- Haskell-Cafe - Haskell.org

ANN: MIME Strike Force
by Jeremy Shaw 20 Mar '07

20 Mar '07

Hello, If you have tried to do any MIME processing using Haskell, you are likely to have found two things: 1) There are a lot of MIME libraries for Haskell 2) None of them do everything you want So, I propose that we form a MIME Strike Force and create the one, true MIME library. The plan is something like this: 1) Document all the things the MIME library needs to support. 2) Pick the technology, and design the infrastructure for supporting these features. For example, I don't think we will be able to use Parsec because: i) We want to support ByteString ii) We want to be able to lazily parse the message 3) Try to reuse as much existing code as possible to implement the design. I have started step 1 by creating this page: http://www.haskell.org/haskellwiki/Libraries_and_tools/MIMEStrikeForce Please add your comments, requirements, etc. If you are interesting in helping contrib ideas, code, or flames, please let me know. If there is enough interest, we should probably set up a mailing list for discussion. j. ps. This *might* make a decent SoC project, but only if it results in the one, true MIME library. We definitely do not need another incomplete MIME library floating around.

4 3

Matrices in Haskell
by Ivan Miljenovic 19 Mar '07

19 Mar '07

Some of you might recall me annoying people on #haskell over the New Year about my Latin Squares project. Well, I'm looking at re-doing it from scratch, but the first thing I need to do is find a new way of representing my square. I have been using a list of lists ([[a]]) to represent a matrix. The problem with this data structure is that I need to be able to manipulate matrices as both row-oriented and column-oriented data structures, as I need to examine the values in the columns as well as in the rows. As it stands, I was doing this by transposing the matrix, manipulating it, then transposing it back again. This is a pain, as it takes up about 15% to 20% of the run time. The other problem with using a list of lists is that the only reason I'm sure that the matrix is valid (i.e. all the rows are the same length, etc.) is because I created it that way, not because the data structure requires it. As such, I'd like to know if there's any way of storing a an n-by-n matrix such that the algorithm/function to get either the rows or the columns is less than O(n^2) like transposition is. I did try using an Array, but my (admittedly hurried and naive) usage of them took longer than a list of lists, was more difficult to manipulate, and wasn't required separate inverse functions to row and cols. Also, since I need to look all throughout the list of values, it's still O(n^2), but this time for both rows and columns. I know that when doing something similar to this in Java a few years ago (a primitive Sudoku solver, to be precise), I could represent the rows and columns as to separate 2D arrays, with rows(i,j) pointing to the same object as cols(j,i). Is something like this possible in Haskell? in particular, I will be storing lists of possible values in the cells of the matrix, so the capability to backtrack would be very nice, as I'd be trying each value at a time to see if it is valid. I'd also want to use such a matrix implementation next semester for a project, which I plan on being a quick comparison of various programming languages as to ease of use and efficiency for matrix-based computing problems. -- Ivan Lazar Miljenovic

6 10

Re: There can be only one fix? Pondering Bekic's lemma
by oleg＠pobox.com 19 Mar '07

19 Mar '07

Nicolas Frisby wrote: > My question is: Given products and a fixed point combinator, can any > pure expression be transformed into a corresponding expression that > has just a single use of fix? Albert Y. C. Lai pointed out model-theoretical and CPU-practical answers. There is also a Haskell-automatic answer. That is, polyvariadic fixpoint, a combinator to obtain a mutually least fixpoint of several terms, is expressible in Haskell itself, via the regular fix: http://okmij.org/ftp/Computation/fixed-point-combinators.html#Poly-variadic I like its inferred type fix':: [[a->b]->a->b] -> [a->b] which is truly the type of the (applicative, or eta-expanded) regular fix with some round parentheses replaced with square ones.

1 0

Haskell and AI
by Adam Wyner 19 Mar '07

19 Mar '07

I agree with Andrew Wagner re: Haskell and AI. There are some relevant resources on Haskell, Maths, and Logic: *The Haskell Road To Logic, Maths And Programming (Paperback) * by Kees Doets <http://www.amazon.com/exec/obidos/search-handle-url/103-5196905-1677457?%5F…> (Author), Jan van Eijck <http://www.amazon.com/exec/obidos/search-handle-url/103-5196905-1677457?%5F…> (Author) http://www.amazon.com/Haskell-Road-Logic-Maths-Programming/dp/0954300696/re… van Eijck has another book on Computational Semantics. Elsewhere, there is work on learning algorithms in Haskell. Broadly, we might start by gathering known resources, projects, and people. We could consider one of the standard AI books as a guide for chapters, sections, and problems to fill in with Haskell approaches. I'd be very interested to contribute to this. Cheers, Adam Wyner Message: 7 Date: Mon, 19 Mar 2007 12:51:19 -0400 From: "Andrew Wagner" <wagner.andrew(a)gmail.com> Subject: Re: [Haskell-cafe] Haskell Chess To: haskell-cafe(a)haskell.org Message-ID: <b8a8636e0703190951y21c7cb01qcb4f5b58089b169d(a)mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Here's my take on this: I've thought for a while now that Haskell needs a general toolkit for AI. After all, functional programming has long been recognized for being good at AI, yet you rarely hear about it being done in Haskell. Anyway, my suggestion would be to concentrate on methods of AI. For example, if we implement alpha-beta polymorphically enough, it should be trivial to use it for any game that it makes sense to use it on. This kind of thing is what I was thinking of when I talked about some "fundamental design" ideas I had. Things like writing a Board type-class, so that you can implement any board representation you want to for chess. Or writing alpha-beta in terms of positions and moves, regardless of what kind of game you're talking about (I also think you simply must use unfoldTree, as this is a beautiful instance of it). Things like learning, and other general strategies, should be developed independently of any particular game, IMHO.

3 3

Re: [Haskell] Haskell Chess
by dons＠cse.unsw.edu.au 19 Mar '07

19 Mar '07

smazanek: > Hello again, > > I got a lot of interesting and useful comments on my posting > about Haskell Chess. Somebody suggested using the program > for benchmarks. Several people asked me to open the program > for contributions. And others were just interested in the exercises. > > It is probably the best to branch the development, so that we > have a teachers' and a hackers' version (ease of understanding > and learning paradigms vs. efficiency and implementation of > full ruleset). I will do the necessary steps next week (svn, etc). > Somebody already offered help for cabalizing hsChess. hsChess has been cabalised, and is available now via darcs; darcs get http://www.cse.unsw.edu.au/~dons/code/contrib/hsChess/ You can also browse the source directly: http://www.cse.unsw.edu.au/~dons/code/contrib/hsChess/ > Several people asked me for the exercises so I started the wiki > page > http://haskell.org/haskellwiki/Learning_Haskell_With_Chess > on this topic. I will add more content when I am back to office. > > Every contribution and discussion is welcome. I am happy to host the code at the above url, unless Steffen, you have a better place to host the repository? -- Don

7 11

Re: [Haskell-cafe] N and R are categories, no?
by Dominic Steinitz 19 Mar '07

19 Mar '07

> I haven't formally checked it, but I would bet that this endofunctor > over N, called Sign, is a monad: Just to be picky a functor isn't a monad. A monad is a triple consisting of a functor and 2 natural transformations which make certain diagrams commute. If you are looking for examples, I always think that a partially ordered set is a good because the objects don't have any elements. A functor is then an order preserving map between 2 ordered sets and monad is then a closure (http://en.wikipedia.org/wiki/Closure_operator) - I didn't know this latter fact until I just looked it up. Dominic.

5 5

N and R are categories, no?
by Steve Downey 19 Mar '07

19 Mar '07

EOk, i'm trying to write down, not another monad tutorial, because I don't know that much yet, but an explication of my current understanding of monads. But before I write down something that is just flat worng, I thought I'd get a cross check. (and I can't get to #haskell) Monads are Functors. Functors are projections from one category to another such that structure is preserved. One example I have in mind is the embedding of the natural numbers into the real numbers. The mapping is so good, that we don't flinch at saying 1 == 1.0. The functor that takes us from N to R is probably a Monad, that is, if N and R are categories. The real hard part is tying together how unit, join and bind produce a spacesuit that can protect apples from nuclear waste. I'm still getting that clear in my head, although my recent blinding flash of obviousness that M a is a type, and that of course types can do interesting things, I think gets me further along.

3 4

Algorithms
by P. R. Stanley 19 Mar '07

19 Mar '07

Hi "When you write a general solution for a class of problems, as opposed to a specific solution to a single problem, you have written an algorithm." Discuss! Regards, Paul

5 6

Haskell and AI
by Andrew Wagner 19 Mar '07

19 Mar '07

> I agree with Andrew Wagner re: Haskell and AI. > > There are some relevant resources on Haskell, Maths, and Logic: > > The Haskell Road To Logic, Maths And Programming (Paperback) > by Kees Doets (Author), Jan van Eijck (Author) > > http://www.amazon.com/Haskell-Road-Logic-Maths-Programming/dp/0954300696/re… > I liked this book. What parts did you have in mind specifically as being relevant to AI? > van Eijck has another book on Computational Semantics. Elsewhere, there is > work on learning algorithms in Haskell. > > Broadly, we might start by gathering known resources, projects, and people. > We could consider one of the standard AI books as a guide for chapters, > sections, and problems to fill in with Haskell approaches. I like this idea. On the other thread [1] Ricardo Herrmann suggested following the structure of "AI: A Modern Approach", the classical text [2] by Russell and Norvig. Do others agree? I think we should carve out a section of the wiki, in the Applications and Libraries section [3] for AI, which should then be split into (at least) a section each for relevant papers, code, interesting AI problems with their solutions, and a table-of-contents type page, structured according to a text like this. Comments? > > I'd be very interested to contribute to this. > Maybe we should assemble an AI strike force? :) > Cheers, > Adam Wyner > [1] http://www.haskell.org/pipermail/haskell-cafe/2007-March/023646.html [2] http://aima.cs.berkeley.edu/ [3] http://www.haskell.org/haskellwiki/Libraries_and_tools

1 0

System.Random StdGen read fails on some strings?
by Fritz Ruehr 19 Mar '07

19 Mar '07

According to the documentation for System.Random (see <http:// haskell.org/ghc/docs/latest/html/libraries/base/System-Random.html>): > In addition, read may be used to map an arbitrary string (not > necessarily one produced by show) onto a value of type StdGen. In > general, the read instance of StdGen has the following properties: > > * It guarantees to succeed on any string. > * ... On the other hand, I get the following on the (admittedly stupid, spur-of-the-moment) String argument "whateva": > Hugs> :l System.Random > System.Random> read "whateva" :: StdGen > > Program error: Prelude.read: no parse > > System.Random> map read $ words "this is a test of the System dot > Random StdGen read, which seems to fail only on ... whateva" :: > [StdGen] > [4580 1,440 1,101 1,4584 1,445 1,1485 1,35261 1,1377 1,32825 > 1,34047 1,13422 1,14037 1,13637 1,469 1,4132 1,4514 1,453 1,626 1, > Program error: Prelude.read: no parse Am I missing something here? Or am I just being punished for the stupidity of my particular choice of String? :) -- Fritz PS: this happens with both Hugs Version: September 2006 and GHC Interactive, version 6.6, running on Mac/Tiger, for what that's worth.

4 4