- Haskell-Cafe - Haskell.org

Haskell's type system compared to CLOS
by Matthias-Christian Ott 11 Aug '09

11 Aug '09

Hi, usually I'm sceptical of programming languages which are not based on the von Neumann architecture, but recently I got interested in functional programming languages. The arrogance of lots of Haskell users, who made me feel that using a programming language other than Haskell is a waste of time, combined with vanguard mathematical notation has been very attractive to me and I have decided to get at least an idea of Haskell and its concepts. Some weeks ago I learned programming in Dylan and was impressed by its object system which is basically a stripped version of CLOS. Multiple dispatch together with a well-thought-out object system is quite powerful, because it removes the burden of including methods in the class definition. At the moment I'm reading the "Functional Programming using Standard ML" and I'm in the chapter on data types. This afternoon it occurred to me that classes and data types are symmetric. In a class hierarchy you start an abstract super class (the most abstract is the class object in some languages) and further specialise them via inheritance; with data types you can start with specialized versions and abstract them via constructors (I'm not sure how message sending to a superclass looks like in this analogy). Anyhow, I also came across an interesting presentation. Andreas Löh and Ralf Hinze state in their presentation "Open data types and open functions" [1]: * OO languages support extension of data, but not of functionality. * FP languages support extension of functionality, but not of data. Their first point refers to the fact that in most object-oriented languages don't allow the separate definition of classes and their respective methods. So to add new functions, you have edit the class definitions. However, in functional programming languages you can easily add new functionality via pattern matching, but have to either introduce new types or new constructors, which again means to modify existing code. In Dylan (and in Common Lisp) you define methods separate from classes and have pattern matching based on types. This solves all mentioned problems. So my question is, how are algebraic data types in Haskell superior to CLOS (despite the fact that CLOS is impure)? How do both compare? What has Haskell to provide what Common Lisp and Dylan haven't? Thanks! Regards, Matthias-Christian [1] http://people.cs.uu.nl/andres/open-japan.pdf

2 1

Galois is Hiring
by Don Stewart 11 Aug '09

11 Aug '09

Galois is continuing to hire. We will also be at ICFP and related events, so come and see Lee Pike or me. We have multiple positions for talented functional programmers (with both junior and senior positions). A strong functional programming ability is an advantage (you will be programming in Haskell), and a good computer science background is required. Experience with compilers, operating systems, networking, security, and formal methods is of particular interest. Galois is based in Portland, Oregon. Our engineers work with functional languages, designing and developing advanced technologies for safety and security-critical systems, networks, and applications. Galois technical staff members play a pivotal role in developing advanced software technology. Enginners work in small team settings, and must successfully interact with clients, partners, and other employees in a highly cooperative, collaborative, and intellectually challenging environment. Technical staff may be called upon to write proposals, gather requirements, and work in all stages of the software development process, from requirements gathering to testing and validation. Additional duties may include project management, technology research and development, and technical infrastructure development. We’re looking for people who can invent, learn, think, and inspire. We reward creativity and thrive on collaboration. We offer great benefits and perks, including a 401K plan, stock options, paid vacation, family health plan, flexible work schedule, a casual work environment, snacks, espresso and foosball. A Masters or Ph.D. degree in Computer Science is desirable. Additionally, a strong programming background and experience with Haskell or other functional programming languages is preferred. You must work well with customers, including building rapport, identifying needs, and communicating with strong written, verbal and presentation skills. Must be highly motivated and able to self-manage to deadlines and quality goals. To learn more about us, visit http://www.galois.com and http://www.galois.com/company/careers The types of technology we use are covered in our blog: http://www.galois.com/blog/ We’d like to hear from you! Send your cover letter and resume to us at jobs2009 at galois.com.

1 0

GTK2HS OpenGL demo doesn't work on Vista when compiled?
by Peter Verswyvelen 11 Aug '09

11 Aug '09

GTK2HS 0.10.1 comes with an OpenGL demo (share/demo/opengl/RotatingCube.hs) On my machine (Vista 32-bit) it runs fine under GHCi, but not when compiling using GHC (no rendering happens, unless I resize the window, then rendering happens from time to time) I'm using GHC 6.10.3 (installed Haskell Platform 2009.2.0.1). Do other people using Windows experience this problem? Cheers, Peter Verswyvelen

1 0

Re: Any comments about Clojure language?
by Mark Volkmann 11 Aug '09

11 Aug '09

> On Mon, 10 Aug 2009 09:35:48 -0700, Don Stewart <dons at galois.com> > wrote: > > Another possible disadvantage of Clojure is the lack of high-quality > free online tutorials. I wrote an extensive intro. to Clojure at http://ociweb.com/mark/clojure/article.html. -- R. Mark Volkmann Object Computing, Inc.

1 0

Re: Re[2]: [Haskell-cafe] unsafeDestructiveAssign?
by Job Vranish 11 Aug '09

11 Aug '09

Opps, didn't "reply to all" I hate it when that happens... On Tue, Aug 11, 2009 at 1:00 PM, Job Vranish <jvranish(a)gmail.com> wrote: > haha, yeah. > > What I'm actually trying to do is make a way to manipulate arbitrary > recursive algebraic data structures as if they were graphs (not for > performance reasons but because it makes manipulating them much easier) > > It's similar to the data-reify package, but I need it to work on any > algebraic datatype. > > I'm not planning on using it for actually production code. I'm mostly just > experimenting with as a potentially interesting language feature. > > - Job > > > On Tue, Aug 11, 2009 at 12:50 PM, Bulat Ziganshin < > bulat.ziganshin(a)gmail.com> wrote: > >> Hello Job, >> >> Tuesday, August 11, 2009, 8:45:29 PM, you wrote: >> > What I am trying to do is hyper unusual and I really do need an >> unsafeHorribleThings to do it. >> >> yes, assigning 5 to 8 is really unusual in math :) >> >> >> -- >> Best regards, >> Bulat mailto:Bulat.Ziganshin@gmail.com >> >> >

1 0

Re: Examples
by John Lato 11 Aug '09

11 Aug '09

Hello, Looking at just examples, there are a few separate issues here. 1. Installation of example source for review 2. Building/installation of example programs For the second issue, the buildExamples flag suggested by Henning works well to actually build the examples with cabal, but the install location is still a problem. Most users wouldn't want example executables to go into their bin directory, especially not in the case of global installs. To the best of my knowledge, Cabal currently doesn't support separate installation locations for multiple executables, nor does it support building but not installing executables. With the first issue, the problem is simpler. There's no location for source code to be installed to. Users can download the package and unpack it themselves, or the developer can include demo source as a "data file" and install it to the datadir (IMO a poor choice). One option is to add support for different installation locations for each executable. Perhaps the --installdir flag could be extended to support a syntax like the following: --installdir=executableName:/path/to/install where if executableName matches one of the executables known to cabal, it will install just that executable to the specified location. This puts a larger burden on users, however, since they would need to know the name(s) of all the demo executables. It also doesn't address the issue of source location. Another way is to add a special type of executable for tests and demo programs to cabal. A user would specify a directory in the .cabal file (or command-line) to which tests would be installed. Cabal would then build the identified tests/demos for the package and put them into that specific directory, as well as the source code. This also answers the question of where to put executables that are Buildable but not Installable. Thoughts? Cheers, John Lato > Date: Sun, 9 Aug 2009 21:07:53 -0400 > From: "John D. Ramsdell" <ramsdell0(a)gmail.com> > Subject: Re: [Haskell-cafe] Examples > To: Haskell Cafe <haskell-cafe(a)haskell.org> > Message-ID: > <7687290b0908091807y1d4a5329w6b452c6a3af5e044(a)mail.gmail.com> > Content-Type: text/plain; charset=UTF-8 > > Maybe in addition to having a buildable boolean in a library or > executable section, there should be an installable boolean. It would > default to true, but when false, the library or executable section is > ignored during package installation. > > John >

1 0

Haskell2Xml
by Dmitry Olshansky 11 Aug '09

11 Aug '09

Hello all, I need a convenient tool to generate Haskell types from XML W3C Schema Definition (xsd) and vice versa - generate instances for Haskell ADT's to make corresponding XML. It is just the same that HaXml do with DTD. I need - using XSD - support for unicode - using xml-attributes as far as elements are very desirable - by my opinion TemplateHaskell for generate instances is prefferable than using DrIFT (which used in HaXml) - both possibilities is great of course - generation of xsd by Haskell type is a good feature also and so on... Does this tool exist? Do some articles / thoughts / standards / recomendations about correspondence between XML and Haskell ADT's exist? Best wishes, Dmitry

7 11

C2HS issues on Hackage
by Michael Snoyman 11 Aug '09

11 Aug '09

Hi all, I've written a Yaml library built on top of libyaml (the same C library at the core of Python's yaml library). All is well on my local system and my server; I'm currently using it in production. However, the build is failing on Hackage with the following build log: http://hackage.haskell.org/packages/archive/yaml/0.0.1/logs/failure/ghc-6.10 . It seems that whichever version of C2HS is installed on the server does not allow "type" (or perhaps other keywords?) to be used as C identifiers. Unfortunately, "type" is the name of a field in a C struct I need to integrate with. It's obviously not vital that I get it to compile on hackage, but 1) I'm worried this is a sign of a bigger problem, and 2) it's always nice to have the API documentation available online. Any suggestions? Thanks, Michael

2 2

Any comments about Clojure language?
by Sukit Tretriluxana 11 Aug '09

11 Aug '09

Hi all, I start reading about Closure language (http://clojure.org) and it seems an interesting language. I don't know much about this language especially in comparison to Haskell feature by feature. Could it perhaps be what Haskell on JVM would have been with the dressing of Lisp syntax? Any one would like to chime in your comments about the language, in comparison to Haskell? Thanks, Ed

4 3

Re: Announce: EnumMap-0.0.1
by Louis Wasserman 10 Aug '09

10 Aug '09

Pardon my asking, but are fromEnum and toEnum guaranteed to preserve order on types like Double? I'd like to see a Double-backed Patricia tree map, but are we certain that EnumMap as presented will behave properly on such types? Louis Wasserman wasserman.louis(a)gmail.com

3 2