aspect composition * Theory of aspect translation (compilation) and rewriting The workshop aims to foster work in foundations, including formal studies, promote the exchange of ideas, and encourage workers in the semantics and formal methods communities to do research in the area of aspect-oriented programming languages. All theoretical and foundational studies of this topic are welcome. The goals of FOAL are to: * Make progress on the foundations of aspect-oriented programming languages. * Exchange ideas about semantics and formal methods for aspect-oriented programming languages. * Foster interest within the programming language theory and types communities in aspect-oriented programming languages. * Foster interest within the formal methods community in aspect-oriented programming and the problems of reasoning about aspect-oriented programs. Workshop Format The planned workshop format is primarily presentation of papers and group discussion. Talks will come in three categories: long (30 minutes plus 15 minutes of discussion), short (20 minutes plus 5 minutes of discussion) and very short (7 minutes plus 3 minutes of discussion). The very short talks will allow for short presentations of topics for which results are not yet available, perhaps for researchers who are seeking feedback on ideas or seek collaborations. We also plan to ensure sufficient time for discussion of each presentation by limiting the number of long talks and having only a few short talks. Submissions Invitation to the workshop will be based on papers selected by the program committee; those wishing to attend but not having a paper to submit should contact the organizers directly to see if there is sufficient space in the workshop. FOAL solicits full-length, short, and very short papers on all areas of formal foundations of AOP languages. Submissions will be read by the program committee and designated reviewers. Papers will be selected for long, short, and very short presentation at the workshop based on their length, scientific merit, innovation, readability, and relevance. Papers previously published or already being reviewed by another conference are not eligible. Some papers may not be selected for presentation, and some may be selected for presentation in shorter talks than their paper length would otherwise command. We will limit the length of paper presentations and the number of papers presented to make sure that there is enough time for discussion. Papers presented at the workshop will be included in a technical report (from Iowa State University). Authors will retain their own copyright to the papers. Publication of papers at other venues will thus remain possible. We will also investigate having a special issue of a journal for revisions of selected papers after the workshop. Authors should note the following details: * Submissions are due no later than 23:00 GMT, Sunday, 16 January 2005. This is a firm deadline. * Authors must indicate whether they wish to be considered for a long, short, or very short presentation. * Papers for long presentations must not exceed 10 pages in length; those for short presentations must not exceed 5 pages in length, and those for very short presentations must not exceed 3 pages in length. * Some papers may not be selected for presentation, and some may be selected for presentation in shorter talks than requested. * We encourage use of the ACM Conference format for submissions, as this will be required for accepted papers. You must add page numbers (which are not part of the standard format) to your submissions, to make adding comments easier. * Submissions are to be sent as PDF (preferred) or postscript attachments in an email to Curtis Clifton, cclifton -at- cs -dot- iastate -dot- edu. We will notify the corresponding author of papers that are selected for presentation at the workshop by 3 February 2005. The early registration deadline for AOSD will be 10 February 2005. FOAL attendees must be registered for AOSD. Final versions of papers for the proceedings will be due on 1 March 2005. For more information, visit the FOAL Workshop home page at http://www.cs.iastate.edu/FOAL. Important Dates Submission Deadline 23:00 GMT, 16 January 2005 Notification of Acceptance 3 February 2005 Final Versions of Papers Due 1 March 2005 Workshop 14 March 2005

1 0

controlling the scope of instances
by Ralf Laemmel 02 Dec '04

02 Dec '04

Hi, I refer to Section 5.3 (title see subject line) in Type classes: exploring the design space by Simon Peyton Jones, Mark P. Jones and Erik Meijer HW 1997. There it is argued very briefly that a type system with controlled scope of instances risks coherence. Likewise, in Wadler's and Blotts original paper on ad-hoc polymorphism, Appendix A.7 sheds light on the related problem of principal typings. Global scope is suggested while wondering about less restrictive conditions that retain principal typing. The interaction with the module system is not really considered. I also know of /Functional Pearl: Implicit Configuration -- or, Type Classes Reflect the Value of Types/ Oleg Kiselyov, Chung-chieh Shan, Haskell Workshop 2004, which sheds some fancy light on the subject. The paper deals with local instances, but not so much with scoping via the module system. Does anyone know of a published, somewhat deeper analysis regarding the possibility to scope instances via the normal module system? For instance, why can't we have as many distinguished "incarnations" of a class in the type system as there are different sets of instances created in various scopes. That is, references to methods are to be *implicitly qualified* by the corresponding class incarnation (i.e., sets of instances). Methods of different incarnations are not to be confused by the type system. This would solve all coherence issues that I can think of. Or is this just getting too complicated? Any pointers greatly appreciated. Regards, Ralf **

1 0

RE: [Haskell] discrepancy between nhc and ghc
by Simon Peyton-Jones 02 Dec '04

02 Dec '04

The Report explicitly specifies that you can export a class method "naked", for exactly the reason you give. So I think GHC is right here. Simon | -----Original Message----- | From: haskell-bounces(a)haskell.org [mailto:haskell-bounces@haskell.org] On Behalf Of John Meacham | Sent: 02 December 2004 01:28 | To: haskell(a)haskell.org | Subject: [Haskell] discrepancy between nhc and ghc | | | | ghc: | import Random(Random(randomR)) -- good | import Random(randomR) -- good | | nhc: (1.16) | import Random(Random(randomR)) -- good | import Random(randomR) -- ERROR! nhc says randomR is not found. | | | I was wondering whether this was intentional and which was correct? I | much much prefer the ghc behavior since it lets you export methods | without exporting the class and lets you hide the implementation detail | of whether a function is implemented as a method or normal declaration. | | | John | | -- | John Meacham - ⑆repetae.net⑆john⑈ | _______________________________________________ | Haskell mailing list | Haskell(a)haskell.org | http://www.haskell.org/mailman/listinfo/haskell

1 0

why no strictness annotations in labelled fields?
by S. Alexander Jacobson 02 Dec '04

02 Dec '04

Is there a good reason one can't do: data Foo = Foo {bar::!String} -Alex- ______________________________________________________________ S. Alexander Jacobson tel:917-770-6565 http://alexjacobson.com

2 1

discrepancy between nhc and ghc
by John Meacham 02 Dec '04

02 Dec '04

ghc: import Random(Random(randomR)) -- good import Random(randomR) -- good nhc: (1.16) import Random(Random(randomR)) -- good import Random(randomR) -- ERROR! nhc says randomR is not found. I was wondering whether this was intentional and which was correct? I much much prefer the ghc behavior since it lets you export methods without exporting the class and lets you hide the implementation detail of whether a function is implemented as a method or normal declaration. John -- John Meacham - ⑆repetae.net⑆john⑈

1 0

pattern matching accross instance declarations
by S. Alexander Jacobson 02 Dec '04

02 Dec '04

Doing read/show on existential types would be a lot easier if pattern matching worked *accross* instance declarations and not only within them For example, this code produces an "Ambiguous type variable" GHC error, but it would be really helpful if it didn't: data MyExistantialType=forall v.(Show v)=>EType v class MyExistentialTypeable a where toMyType::String->MyExistantialType instance MyExistentialTypeable String where toMyType "String" = EType "foo" instance MyExistentialTypeable Int where toMyType "Int" = EType "bar" tVal (EType v)=show v main = print $ tVal $ toMyType "String" I know Haskell has a general prohibition on defining functions in multiple locations, but instance declarations already violate it. Why not go all the way? My current approach is manually writing lookup functions in addition to per-instance functions and that is both annoying and error-prone. -Alex- ______________________________________________________________ S. Alexander Jacobson tel:917-770-6565 http://alexjacobson.com

2 2