Hi fellow Haskellers, I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake. If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too. Purposes which I consider "Haskell for Haskell's sake" include: - writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :) Thanks, - Hal -- Hal Daume III | hdaume@isi.edu "Arrest this man, he talks in maths." | www.isi.edu/~hdaume
--- Hal Daume III <hdaume@ISI.EDU> wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
In our office we use Haskell for writing of tools but the main development is done in C++, C# and SQL. Probably we can use Haskell for large projects but there are still missing many features: - library for Graphical User Interface - Designer for dialogs and forms - Complete and comfortable IDE - Library and tools for Database access This is my motivation to start development of HToolkit. Krasimir __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com
----- Original Message -----
--- Hal Daume III <hdaume@ISI.EDU> wrote: Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
Simple. Haskell is the first language programming I ever learned, It is very easy to write because it does not have a gui or stuff that needs mouch knowledge, plus it is system free so I can compile it in any possible system (Windows, Unix, Linux...). Not to mention that I am not very good with programming which makes it easier for me (Funktionall Programming rules). Best Regards Alex
We are just a software company that builds multi platform (Unix - AIX Solaris HP-UX Linux, Windows) report generators for various formats, including Excel and PDF. Our product is not written in Haskell, but we do all our research by using Haskell as a prototype language for our ideas. In our minds, Haskell is the best language you can find now, and we thought to choose Haskell as a language for all our projects. But we decided not to do so because of couple things: 1. The language runtime that we have to include in every our projects 2. The language does still lack many useful libraries, like database access, which are essentials for any business development. Thanks, Andrey Dadakov adadakov@awaresw.com Aware Software, Inc http://www.awaresw.com -----Original Message----- From: haskell-admin@haskell.org [mailto:haskell-admin@haskell.org] On Behalf Of Hal Daume III Sent: Friday, August 29, 2003 5:39 PM To: Haskell Mailing List Subject: Haskell for non-Haskell's sake Hi fellow Haskellers, I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake. If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too. Purposes which I consider "Haskell for Haskell's sake" include: - writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :) Thanks, - Hal -- Hal Daume III | hdaume@isi.edu "Arrest this man, he talks in maths." | www.isi.edu/~hdaume _______________________________________________ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
On 2003-08-29 at 17:39PDT Hal Daume III wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear.
In addition to the things you exclude, I've written (and sold!) a custom web-server log-analysis programme. I use Haskell for managing some personal data, and also use it a great deal for what prof. Wilkes calls nonce programming. Jón -- Jón Fairbairn Jon.Fairbairn@cl.cam.ac.uk
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
I have used Haskell for: - Knit (http://www.cs.utah.edu/flux/knit/) A component language for C based on the Unit model used in MzScheme. (Roughly comparable to using ML functors for every module.) This was used to break the OSKit (a huge collection of OS kernel components extracted from Linux, Mach and *BSD) into finer grained components. A new version of Knit which allows import and export of types is almost ready. (The first version could only import/export functions, variables and opaque types.) [Knit is a language needing all the bits that Haskell is good for writing (lexing, parsing, static typechecking, desugaring, optimizing, generating 'code', etc.) Uses an external program written in C to rename symbols in ELF-format object files.] - CMI (due for release soon) A tunable cross-module inliner for C. Can be used to speed up your code. Can be used to reduce the stack consumption of embedded systems (http://www.cs.utah.edu/~regehr/stacktool/) Can be used to let you adopt a better programming style in C by letting you put aside some performance concerns. [Works by parsing C source code, merging multiple files into one, topologically sorting function definitions, pretty-printing, etc. All the things Haskell is good for!] - ADL An architecture description language which focusses on describing the hierarchies of schedulers and composition of software components found in embedded systems. (Background: http://www.cs.utah.edu/flux/papers/cee-flux-tn-02-02/) [Again, ADL is a language so it has all the usual bits Haskell is good for: lexer, parser, dynamic typechecking, dimension types, interpreter, etc. Also uses Haskell's ability to call external tools so it can use external analysis tools written in Haskell (like TSL, below), or written in C, perl, whatever.] - Task/Scheduler Logic (TSL) A language/ logic/ tool for reasoning about hierarchies of schedulers found in embedded systems. [Uses Haskell's strengths for implementing languages and for symbol manipulation (contains a small forward-chaining inference engine).] - FVision (Visual tracking) Given a bunch of simple image tracking primitives (written in C++ and assembly, part of the larger XVision system), build complex feedback loops, hierarchies, etc. to create more robust, flexible, sophisticated tracking systems. http://www.reid-consulting-uk.ltd.uk/alastair/publications/padl01/index.html Uses Haskell's ability to 'embed' domiain specific languages inside it. [One could argue that this project was just 'Haskell for Haskell's sake' but it's worth pointing out that it lead to a complete redesign of XVision along the lines I had developed in the Haskell version.] -- Alastair Reid
On Sat, 30 Aug 2003, Alastair Reid wrote:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
[snip]
- FVision (Visual tracking)
Given a bunch of simple image tracking primitives (written in C++ and assembly, part of the larger XVision system), build complex feedback loops, hierarchies, etc. to create more robust, flexible, sophisticated tracking systems. http://www.reid-consulting-uk.ltd.uk/alastair/publications/padl01/index.html
Uses Haskell's ability to 'embed' domiain specific languages inside it.
[One could argue that this project was just 'Haskell for Haskell's sake' but it's worth pointing out that it lead to a complete redesign of XVision along the lines I had developed in the Haskell version.]
I do research in computer vision/image processing and I've also used Haskell quite a lot for doing prototyping of algorithms. I'm doing sort of the opposite thing to Alastair: he's taking established low-level image analysis techniques (written in C/C++) and combining them in more effective ways using Haskell as a language for doing higher level processing. (Apologies if this is an incorrect understanding.) I work on more effective low-level image processing algorithms with a higher-level stuff that's simple and stable enough that coding it in C++ doesn't cause a problem. I do extensive prototyping using simple Haskell implementations of ideas; once I'm reasonably happy that the idea has a chance of working I then convert it to C++. I have to convert to C++ for `real work' because (a) Haskell is too slow for most of the low-level stuff, particularly `semi real-time' image processing and (b) no-one else here knows Haskell so if I want to be able to share code on common projects I need either C or C++. I want eventually to be able to plug in Haskell code prototypes into the overall C++ structure to be able to do more testing before moving to C++, but that awaits me having enough free time to study the Haskell FFI, etc... I'm very impressed with the FVision stuff and I've contrasted what I do with the it just to show Haskell is being used for BOTH high and low-level areas. I also use Haskell for some `scripting-stuff level tasks' like autogenerating makefiles and processing log files. I write both Perl and Python code where they seems best, so I can reasonably say that in those cases where I use Haskell it's because I think it's easier for me than those languages. ___cheers,_dave_________________________________________________________ www.cs.bris.ac.uk/~tweed/ | `It's no good going home to practise email:tweed@cs.bris.ac.uk | a Special Outdoor Song which Has To Be work tel:(0117) 954-5250 | Sung In The Snow' -- Winnie the Pooh
Hi Hal (et al.) I am using it to write a compiler and interpretor for a quantum programming language, based on the semantics of the paper by Peter Selinger. (See http://quasar.mathstat.uottawa.ca/~selinger/papers.html#qpl for details on the semantics) On 29 Aug, Hal Daume III wrote:
Hi fellow Haskellers,
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Purposes which I consider "Haskell for Haskell's sake" include:
- writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :)
-- Brett G. Giles Grad Student, University of Calgary Formal Methods, Category Theory, Semantics of Programming http://www.cpsc.ucalgary.ca/~gilesb mailto:gilesb@cpsc.ucalgary.ca
On Fri, 29 Aug 2003, Hal Daume III wrote:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Purposes which I consider "Haskell for Haskell's sake" include:
- writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :)
I'm writing a compiler in it for my dissertation project at uni - the source language is one of my own design. -- flippa@flippac.org
I use haskell when I have to write a program myself and quickly. So I was very happy when I saw wxwindows bindings, because I wrote a frontend for mame with it, and it took three days to get something satisfying. We need some "ordinary people" use for haskell sometimes ;) V.
Hello, 1 I wrote Haskell programs to compute matrix elements of operators (in physics). 2 I use Haskell for generating figures (Functional Metapost). 3 For generating HTML summaries out of some data. 4 For common text processing as an advanced sed. Actually, I do not use Haskell for Haskell at all... Sorry for that. :( Feri.
At 17:39 29/08/03 -0700, Hal Daume III wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Rather than repeat myself, I refer you to my "individual Haskellers" entry in: http://haskell.cs.yale.edu/communities/05-2003/html/report.html#sect6.5 (Hmmm... I notice no direct URI reference is available... just scan down a section from the above link.) In short, I'm exploring use of Haskell as "a scripting language for the semantic web". #g ------------ Graham Klyne GK@NineByNine.org
Hi Hal, On Saturday 30 August 2003 01:39, Hal Daume III wrote:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Together with Ralf Laemmel I have applied Haskell for processing not only Haskell itself, but also Cobol and Java. See: http://homepages.cwi.nl/~ralf/padl03/ And a quote from the abstract: In a selection of case studies, we demonstrate that typed functional programming in Haskell, augmented with Strafunski's support for generic traversal and external components, is very appropriate for the development of practical language processors. In particular, we discuss using Haskell for Cobol reverse engineering, Java code metrics, and Haskell re-engineering. Then again, we are not using only Haskell, since we resort to external components e.g. for parsing. Regards, Joost -- Dr. ir. Joost Visser | Departamento de Informática phone +351-253-604461 | Universidade do Minho fax +351-253-604471 | mailto:Joost.Visser@di.uminho.pt mobile +351-91-6253618 | http://alfa.di.uminho.pt/~joost.visser
Since the opening of this thread by Hal Daume 11 (binary), we see a constant flow of interesting contributions/confessions. Plenty of applications, it seems that Haskell is really used in a wider context than we might think. It is a pleasure to read all this. I have just one question thus. Why the application-oriented papers devoted to Haskell at ICFP, including the Haskell workshop are rather rare? People are reluctant to contribute, or the reviewers are not so fascinated? (Well, it happened once to me, but it had *nothing to do with Haskell*, and nothing to do with ICFP, just some other workshop, elsewhere. Simply a BTW remark. One reviewer wrote: "this is just an application work, not a scientific paper". Presumably this reviewer has his particular visions what a science is, but I don't believe that such people dominate in the milieu of FPL. I believe that it would be interesting to organize some workshops on "practical" applications of functional programming...) Jerzy Karczmarczuk Caen, France
At 16:15 01/09/03 +0200, Jerzy Karczmarczuk wrote:
Since the opening of this thread by Hal Daume 11 (binary), we see a constant flow of interesting contributions/confessions. Plenty of applications, it seems that Haskell is really used in a wider context than we might think. It is a pleasure to read all this.
I have just one question thus. Why the application-oriented papers devoted to Haskell at ICFP, including the Haskell workshop are rather rare?
There are many conferences more closely related to my application interest that would be more useful for me to attend (if I had the funding...) #g ------------ Graham Klyne GK@NineByNine.org
In article <3F53547C.6080203@info.unicaen.fr>, Jerzy Karczmarczuk <karczma@info.unicaen.fr> wrote:
I have just one question thus. Why the application-oriented papers devoted to Haskell at ICFP, including the Haskell workshop are rather rare?
Perhaps people who are busy writing applications don't really bother with papers? Most C++ and Java programmers don't write papers... -- Ashley Yakeley, Seattle WA
On 30-Aug-2003 Hal Daume III wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
I have been working on a software-engineering tool that is intended to derive SQL-DDL-Statements, XML-Schema-Documents, and Java classes from a common schema definition which is written in some UML-like language (with a proper semantics of course). The tool will furthermore generate java-code for the conversion of complex objects (that is, instances of certain subgraphs of the eschema) along the paths SQL<->Java, SQL<->XML, Java<->XML. The project will however be cancelled before it reaches a state where the software could actually deployed. Everything except of the frontend that is a graphical editor for schemata, has been written in Haskell. It is surely not "Haskell for Haskell" but "Haskell for the Rest of the World". Elke Kasimir. Software Development EsPresto AG ----------------------------------------------------------------- kasimir@espresto.com Breite Str. 30-31 Tel/Fax: +49-30-90 226-750/-760 10178 Berlin/Germany
Hi Hal and others, We would like to hear your thoughts on the viability of a conference or workshop dedicated to applications of Haskell for non-Haskell purposes. On Saturday 30 August 2003 01:39, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
This bias seems to exist not only in the Communities & Activities reports, but also in the Haskell mailing lists and in the Haskell-related events, such as the Haskell Workshop. One could easily deduce that Haskell is still very much an academic language, of interest to language _designers_ more than to language _users_. However, the reactions to your inquiry about use of Haskell for non-Haskell purposes suggests that a significant group of language _users_ does actually exist, though their voice is not heard too often. We think (hope) there could be room for a "Haskell One" event, quite a bit more modest than JavaOne, but similarly intended to bring together Haskell Developers. Submissions would be explicitly judged by how practical and how "proven" the described technology is. Experience reports would be more than welcome. There could be interactive tool-demo's, hands-on lab sessions, a programming contest, etc. If the focus is on applications only (excluding theory, language-design, and anything that is Haskell-for-Haskell's sake), this HaskellOne event could maybe be a satelite to PLI. We are interested to know your thoughts on such an event, and whether the outcome of your poll suggests it could be viable. Regards, Joost Visser & João Saraiva PS: If sufficient interest exists, we are willing to contribute to the origanization. -- Dr. ir. Joost Visser | Departamento de Informática phone +351-253-604461 | Universidade do Minho fax +351-253-604471 | mailto:Joost.Visser@di.uminho.pt mobile +351-91-6253618 | http://alfa.di.uminho.pt/~joost.visser
On Mon, 1 Sep 2003, Joost Visser wrote:
Hi Hal and others,
We would like to hear your thoughts on the viability of a conference or workshop dedicated to applications of Haskell for non-Haskell purposes.
On Saturday 30 August 2003 01:39, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
This bias seems to exist not only in the Communities & Activities reports, but also in the Haskell mailing lists and in the Haskell-related events, such as the Haskell Workshop. One could easily deduce that Haskell is still very much an academic language, of interest to language _designers_ more than to language _users_. However, the reactions to your inquiry about use of Haskell for non-Haskell purposes suggests that a significant group of language _users_ does actually exist, though their voice is not heard too often.
Personal viewpoint: I think I see a reasonable number of people asking questions about how to acheive what they need to in Haskell, which whilst it isn't often explicitly stated, often appears to be because they've got a task that they aren't `doing in Haskell for Haskell's sake'. When making your contribution is spending 10 minutes writing an e-mail (such as this one) there's no problem making your voice heard, and it's nice think you're being an active member of a very nice and helpful community. When it's writing a paper for a conference, which requires weeks of concerted effort, requires that you (& the reviewers :-) ) beleive you've done something worth telling other people about, finding funding to attend the conference (which may be funding which could be used to attend a conference in an area where you are a specialist), and when your peers in your `proper subject area' won't be interested in the result of all this work, it seems natural (though not of course desirable) that most `pure users' of Haskell don't have enough desire to do the work to make their voice heard that way. To put it in context, I wouldn't expect virtually anyone on the list who work in some area (e.g., Hal appears to work in Natural Language Processing) to have attended a conference for the language they did a particular piece of software in, when that language was Java, C++, Perl, Python (and I know there are conferences for those languages). This isn't to put anyone off the idea of a Haskell applications conference as such, I just think that before beginning there should be a convincing rebuttal of the points above. There may well be one; it's VERY possible I'm wrong/atypical. ___cheers,_dave_________________________________________________________ www.cs.bris.ac.uk/~tweed/ | `It's no good going home to practise email:tweed@cs.bris.ac.uk | a Special Outdoor Song which Has To Be work tel:(0117) 954-5250 | Sung In The Snow' -- Winnie the Pooh
[this seemed to be flowing along nicely, but now that the thread has moved from information to organisation and meta-discussion, I'd like to add a few comments, and an invitation]
On Saturday 30 August 2003 01:39, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
This bias seems to exist not only in the Communities & Activities reports, but
The bias is entirely in what readers of this mailing list are contributing to these reports. The editor has certainly been more than willing to include "other" applications of Haskell. I know there are lots of other interesting things going on out there, but have so far not been able to reach these people: - some of them read/attend none of the Haskell list, c.l.f, or Haskell Workshop, so they simply never hear about these report *unless _you_ forward the invitations to participate* - some read the reports and the calls for contributions, but don't think of their work as particularly interesting to a wider community, or as not substantial enough So, I do hope that those who've answered Hal's call (or have been thinking about answering) *will* contribute to the next edition of the Haskell Communities & Activities Report! I'll be expecting your email in my mailbox in the last 2 weeks of October :-)
also in the Haskell mailing lists and in the Haskell-related events, such as the Haskell Workshop.
I'm not sure about this list, but as for the Haskell workshop, this year we had (http://www.cs.uu.nl/~johanj/HaskellWorkshop/cfp03.html): - 4 presentations on applications of Haskell, to gaming, quantum mechanics, quantum computing, web site development - 8 presentations on programming techniques, tools, debugging, and libraries - 3 presentations on language design issues (strict language, records, lack of principal types) How does this relate to the bias you see?
However, the reactions to your inquiry about use of Haskell for non-Haskell purposes suggests that a significant group of language _users_ does actually exist, though their voice is not heard too often.
Indeed. So, please let yourself be heard!-) Or if you know of someone else who does interesting Haskell stuff, ask them to talk about their work. It is the collection of all these small fragments that makes the HC&A Reports useful!
When making your contribution is spending 10 minutes writing an e-mail (such as this one) there's no problem making your voice heard, and it's nice think you're being an active member of a very nice and helpful community.
Just the way the HC&A Reports are intended to work (if the 10 minute email lacks any information, the editor will get back to you). As for Haskell Applications events: IMHO, Haskell has grown beyond that (is an application more interesting for the sake of being implemented in Haskell?). The point of applications is not the language, and some of us have already presented Haskell applications at domain-specific events, which is the way to go. Where there is interaction between the application and the language (tools missing, features inadequate) or where applications are facilitates by language and tool developments, the Haskell workshop and IFL are good places to present and discuss that work. Potential benefits of a Haskell applications event would be for - advertizing: having several application presentations in a single venue - contacts: getting to meet other Haskell "applicators" The main problem: "real" Haskell developers are reluctant to talk about their work (and sometimes business interests are in the way). The comparison to JaffaOne is misleading, methinks: are there enough professional Haskell developers who can afford to/have to attend such an event with the goal of keeping up-to-date with their main tool? And would there be enough presenters to make such a visit worthwhile for the professional participants? Perhaps an add-on to the Advanced Functional Programming Summerschools might work? I guess I'd prefer a Haskell quarterly "magazine", with editing, but emphasizing use over academic criteria when evaluating submissions. Cheers, Claus -- http://www.haskell.org/communities/
I use Haskell to design and verify circuits that are used at my company and by our customers. A Haskell-based methodology for producing circuits has proved to be successful in some situations when a conventional flow based on Java or hardware description languages (VHDL and Verilog) was not able to deliver a solution of suitably high quality (speed or area). Some information at http://www.xilinx.com/labs/lava A public release will be made available before the end of the year. Cheers, Satnam Hal Daume III wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Purposes which I consider "Haskell for Haskell's sake" include:
- writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :)
Thanks,
- Hal
-- Hal Daume III | hdaume@isi.edu "Arrest this man, he talks in maths." | www.isi.edu/~hdaume
_______________________________________________ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
On Fri, Aug 29, 2003 at 05:39:09PM -0700, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake. If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too. Purposes which I consider "Haskell for Haskell's sake" include: - writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :)
I use Haskell interpreters as a "more expressive calculator", also do basic scripting such as vast numbers regularly improvised log processing scripts (usually for logs of debugging output and other similar dumps), as well as some somewhat less frequently improvised network "scripts" or script-like things (both clients and servers). i.e. where others use perl and expect. Haskell's faster for me to write. -- wli
On Fri, Aug 29, 2003 at 05:39:09PM -0700, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell
I work at Galois Connections <http://galois.com> and much of the software we write (mostly government contracting) is written in Haskell. I've written an ASN.1 parser prototype, a remote-procedure-call framework, and a secure web-server back-end. -- Brett Letner Galois Connections, Inc. http://www.galois.com mailto:bletner@galois.com phone:(503)626-6616 ext.110
G'day all. On Fri, Aug 29, 2003 at 05:39:09PM -0700, Hal Daume III wrote:
I'm attempting to get a sense of the topology of the Haskell community.
I used Haskell to write a compiler for the RenderMan shading language for a former employer. Unfortunately, the compiler never shipped. I still own the IP, though, so it may yet end up open source. Cheers, Andrew Bromage
Hello, I am a student from Germany and I have used Haskell for several purposes as well: - to implement and compare algorithms quickly, e.g. Travelling Salesman, Sorting, etc. - to calculate state spaces and blocking probabilities in networks - to solve some of our cryptography and computability excersises :-) - to solve several programs of the online judge, although Haskell is currently not supported Further on I have written an interpreter for RATH (exploring finite relation algebras using tools written in Haskell) and in this time I use Haskell in my diplome thesis (but unfortunately for the sake of Haskell :-p). Some time ago, we had some problems with our news server and installed a new inn. This new inn was totally incompatible with the old one (or we were to clumsy to figure it out) and so the old articles seem to be lost. So it was our task to write a script, which should transform the backuped, old articles on the server into expect-scripts. I did this in Haskell and it worked satisfactorily :-) I call it the "News-Reposter" *g*. Haskell is my programming language of choice. It is possible, to solve problems quickly and the algorithms look so beautiful. I miss convenient and standardized libraries for gui-programming! I think, this is a serious problem. Bye, Steffen Mazanek P.S. I do my best to motivate other people to give Haskell a try, e.g. during a lecture about document-description-languages I had provided a funny example, how useful HaXml is :-) The program reads a xml-file with descriptions (age, iq, bust size *g*, all values are only estimated) of some famous women (Britney, Madonna, etc.) and puts out a nicely-formatted html-file with a final evaluation (value=iq*bustsize-100*|age-22| *g*). In case, that a woman is reading this list: One could easily invent a formula as well, which evaluates men.
The Timber group at OHSU/OGI are using Haskell to write their Timber compiler and Timber VM executable specification. Andy Gill Hal Daume III wrote:
Hi fellow Haskellers,
I'm attempting to get a sense of the topology of the Haskell community. Based on the Haskell Communities & Activities reports, it seems that the large majority of people use Haskell for Haskell's sake.
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear. I don't need a long report, anything from a simple "I do" to a paragraph would be fine, and if you want to remain anonymous that's fine, too.
Purposes which I consider "Haskell for Haskell's sake" include:
- writing Haskell compilers/interpreters - developing libraries for Haskell - writing Haskell debuggers, tracers, profilers or other tools - more or less anything with matches /.*Haskell.*/, other than /in Haskell$/ :)
Thanks,
- Hal
-- Hal Daume III | hdaume@isi.edu "Arrest this man, he talks in maths." | www.isi.edu/~hdaume
_______________________________________________ Haskell mailing list Haskell@haskell.org http://www.haskell.org/mailman/listinfo/haskell
* Hal Daume III <hdaume@ISI.EDU> [2003-08-29 17:39 -0700]:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear.
I have written checkrdf[1], a tool for downloading and processing RSS files from various newstickers. checkrdf uses HaXml for processing RSS XML files. A shell script is used as wrapper and the XML files are fetched by fetch(1) or wget(1). I have developed Editor Combinators[2][3] together with Wolfram Kahl and Jan Scheffczyk for my diploma thesis. Currently, I am using Haskell for my PhD work where I use Haskell as "Runtime Environment" for code generated by HOPS[4]. I have done statistical evaluation of medical datasets (sensitivity, specitivity, predictive values and the like) in Haskell. For small tasks (processing textfiles and the like) I also use Haskell or zsh[5] scripts. Regards, Olli 1. http://checkrdf.sourceforge.net/ 2. http://www2-data.informatik.unibw-muenchen.de/EdComb/ 3. http://www2-data.informatik.unibw-muenchen.de/People/obraun/diploma_thesis.p... 4. http://www2-data.informatik.unibw-muenchen.de/kahl/HOPS/ 5. http://www.zsh.org/ -- Oliver Braun -- obraun @ { unsane.org | FreeBSD.org | haskell.org }
Hal Daume III wrote:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear.
Haskell is the implementation language behind PXSL, the Parsimonious XML Shorthand Language: PXSL ("pixel") is a convenient shorthand for writing markup-heavy XML documents. It provides XML authors and programmers with a simple, concise syntax that they can use to create XML documents. For more advanced users, PXSL offers customizable shortcuts and sophisticated refactoring tools like functional macros that can markedly reduce the size and complexity of markup-dense XML documents. http://community.moertel.com/ss/space/pxsl PXSL also borrows Haskell's layout rule, which you can see in this brief example comparing some MathML in XML and PXSL: MathML in XML MathML in PXSL <declare type="fn"> declare -type=fn <ci> f </ci> ci << f >> <lambda> lambda <bvar><ci> x </ci></bvar> bvar <apply> ci << x >> <plus/> apply <apply> plus <power/> apply <ci> x </ci> power <cn> 2 </cn> ci << x >> </apply> cn << 2 >> <ci> x </ci> ci << x >> <cn> 3 </cn> cn << 3 >> </apply> </lambda> </declare> I used Haskell to write PXSL because (a) I like Haskell, (b) it made writing the parser easy (thanks, Parsec!), (c) it made the macro system easy, and (d) did I mention that I like Haskell? Cheers, Tom
On Fri, Aug 29, 2003 at 05:39:09PM -0700, Hal Daume III wrote:
If you use Haskell for a purpose *other than* one of those listed below, I'd love to hear.
In my job (website traffic measurement) the official programming language is C++ (also PHP and Java, but I don't touch these) and AFAIK I am the only one that knows and likes Haskell here. Because of this I can use Haskell only for programs which are not critical for the company. I would rather use Haskell for a couple of applications I have done in C++, but there are applications with performance requirements beyond GHC's abilities, I'm afraid (for example imagine a database with more than 10,000,000,000 records, about 400,000,000 updates/inserts every day, many statistics which have to be recomputed after updates, and this has to work on an off-the-shelf PC :). So far I have used Haskell for such not-for-haskell's-sake tasks: - various networking applications For example, I created a pure (concurrent) haskell dns resolver library which I find very efficient for resolving tons of IP addresses. I parse DNS messages declaratively using Parsec :). It's a bit incomplete now, but someday I will fix this. As I have written it in my free time, I can release it as open source. - creating prototype web interfaces based on WASH for presenting statistics computed by other programs - text/data processing, like converting files from one format to another, or performing transformations of simple databases in text files - general scripting (things I would do in Perl two years ago) - implementing algorithms which would be tiresome to do in C++ and/or don't have to be (crashing :) so fast - making prototypes of algoritms, ideas, etc. - various calculations in GHCi - etc. It's about 20 KLOC of code in total (without the throw-away ones). Best regards, Tom -- .signature: Too many levels of symbolic links
Hal (and other interested parties): I used Haskell to implement a model checker for a group of logics of time and knowledge. In practice these are a bunch of extensions to the classic CTL algorithms implemented in SMV [1]. The program itself (in terms of LOC) looks mostly like a compiler, and so the standard arguments apply. However, Haskell is absolutely perfect for implementing the SMV algorithms (and our variants) as they are cast in terms of fixpoints of boolean functionals. Moreover performance isn't an issue as a C library does the heavy lifting. (The program uses BDDs at the moment.) WRT Haskell, the only worry I had was that the program might leak space in difficult-to-resolve ways, and the usual problem of BDD variable ordering being difficult to control. cheers Peter. [1] SMV itself is actually a LISP program written in C. (I think this is self-evident from even a cursory glance, but the real giveaway is their AST: who else, apart from a seasoned LISP hacker, would use CONS cells for everything?)
participants (27)
-
Alastair Reid -
Alexandre Weffort Thenorio -
Andrew J Bromage -
Andrey Dadakov -
Andy Gill -
Ashley Yakeley -
Brett Letner -
C.Reinke -
D. Tweed -
Elke Kasimir -
Ferenc Wagner -
gilesb＠cpsc.ucalgary.ca -
Graham Klyne -
Hal Daume III -
Jerzy Karczmarczuk -
Jon Fairbairn -
Joost Visser -
Krasimir Angelov -
Nick Name -
Oliver Braun -
Peter Gammie -
Philippa Cowderoy -
Satnam Singh -
Steffen Mazanek -
Tom Moertel -
Tomasz Zielonka -
William Lee Irwin III