February 2003 - Haskell

Prototyping language extention
by Dmitry Malenko 10 Feb '03

10 Feb '03

Hi, All! As part of my research I'm going to create prototype system implementing some extentions concerning at most module system to the language and to toy with it a little. So I wonder what is the best way to do that? Should I cope with compiler sources, or write an interpeter from scratch, or maybe something else? Any thoughts or pointers would be apprciated. -- Best regards, Dmitry Malenko. +----------------------------------------------------+ Registered Linux user #258004 ALT Linux Team http://www.altlinux.ru +----------------------------------------------------+ Staff meeting in the conference room in %d minutes.

2 1

RE: filter and Prelude index in H98 report?
by Simon Peyton-Jones 10 Feb '03

10 Feb '03

No there isn't. Its in the revised version at http://research.microsoft.com/~simonpj/haskell98-revised/haskell98-repor t-html/index.html I'll ask John to upload it. Simon | -----Original Message----- | From: Antony Courtney [mailto:antony@apocalypse.org] | Sent: 07 February 2003 22:04 | To: haskell(a)haskell.org | Subject: filter and Prelude index in H98 report? | | Hi all, | | Is there any deep reason why "filter" does not appear in the "Index of | the Haskell 98 Prelude", found at: | | http://www.haskell.org/onlinereport/prelude-index.html | | Thanks, | | -antony | | -- | Antony Courtney | Grad. Student, Dept. of Computer Science, Yale University | antony(a)apocalypse.org http://www.apocalypse.org/pub/u/antony | | _______________________________________________ | Haskell mailing list | Haskell(a)haskell.org | http://www.haskell.org/mailman/listinfo/haskell

1 0

[OT] Teaching Haskell in High School
by Hal Daume III 09 Feb '03

09 Feb '03

Hi all, Before getting in to this, let me preface my question(s) with a note that I have checked through the Haskell in Education web page and have found various links off there of interest (and I've googled, etc. In short: I've done my homework). That said, I've been in rather close correspondence with my math/computer science teacher from high school. When I first took CS there, they taught Pascal (a year early they had been teaching Scheme). They switched over to VB (alas) recently and have been teaching that for a few years now. The teacher really wants to get away from VB, but is having a somewhat difficult time deciding what to go to. The two most promising options are Haskell and Java. Aside from hype, etc., the primary advantage to Java is that the Advanced Placement (AP) tests are in Java. For those of you unfamiliar with these, high school students can take AP tests and then (typically) skip out of first semester college courses. They're essentially proficiency exams. The way the computer science curriculum is set up at my old school is essentially as either (a) an elective or (b) a replacement for senior year math. The students in the course are usually about 2/3 juniors (16 year olds) taking it as an elective and 1/3 seniors who want to get our of senior year math :). Either way, they've both taken differential calculus, algebra, etc. Note, however, that high school math in the states is very rudimentary when it comes to things like "induction" and "proofs" and things of this sort. Due to the fact that CS is essentially an alternative math course, I think it would be interesting to teach Haskell. It would enable the instruction of things the students wouldn't have come across in their ordinary math studies, etc. However, I'm also well aware that Haskell is very difficult to learn (and, I'd imagine, to teach). Given that this would in large part be a first language for them and that they won't have a college-level math background, do you think it would be too much to attempt to teach Haskell at this level, and stick with Java? I'm really interested in any comments/experience/etc. that people have that might assist the teacher (and, to some extent, me) make this decision. Thanks in advance! - Hal

7 16

Re: time since the epoch
by Peter Thiemann 09 Feb '03

09 Feb '03

>>>>> "SM" == Simon Marlow <simonmar(a)microsoft.com> writes: >> Ha! After playing with this, I discovered that only the seconds were >> set and all other fields remained untouched. At least in ghc's >> implementation. Interestingly, TimeDiff derives Eq and Ord, but I'd >> better not ask for their implementation... SM> TimeDiff has a perfectly well-defined meaning for Eq and Ord: one SM> TimeDiff is equal to another iff all the fields are respectively equal. Yes, sure, Eq and Ord are defined by deriving (Eq, Ord, ...) but this does not make much sense if the representation for a particular duration is not unique. All I'm asking for is a unique representation for a duration and I do not see how you can do this using not uniquely defined terms like "month" and "year". As an aside, you can easily define convenience functions that map your (application's) idea of months and years to seconds, but it would be much harder the other way round. >> There is also a more fundamental problem with the TimeDiff data >> type. While seconds, minutes, hours, and days are clearly specified >> amounts of time, the duration of a month or a year may vary depending >> on the reference point where the time difference is applied. SM> This isn't a problem with the spec, I think. A TimeDiff of "1 month" is SM> precisely a difference, which, when added to a given ClockTime, produces SM> a ClockTime which is one month later (with respect to some timezone, SM> presumably UTC since it isn't specified otherwise). That is, January SM> 12th 12:00 UTC becomes February 12th 12:00 UTC, and so on. Adding one SM> month to certain ClockTimes is meaningless: eg. adding one month to SM> January 31st doesn't work. Unfortunately, with UTC, this goes on to make adding *anything* except seconds to certain ClockTimes (namely the instants of leap seconds, eg, what happens if you add 1 minute to 23:59:60 ? This is a perfectly legal UTC time on a day with a leap second) meaningless. My take would be to look at an established standard for expressing time durations (eg, ISO8601) and just adhere to that (but not the POSIX standard, please!). Personally, I think that it is unacceptable to make something like addToClockTime a partial function. SM> At least, that's how I believe it is supposed to work. GHC's SM> implementation *is* completely wrong (which is perhaps where some of the SM> confusion comes from), but the TimeExts library can be used instead of SM> TimeDiffs. Well, so you agree that the behavior of the Time module is unspecified! Of course, I'm not blaming GHC's implementation which cannot be right under these circumstances. >> My conclusion is that time differences really should be measured in >> seconds and picoseconds. >> >> type TimeDiff = (Integer, Integer) SM> We already have this kind of TimeDiff: just don't use anything except SM> the seconds and picoseconds fields. Indeed, I found by experimentation that diffClockTimes only fills in these two fields so that my attempt to compare the result with TimeDiff{ 0... tdDay = 14 ... 0 } was doomed. Unfortunately, the tdSec field has type Int, so its range is too small to be useful. SM> I agree with John Meacham in that it ought to be possible to get a raw SM> time in terms of seconds/picoseconds since the epoch, and indeed GHC SM> lets you do that: importing System.Time lets you get at the SM> representation of ClockTime: SM> data ClockTime SM> = TOD Integer -- Seconds since 00:00:00 on 1 Jan 1970 SM> Integer -- Picoseconds with the specified second SM> This is a perfectly reasonable definition of absolute time, we don't SM> need to talk about TAI at all, except to define what the epoch means. Agreed! So what is epoch? Is it "Seconds since 00:00:00 on 1 Jan 1970 [TAI]" ?

2 1

RE: time since the epoch
by Simon Peyton-Jones 09 Feb '03

09 Feb '03

| the haskell 98 time library is horribly broken, if you are using ghc, | you can deconstruct the time constructor which has an Integer containing | the number of seconds since epoch... otherwise you can use | ... | I dont supose this could be considered a typo in the haskell 98 report? | it is an embarasing thing for a language to not be able to do... Alas, no. CUP is, even as I type, printing the Haskell 98 Report. We'll have to think about how to accumulate further bugs/suggestions for H98. Meanwhile, I suspect there's an opportunity for someone (or a small group) to suggest a new Time library that really does the business, and provide an implementation. If it's sufficiently persuasive, all the implementations will adopt it and it can become a de-facto standard. The implementation is important because Time is a weird enough thing that only an expert can implement the spec! I think the current Time library suffered because no one who Really Cared about time was involved in its design or implementation. Simon

7 8

Re: FiniteMap performance WITHOUT any copying?
by George Russell 08 Feb '03

08 Feb '03

Magnus wrote (snipped) > Now, Haskell has a garbage collector, so Haskell must know how many > pointers there are to all objects in the heap (or?). Then, if a finite map > for example only has one single pointer to it, then it ought to be all > right to modify the finite map (or whatever datastructure we are > considering), I mean really modify the map without making any copies, just > like in imperative languages. Perhaps there might be pointers to nodes > inside the tree and I guess that could complicate the matter somewhat. But > for Haskell arrays it ought to be possible to really modify the array if > it is used by only one pointer ? > > Are such optimizations possible, and if they are, are they already > implemented in for example GHC ? Or am I wrong somewhere ? I think it unlikely that such an optimisation is implemented for GHC, which is as a rule extremely unwilling to modify "functional" data. The trouble with this kind of thing is that although it seems neat, it is hard to implement, and also opens up cans of worms all over the place; for example by making generational garbage collection harder to implement. In fact I don't think GHC does know how many pointers there are to all objects in the heap. That would only be useful if it did GC by reference-counting, which it doesn't. Reference counting is in fact rather expensive; as well as requiring an extra field on every item to hold the number of references, you need to add lots of operations to increment these fields every time you create or delete a new reference. Also of course reference counting will not be able to free circular structures, at least not in general with data being modified.

2 1

Re: FiniteMap performance WITHOUT any copying?
by Zdenek Dvorak 08 Feb '03

08 Feb '03

Hello, >Now, Haskell has a garbage collector, so Haskell must know how many >pointers there are to all objects in the heap (or?). Then, if a finite map >for example only has one single pointer to it, then it ought to be all >right to modify the finite map (or whatever datastructure we are >considering), I mean really modify the map without making any copies, just >like in imperative languages. Perhaps there might be pointers to nodes >inside the tree and I guess that could complicate the matter somewhat. But >for Haskell arrays it ought to be possible to really modify the array if >it is used by only one pointer ? > >Are such optimizations possible, and if they are, are they already >implemented in for example GHC ? Or am I wrong somewhere ? You are more or less right; except that -- the Haskell implementations I have seen do not keep number of pointers to objects (gc in principle works by traversing the graph made by pointers, marking reachable nodes and finally removing the unreachable ones; just keeping the number of references is not sufficient due to cyclic structures) -- due to lazy evaluation, the situation where there would be just one pointer to array is rare Zdenek _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

1 0

filter and Prelude index in H98 report?
by Antony Courtney 07 Feb '03

07 Feb '03

Hi all, Is there any deep reason why "filter" does not appear in the "Index of the Haskell 98 Prelude", found at: http://www.haskell.org/onlinereport/prelude-index.html Thanks, -antony -- Antony Courtney Grad. Student, Dept. of Computer Science, Yale University antony(a)apocalypse.org http://www.apocalypse.org/pub/u/antony

1 0

SBLP 2003 - extended deadline
by Lucilia Camarao de Figueiredo 07 Feb '03

07 Feb '03

CALL FOR PAPERS SBLP 2003 - 7th Brazilian Symposium on Programming Languages Ouro Preto, MG, Brazil - May 28-30, 2003 http://www.inf.pucminas.br/sblp2003 ******** Extended deadline: February 24th ****** ***** PROCEEDINGS WILL BE PUBLISHED AS A VOLUME OF THE ***** "Journal of Universal Computer Science" (www.jucs.org) Scope The Brazilian Symposium on Programming Languages (SBLP), sponsored by the Brazilian Computer Society (SBC), is a forum where researchers, educators, and practitioners can present and discuss the fundamental principles and innovations in the design, analysis and implementation of programming languages and systems. We invite authors to contribute to SBLP 2003 with the following kinds of submissions: - Technical research papers introducing original research results on topics such as (but not limited to): programming language design and implementation, formal semantics of programming languages, domain-specific programming languages, programming languages for mobile, WWW, and network computing, script languages, new programming models, theoretical foundations of programming languages and teaching programming languages. - Tutorial proposals on subjects related to programming languages, APIs, tools, environments or theories, at graduate level. Important Dates - Deadline for submissions (extended): *** February 24th, 2003 *** - Notification of acceptance: March 28th, 2003 - Camera-ready version: April 17th, 2003 Invited Speakers Nick Benton Programming Principles and Tools group Microsoft Research, Cambridge http://research.microsoft.com/~nick Simon Thompson Computing Laboratory University of Kent http://www.cs.ukc.ac.uk/people/staff/sjt Submission Guidelines Contributions can be written in Portuguese, English or Spanish. Papers should have at most 14 pages. All papers will be published in the conference proceedings. Tutorial submissions may be in the form of a proposal. The final version of a selected tutorial should contain at most 30 pages and will be published in separate proceedings. Please check the symposium web site for more information about submissions guidelines and procedures. Program Committee Luis Barbosa, Universidade do Minho Roberto Bigonha, UFMG Mariza Bigonha, UFMG Paulo Blauth Menezes, UFRGS Paulo Borba, UFPE Christiano Braga, UFF Carlos Camarao, UFMG Renato Cerqueira, PUC-Rio Jose Luiz Fiadeiro, University of Leicester Alex Garcia, IME Jose Guimaraes, UFSCAR Hermann Haeusler, PUC-Rio Roberto Ierusalimschy, PUC-Rio (chair) Jose Emilio Labra, Univesidad de Oviedo Rafael Dueire Lins, UFPE Alfio Martini, PUC-RS Ricardo Massa, UPE Dale Miller, INRIA Hermano Moura, UFPE Martin Musicante, UFPR David Naumann, Stevens Institute of Technology Catuscia Palamidessi, INRIA Futurs Uday Reddy, University of Birmingham Noemi Rodriguez, PUC-Rio Cecilia Rubira, Unicamp Andre Santos, UFPE Marco Tulio Valente, PUC-MG Organizing Committee Lucilia Figueiredo (UFOP) - Chair Marco Tulio Valente (PUC Minas) Mariza Bigonha (UFMG) Symposium Location Listed by UNESCO as a World Heritage landmark, Ouro Preto ("black gold" in portuguese) gathers one the largest collections of baroque art in the world. Its churches, museums and public monuments attract tourists from all over the world. The city is located 100 km from Belo Horizonte, one of the largest cities from Brazil, in the state of Minas Gerais. More Information For further information, visit the symposium web site at http://www.inf.pucminas.br/sblp2003 or send e-mail to sblp2003(a)pucminas.br

1 0

RE: [OT] Teaching Haskell in High School
by Kevin Millikin 07 Feb '03

07 Feb '03

On Friday, February 07, 2003 7:41 AM, Arjan van IJzendoorn [SMTP:afie@cs.uu.nl] wrote: > The quality of the messages, however, is not as good as I would have > expected. I am a beginning Scheme programmer ... Then you should have set the language to "Beginning Student" or "Beginning Student with List Abbreviations". Seriously. A nice thing about the DrScheme environment is that it allows languages to be layered. Instructors don't have to wave their arms ands say "pay no attention to 'public static void' and 'String[] args' and 'class TestExp {'" just to begin writing simple programs. The DrScheme languages are restricted subsets of Scheme. The same could certainly be done in a learning environment for Haskell, but is more difficult in Java or C++. Even if the student is told to "ignore" some things for now, many of the problems they have are often caused by not understanding the things they're supposed to be ignoring--which can be incredibly frustrating. If you had set the DrScheme language to one of the two "Beginning Student" levels, you would have seen: function call: expected a defined name or a primitive operation name after an open parenthesis, but found a number And "+1" would have been highlighted. This would have been reported when you entered the *definition* for length2, not when you applied it. Switch to "Intermediate Student", and you get: function call: expected a name after an open parenthesis, but found a number With "+1" highlighted, at definition of length2. Switch to "Intermediate Student with lambda" or "Advanced Student", and the definition is accepted, but the application (to a non-empty list) reports: procedure application: expected procedure, given: 1; arguments were: 0 > I think it is important that error messages show exactly what was > entered by the user (+1 and not 1) and that's why in Helium we > remember each little detail in the abstract syntax tree. Excellent idea. > I still don't know what "arguments were: 1" means. Does it refer to > the number of arguments? Are you sure you it said that? Mine doesn't ;). I expect "arguments were: 0". It was complaining about trying to apply "1" to the argument that is the value of (length2 empty), which should be zero, no?

1 0