Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
Manuel M T Chakravarty
[go: Go Back, main page]

Submit a paper to
Trends in Functional Programming (TFP 2008)!
[submission deadline: 3rd March 2008 - PASSED]

News and Current Work

[April 2008] Type Checking with Open Type Functions describes our extension of GHC's type checker with support for type synonym families. (Accepted for ICFP'08.)

[February 2008] GHC now supports the creation of Mac installer packages.

[November 2007] Duncan Coutts just released C->Haskell version 0.15.1, which builds with GHC 6.4 through to 6.8 and also properly supports OpenBSD and Windows. Morover, the documentation has been updated.

[November 2007] Partial Vectorisation of Haskell Programs introduces a novel method to vectorise irregular array programs for parallel execution when vectorisation is only possible for part of the program. (Accepted for presentation at DAMP'08.)

[November 2007] Final version of Specialising Simulator Generators for High-Performance Monte-Carlo Methods. This paper describes how we designed and implemented the world's fastest Monte-Carlo simulator for polymerisation kinetics - in Haskell. (To appear in PADL08.)

[October 2007] ML Modules and Haskell Type Classes: A Constructive Comparison (main credit to Stefan Wehr). The first rigorously formal comparison between ML modules and Haskell type classes. Besides settling the question as to how these two language features compare, the paper suggests how ML and Haskell may be extended. (Submitted.)

[September 2007] New paper making progress Towards Open Type Functions for Haskell. (Presented at IFL'07.)

[August 2007] Type synonym families are now available in GHC's development version.

[August 2007] Duncan Coutts just released C->Haskell version 0.15.0, which includes his new parser with excellent gcc compatibility and an improved build system.

[August 2007] Specialising Simulator Generators for High-Performance Monte-Carlo Methods is thoroughly revised version of our previous technical report on this topic, which has more detail on the implementation of the polymerisation simulator. (To appear in PADL08.)

[June 2007] New draft paper on Solving the Expression Problem with True Separate Compilation is available.

[June 2007] In joint work with the Centre for Advanced Macromolecular Design @ UNSW, we describe A Parallelised High Performance Monte Carlo Simulation Approach for Complex Polymerisation Kinetics - see also the project page at CAMD which offers the source code of the system for download. (To appear in Macromolecular Theory & Simulation.)

[May 2007] I started the tumblelog called Just Testing. Somehow I want to integrate with this page, but I am not sure how...

[April 2007] Technical report on Generative Code Specialisation for High-Performance Monte-Carlo Simulations. We show how to outperform high-performance implementations in low-level languages, such as C, by implementing simulator generators in Haskell.

[March 2007] The first release, version 0.1, of the concurrent, high-performance library package ndp was released today. For further information, see the Data Parallel Haskell homepage.

[March 2007] Folks interested in using nested data parallelism for programming many-core architectures may want to have a look at my talk Compiling Nested Data Parallelism by Program Transformation presented at the Data-Parallel Programming Models for Many-Core Architectures workshop. (The slides for all talks from the workshop are also available online.)

[January 2007] The final version of Data Parallel Haskell: a status report describes the design, implementation, and first results of our current effort to realise nested data parallel programming for Haskell on SMP, and especially, multicore computers. (To appear at DAMP 2007.) [More information about Data Parallel Haskell is available.]

[November 2006] Final version of System F with Type Equality Coercions (with Martin Sulzmann, Simon Peyton Jones, and Kevin Donnelly). We show that System FC, an extension of System F with type-equality coercions, is a suitable typed intermediate language for compiling generalised abstract data types, functional dependencies, and associated types. This is GHC's new IL; see also the corresponding GHC devel wiki page. (To appear at TLDI'07.)

[November 2006] Learn how to use type-indexed data types, a moderate extension to our earlier proposal for associated data types, with GHC.

[October 2006] Final version of Modular Type Classes (main credit to Derek Dreyer and Robert Harper). We reconstruct Haskell-style type classes as a mode of use of ML-style modules, providing a seamless integration of these two important linguistic tools. (To appear at POPL'07.)

[October 2006] Curious about array programming and data parallelism in the Glasgow Haskell Compiler? Check out the new Haskell wiki page on Data Parallel Haskell.

[September 2006] Type equality coercions are now fully implemented in GHC. They are used to implement newtypes, GADTs, and indexed data types and newtypes (including associated data types). We are still working at implementing general type functions (i.e., indexed type synonyms).

[August 2006] Type equality coercions are now being put to good use inside GHC with GADT support being functional and type functions being worked on.

[July 2006] Preprint of Running the Manual: An Approach to High-Assurance Microkernel Development (main credit to Philip Derrin, Kevin Elphinstone, Gerwin Klein, and David Cock). We introduce a development model of rapid prototyping and iterative refinement of a microkernel in Haskell, such that real user-level code can be executed on a simulator and a formal specification for Isabelle/HOL can be derived semi-automatically (Appeared at 2006 Haskell Workshop.)

More publications...
More software...

Roles

Free and Open Source Software

GNU Project Fedora Project ohloh profile for chak
Linux Rules! slashdot.org
haskell.org

Manuel M T Chakravarty

aka Chilli :-)
University of New South Wales
Faculty of Engineering
School of Computer Science & Engineering
Programming Languages and Systems
How to reach me/where to find me
Please no attachments in proprietary formats.
Here is why.

Quotes

"Functional programming is an attitude, not a fashion style."

Don Stewart on his blog

"In a concurrent world, imperative is the wrong default!"

Tim Sweeney (of Epic Games). The Next Mainstream Programming Language: A Game Developer's Perspective. Invited talk at ACM SIGPLAN-SIGACT Principles of Programming Languages (POPL'06).

More quotes...

Disclaimer

These pages reflect my personal, but not necessarily my affiliation's point of view.

• Copyright [2005..2006] Manuel M T Chakravarty • Last modified: Sun Apr 8 14:42:47 EST 2007