December 2010 - Beginners

Struggling with Haskell updates on Macintosh
by David Jameson 18 Dec '10

18 Dec '10

Hi, I'm trying to teach myself Haskell and have installed the platform on my Mac along with Leksah. I'm trying to use cabal to get some updates (in particular mtl so that I can get the gtk library working) and cabal wants to update itself. However, when I try to do this, I'm running into various issues 1) I get a lot of warnings about the Module Prelude being deprecated (possibly due to some legacy haskell installs?) 2) The update is being installed into a different location ~/.cabal rather than to /usr/local/bin/cabal (which itself is an alias to /Library/Frameworks/HaskellPlatform.framework/bin/cabal) so no matter how many times I try to update, I keep getting told there's a new version. So how can I make the update go into the correct location? Many thanks, David Jameson

2 1

Re: [Haskell-beginners] How to think in Haskell
by Russ Abbott 18 Dec '10

18 Dec '10

It seems to me that among the basic concepts to focus on when learning Haskell are Currying (or partial evaluation), functional composition as a way to produce another function (and not just as an evaluation strategy), and passing functions are arguments. In all three one has to think in terms of functions being values that are created dynamically and passed around. A good way to practice is to focus on writing point-free code and to use the $ syntax notation as much as possible. $ is nice in that it is the apply function. Thinking of it in those terms helps build intuition since (again) its first argument is a function that is passed to it. A nice thing about these concepts is that they all exist and can be played with at the code level. They can all be explored at the level of the operational semantics of the language--i.e., how it works when you write code. Virtually no theory is required to learn or use them. * -- Russ* * * On Sat, Dec 18, 2010 at 3:00 AM, <beginners-request(a)haskell.org> wrote: > Send Beginners mailing list submissions to > beginners(a)haskell.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.haskell.org/mailman/listinfo/beginners > or, via email, send a message with subject or body 'help' to > beginners-request(a)haskell.org > > You can reach the person managing the list at > beginners-owner(a)haskell.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Beginners digest..." > > > Today's Topics: > > 1. Re: How to think in Haskell (David Virebayre) > 2. How to think in Haskell (Jun HU) (Sean Charles) > 3. Re: How to think in Haskell (Jun HU) (Gregg Reynolds) > 4. Re: How to think in Haskell (Ertugrul Soeylemez) > 5. Computational Linguistic (Luca Ciciriello) > 6. Re: Computational Linguistic (SOLVED) (Luca Ciciriello) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 17 Dec 2010 13:10:45 +0100 > From: David Virebayre <dav.vire+haskell(a)gmail.com<dav.vire%2Bhaskell(a)gmail.com> > > > Subject: Re: [Haskell-beginners] How to think in Haskell > To: Brent Yorgey <byorgey(a)seas.upenn.edu> > Cc: beginners(a)haskell.org > Message-ID: > <AANLkTik3yaYtdOJ5ZOtxwj8xn5m7CjGrjL32RD80_0+Z(a)mail.gmail.com<AANLkTik3yaYtdOJ5ZOtxwj8xn5m7CjGrjL32RD80_0%2BZ(a)mail.gmail.com> > > > Content-Type: text/plain; charset=UTF-8 > > 2010/12/16 Brent Yorgey <byorgey(a)seas.upenn.edu>: > > On Thu, Dec 16, 2010 at 04:41:24PM +0100, Luca Ciciriello wrote: > >> Hi, here my path (in correct order) I followed to learn to think > functionally: > > >> 1) Curry-Howard Isomorphism > >> 2) Type Theory & Functional Programming > >> 3) The Hindley-Milner Type inference algorithm > >> 4) Basic Category Theory > >> 5) Notions of computation and monads > >> 6) Denotational semantics > >> 7) Monads for functional programming > >> 8) Theorems for free. > >> 9) A History of Haskell: Being Lazy With Class > > > These are all wonderful topics. ?But I strongly disagree with the > > notion that one must understand all of these before even starting a > > Haskell tutorial (!), or even that one must understand all of these to > > be able to "think functionally" in some sense. > > ... aaaaand I'm going to agree with Brent here. > > It took me a while to have many things "click", but I found it hard to > thoroughly read many papers. The mathematical notation which I'm not > familiar with often is a problem to stay focused. > > What helped for me was reading tutorials and practice. Reading about > people's problems on Caf?, and reading the replies. Re-reading several > times a tutorial, weeks apart. I wish I had Real world haskell was I > was starting, but I have it now and I'm glad I do. > > I can't claim I understand more than Haskell 98, but it's more than > enough for a start. > > David. > > > > ------------------------------ > > Message: 2 > Date: Fri, 17 Dec 2010 14:44:09 +0000 > From: Sean Charles <sean(a)objitsu.com> > Subject: [Haskell-beginners] How to think in Haskell (Jun HU) > To: beginners(a)haskell.org > Message-ID: <1292597049.3903.18.camel@sean-desktop> > Content-Type: text/plain; charset="UTF-8" > > I learned Smalltalk in 1999, then I found LISP because Alan Kay said it > was the most awesome language ever invented. That got me hooked into > using s-expressions for everything! > > Then I found Erlang and that was just brilliant, although I still missed > the s-expression format of LISP. > > Then, about seven months ago I decided to try Haskell, I bought the Real > World Haskell book and I've been trying to get better every day I can. > > I've been a software develoepr for aobut twenty-six years, from assembly > language on chips and DEC/VAX right through to today and without doubt > Haskell has been my single most frustrating learning experience ever. > > And that my friends is a compliment to Haskell! :) > > I am *not* a mathematician, I did 'A'-level maths and physics and > computers back when Clive Sinclair was a force to be reckoned with. > To be told I have to read a big list of stuff with titles that would > make people think I was speaking in tongues doesn't cut it for me. For > my day job I *have* to cut PHP/Drupal, so I applied learning Haskell and > made myself some tools that generate Drupal code! > > Make it relevant to what you do everyday, that way it will carry more > meaning for you, why spend hours head-scratching over some fancy pants > monadic kung-fu lesson when you can as easily learn to read a file and > process each line, do something with it and then create a new file. > > First rule of optimisation: Don't do it. It was how I felt learning LISP > too, there are so many functions to do things, which is the 'right' one? > Make it work first then you can step back and say, "Can I do this more > succinctly using more advanced language features?" > > Don't try to learn it all at once, it's depressing! I know! > > > BIG HELP: The single biggest thing that helped me was to download the > PDF slide-notes and watch both parts of Simon Peyton Jones talks, all > available here: > http://notes-on-haskell.blogspot.com/2007/08/more-spj.html > > It was when I realises that, unlike C++ for example, where there is an > implicit 'this' pointer stuffed in the stack, in Haskell programs there > is an implicit pointer to the type information being passed around which > made lots of things click for me like "How does it know?" LOL > > Haskell is the classic elephant sandwich; a formidable task, but > everytime I understand something new it feels great! > > Best of luck! > :) > Sean Charles > > > > > > > ------------------------------ > > Message: 3 > Date: Fri, 17 Dec 2010 09:18:42 -0600 > From: Gregg Reynolds <dev(a)mobileink.com> > Subject: Re: [Haskell-beginners] How to think in Haskell (Jun HU) > To: sean(a)objitsu.com > Cc: beginners(a)haskell.org > Message-ID: > <AANLkTimXZj__z2uzJB-T9tH0f-ftkcAUVCMb-nuFc3QQ(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > On Fri, Dec 17, 2010 at 8:44 AM, Sean Charles <sean(a)objitsu.com> wrote: > > > > > I've been a software develoepr for aobut twenty-six years, from assembly > > language on chips and DEC/VAX right through to today and without doubt > > Haskell has been my single most frustrating learning experience ever. > > > > And that my friends is a compliment to Haskell! :) > > > > Don't try to learn it all at once, it's depressing! I know! > > > > I took my first crack at Haskell 5+ years ago, was befuddled by the term > "constructor" and left completely in the dark by monads. So I moved on to > other things. It took me about three tries and a lot of research into > various corners of computer science (I was a liberal arts major) to reach > the point where I understood the terminology and began to grasp the big > picture. Fortunately there's a lot more information available for > beginners > now. > > Even though the likelihood of me ever getting paid to program in Haskell is > nil, what I've gained from studying it vastly outweighs the considerable > effort I put into it. I thought I was learning another programming > language; in fact, I was learning to think about computation, logic, and > mathematics in ways that were completely new and very enlightening to me. > To really really understand it all, you have to delve into lambda calculus > most obviously, but also intuitionistic mathematics and logic, proof > theory, > category theory, and various other fascinating topics. In 100 or even 10 > years such stuff will form part of a basic educational program, but at the > moment it's like a secret world of magical beasts. The down side is > there's > nobody to talk to about it (except via mailing lists) since almost nobody > outside of specialists even knows this world exists. > > > > > > BIG HELP: The single biggest thing that helped me was to download the > > PDF slide-notes and watch both parts of Simon Peyton Jones talks, all > > available here: > > http://notes-on-haskell.blogspot.com/2007/08/more-spj.html > > > His book "Implementation of Functional > Languages< > http://research.microsoft.com/en-us/um/people/simonpj/papers/slpj-book-1987… > >" > is also very helpful, even if you have no interest in actually implementing > functional languages. It (and various other pubs) has proven very helpful > in deciphering some of the arcane bits of terminology that often pop up in > discussions of Haskell and FP, such as boxing, bottom, "weak head normal > form", etc. You can skim the gory details and still get a lot out of it. > Plus, I just found the tutorial > version< > http://research.microsoft.com/en-us/um/people/simonpj/papers/pj-lester-book/ > > > . > > -Gregg >

1 0

Computational Linguistic
by Luca Ciciriello 18 Dec '10

18 Dec '10

Hi All. It's my impression or all "items" about computational linguistic are been ripped away from Hackage? Luca.

1 1

How to think in Haskell (Jun HU)
by Sean Charles 17 Dec '10

17 Dec '10

I learned Smalltalk in 1999, then I found LISP because Alan Kay said it was the most awesome language ever invented. That got me hooked into using s-expressions for everything! Then I found Erlang and that was just brilliant, although I still missed the s-expression format of LISP. Then, about seven months ago I decided to try Haskell, I bought the Real World Haskell book and I've been trying to get better every day I can. I've been a software develoepr for aobut twenty-six years, from assembly language on chips and DEC/VAX right through to today and without doubt Haskell has been my single most frustrating learning experience ever. And that my friends is a compliment to Haskell! :) I am *not* a mathematician, I did 'A'-level maths and physics and computers back when Clive Sinclair was a force to be reckoned with. To be told I have to read a big list of stuff with titles that would make people think I was speaking in tongues doesn't cut it for me. For my day job I *have* to cut PHP/Drupal, so I applied learning Haskell and made myself some tools that generate Drupal code! Make it relevant to what you do everyday, that way it will carry more meaning for you, why spend hours head-scratching over some fancy pants monadic kung-fu lesson when you can as easily learn to read a file and process each line, do something with it and then create a new file. First rule of optimisation: Don't do it. It was how I felt learning LISP too, there are so many functions to do things, which is the 'right' one? Make it work first then you can step back and say, "Can I do this more succinctly using more advanced language features?" Don't try to learn it all at once, it's depressing! I know! BIG HELP: The single biggest thing that helped me was to download the PDF slide-notes and watch both parts of Simon Peyton Jones talks, all available here: http://notes-on-haskell.blogspot.com/2007/08/more-spj.html It was when I realises that, unlike C++ for example, where there is an implicit 'this' pointer stuffed in the stack, in Haskell programs there is an implicit pointer to the type information being passed around which made lots of things click for me like "How does it know?" LOL Haskell is the classic elephant sandwich; a formidable task, but everytime I understand something new it feels great! Best of luck! :) Sean Charles

2 1

polymorphism vs. overloading
by Dennis Raddle 17 Dec '10

17 Dec '10

I looked up the difference between polymorphism and overloading (speaking of all languages in general), and apparently polymorphism is a general term for an operation that can accept arguments of different types. Overloading is a form of "ad-hoc polymorphism" in which the compiler chooses specific code to be executed by inspecting the arguments at compile time. Apparently "true polymorphism" means exactly the same code is executed no matter what the argument types, such as implemented by inheritance in C++ (*) Now my understanding of Haskell's classes is that the compiler chooses the right instance of the class at compile time. Does that mean that Haskell implements only ad-hoc polymorphism? (*) Another question: if someone says "What feature of C++ makes polymorphism possible?" What is the correct answer? Inheritance? Virtual methods? Some combination? -D

2 1

polymorphism vs. overloading
by Eugene Perederey 16 Dec '10

16 Dec '10

Haskell supports parametric polymorphism in the sense as, for example, operations on lists. Thus function List.length::[a]->Int is polymorphic. The difference between polymorphism and overloading is that polymorphism basically means that there is one algorithm for different types of operands whereas overloading means that for one symbol there are different implementations. Haskell compiler converts an overloaded function in such way that the function is passed an additional argument -- a dictionary that contains implementation of operations defined in the class. Consider function square: square :: Num n => n -> n square x = x*x class Num a where (+) :: a -> a -> a (-) :: a -> a -> a negate :: a->a ...etc... instance Num Int where (+) = intPlus a b (*) = intMult a b negate a = intNeg a ...etc... where functions intPlus, intMult, intNeg are defined somewhere else. So at compile time the type of square :: Num n => n -> n is replaced with square :: numDict n -> n -> n where numDict a is a parametric data type with defined accessor functions that return implementation of operations defined in the class. On 16 December 2010 19:34, Dennis Raddle <dennis.raddle(a)gmail.com> wrote: > I looked up the difference between polymorphism and overloading (speaking of > all languages in general), and apparently polymorphism is a general term for > an operation that can accept arguments of different types. Overloading is a > form of "ad-hoc polymorphism" in which the compiler chooses specific code to > be executed by inspecting the arguments at compile time. Apparently "true > polymorphism" means exactly the same code is executed no matter what the > argument types, such as implemented by inheritance in C++ (*) Now my > understanding of Haskell's classes is that the compiler chooses the right > instance of the class at compile time. Does that mean that Haskell > implements only ad-hoc polymorphism? > > > (*) Another question: if someone says "What feature of C++ makes > polymorphism possible?" What is the correct answer? Inheritance? Virtual > methods? Some combination? > > -D > > > _______________________________________________ > Beginners mailing list > Beginners(a)haskell.org > http://www.haskell.org/mailman/listinfo/beginners > > -- Best, Eugene Perederey -- Best, Eugene Perederey

1 0

mapping over newtypes
by Tim Baumgartner 15 Dec '10

15 Dec '10

Hello Haskellers, I defined newtype Banana = Banana Something When I work with it, I sometimes use a function mapBanana f (Banana b) = Banana (f b) which is of course discarded by the compiler. Is this usual practice or are there any better solutions? Regards Tim

4 5

haven
by jean verdier 15 Dec '10

15 Dec '10

I was reading http://www.haskell.org/haskellwiki/Applications_and_libraries/Graphics and i'm interested in haven but the link is dead: http://haskell.org/haven/ Can someone give me some some pointer ?

3 2

Type unions
by Russ Abbott 15 Dec '10

15 Dec '10

Is there a way to get this to work? data A = Aconstructor Int data B = Bconstructor Int data AorB = A | B f :: Int -> AorB f x | even x = Aconstructor x | otherwise = Bconstructor x I get this diagnostic. Couldn't match expected type `AorB' against inferred type `A' Since AorB is A or B, why is this not permitted? If instead I write data AorB = Aconstructor Int | Bconstructor Int everything works out ok. But what if I want separate types for A and B? Thanks, * -- Russ *

8 21

Re: [Haskell-beginners] Equivalence of Inheritance
by Russ Abbott 14 Dec '10

14 Dec '10

That's good. (It's more or less the way I was doing it.) What I wanted to avoid was this. getGenderSpecificCondition ( Man _ _ cond) = cond getGenderSpecificCondition (Woman _ _ cond) = cond I know it seems like a small thing, but I would like to be able to write it like this. getGenderSpecificCondition p | p == (Man _ _ cond) = cond | p == (Woman _ _ cond) = cond But that's not legal syntax. A pattern can't appear in that context. But this does the job. getGenderSpecificCondition :: Person -> Condition getGenderSpecificCondition p | isMan p = prostateCondition p | isWoman p = ovaryCondition p isMan ( Man _ _ cond) = True isMan _ = False isWoman (Woman _ _ cond) = True isWoman _ = False That works! prostateCondition and ovaryCondition are both defined on Person. (I'm surprised to see that.) *Person> Group [Man "Harry" 32 OK, Woman "Sally" 29 Good] Harry(32, OK) Sally(29, Good) Also *Person> prostateCondition (Woman "Sally" 29 Good) *** Exception: No match in record selector prostateCondition *Person> prostateCondition (Man "Harry" 29 Good) Good *-- Russ * * * On Tue, Dec 14, 2010 at 12:31 PM, Michael Katelman <katelman(a)uiuc.edu>wrote: > Perhaps this? > > https://gist.github.com/741048 > > -Mike > > On Tue, Dec 14, 2010 at 2:27 PM, Russ Abbott <russ.abbott(a)gmail.com> > wrote: > > What I'm after is a version of my example that compiles. Can you make > one? > > > > -- Russ > > > > > > On Tue, Dec 14, 2010 at 12:18 PM, Antoine Latter <aslatter(a)gmail.com> > wrote: > >> > >> Sorry, I really don't know enough about what you're after to attempt > that. > >> > >> But you'll need to change you're signatures of the form: > >> > >> > function :: Person -> Foo > >> > >> to something of the form: > >> > >> > function :: Person p => p -> Foo > >> > >> Because again, a type class can not be used as a type. > >> > >> Antoine > >> > >> On Tue, Dec 14, 2010 at 2:12 PM, Russ Abbott <russ.abbott(a)gmail.com> > >> wrote: > >> > What got fouled up is all the adjustments I had to make to the other > >> > declarations. > >> > Can you complete the example so that it compiles using > >> > > >> > class Person p where ... > >> > > >> > I'd very much like to see an example that actually compiles. > >> > > >> > Thanks. > >> > -- Russ > >> > > >> > On Tue, Dec 14, 2010 at 11:58 AM, Antoine Latter <aslatter(a)gmail.com> > >> > wrote: > >> >> > >> >> On Tue, Dec 14, 2010 at 1:52 PM, Russ Abbott <russ.abbott(a)gmail.com> > >> >> wrote: > >> >> > If gender is a field in a Person type, then a Person must have both > >> >> > an > >> >> > ovaryCondition and a prostateCondition. That seems awkward. > >> >> > Regarding > >> >> > class Person p where > >> >> > I started down that path but got completely fouled up. > >> >> > >> >> How did this get fouled up? Every class declaration must take > >> >> arguments - here, 'p' is the argument for the class. > >> >> > >> >> Thanks, > >> >> Antoine > >> > > >> > > > > > > > _______________________________________________ > > Beginners mailing list > > Beginners(a)haskell.org > > http://www.haskell.org/mailman/listinfo/beginners > > > > >

2 4