February 2006 - Haskell

Type synonyms in the instance head
by Mads Lindstrøm 02 Feb '06

02 Feb '06

Hi all In this url http://www.haskell.org/ghc/docs/latest/html/users_guide/type-extensions.htm… one can read about "Type synonyms in the instance head". A quote: "You cannot, for example, write: type P a = [[a]] instance Monad P where ... This design decision is independent of all the others, and easily reversed, but it makes sense to me." It does not make sense to me! Could anybody explain why this makes sense? Below, I will try to explain why it does not make sense to me. If I have: data LongAndUglyName a = ... type ShortAndTellingName a = LongAndUglyName a It seems to make sensible to: instance Foo ShortAndTellingName Also it is called a type _synonym_ and the type synonym and what it is synonym for, should therefore be interchangeable. This restriction do not only apply to instance heads, but also to contexts, as in: instance (Foo LongAndUgleName) => class something ... Here again LongAndUglyName cannot be replaced by ShortAndTellingName. This can be annoying if you have a lot of these instance declarations. Also what the type synonym is synonym for, may not have any name and thus properly not be telling at all, as in: type TellingName a = \a -> String /Mads Lindstrøm

4 3

Idioms and Proposals
by Ashley Yakeley 01 Feb '06

01 Feb '06

I've added a few Haskell idioms and proposals (well, ideas for changes) to the wiki: http://haskell.org/haskellwiki/Category:Idioms http://haskell.org/haskellwiki/Category:Proposals I particularly wanted to include "perennial" ideas that often seem to get discovered independently (for instance, type arithmetic). Of the proposals, I think only MonadPlus reform might be worth moving to the Haskell Prime process. Please add, improve, etc.

1 0

Internships at Microsoft Research, Cambridge
by Simon Peyton-Jones 01 Feb '06

01 Feb '06

Internships are available at Microsoft Research, Cambridge, for this summer. More information is available here: http://research.microsoft.com/aboutmsr/jobs/internships/ Internships are usually for graduate students (PhD or Masters) and last 12 weeks (although longer projects are also possible). Furthermore, although there is a deadline of *28 Feb* for the main batch of summer internships, MSR is now taking interns year-round, a non-summer internship may even be easier to arrange (because it misses the Big Rush). The reason for this mail is to highlight the fact that Simon & I would particularly like to encourage applicants interested in working on Haskell-related projects. We have a number of projects in mind, but feel free to suggest your own: - Parallel garbage collection: to complement GHC's new support for SMP parallelism, replacing single-threaded garbage collection with parallel GC is a high priority goal, with plenty of interesting research too. - Parallel Haskell applications: investigate using SMP/multicore parallelism for real, either implicit parallelism or using concurrency & STM, by tackling some applications. - Haskell debugging in GHCi: we have some ideas for building a debugger into GHCi, there are some research angles here too. - GHC as a library: we have made a start on providing a library interface to GHC, but there is much left to do. Re-targetting Haddock on top of GHC would make a good project, for example. - C-- code generation: we have a project student working on C-- code gen, but there's a lot to do here. The ultimate goal is to divide the code generator into two: first generate C-- and then CPS-convert it into C-- with only tail calls, suitable for generating code using the existing paths (native code and via-C) too. Let us know by email if you apply. Good luck! Simon Peyton Jones and Simon Marlow

1 0

TYPES 2006 Workshop
by peter morris 01 Feb '06

01 Feb '06

Registration for TYPES 2006 is now open. Early registration until 15/3/2006. Cheers, The TYPES06 Team -------------------------------------------------------------------------- TYPES 2006 Main Conference of the Types Project Nottingham, UK, 18-21 April 2006 http://www.cs.nott.ac.uk/types06/ This is the latest meeting in a series that started 1992, the last conference was in December 2004 in Paris. The topic of the meeting is formal reasoning and computer programming based on Type Theory : languages and computerised tools for reasoning, and applications in several domains such as analysis of programming languages, certified software, formalisation of mathematics and mathematics education. TYPES 2006 is colocated with TFP 2006 (Trends in Functional Programming) and we plan to hold a joint session on Dependently Typed Programming. The conference takes place at Jubilee campus of the University of Nottingham, on-site accomodation will be available together with the registration. For more information see: http://www.cs.nott.ac.uk/types06/ Registration is open now. You will be able to submit your talk and abstract together with your registration. We will try to accomodate all talks which fit into the scope of the TYPES project. There will also be invited lectures from Bart Jacobs, Simon Peyton Jones (joint with TFP) and Hongwei Xi. Please direct all emails related to TYPES 2006 to types06(a)cs.nott.ac.uk Cheers, The Organisation Comittee Thorsten Altenkirch, James Chapman, Conor McBride, Peter Morris and Wouter Swierstra .

1 0

Everything Your Professor Failed to Tell You About Functional Programming
by Shannon -jj Behrens 01 Feb '06

01 Feb '06

Shameless plug: I wrote an article for "Linux Journal" called "Everything Your Professor Failed to Tell You About Functional Programming". Hopefully you will enjoy the article. http://www.linuxjournal.com/article/8850 (By the way, I'm not on the cafe mailing list, so please cc me on any replies.)

3 2