Re: [Haskell-cafe] A Mascot
Hi all, This is my sister's proposition: http://origami.bieszczady.pl/images/The_Lamb_Da.png What do you think? Best, Karol Samborski ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
Tim Baumgartner wrote: Thanks a lot! Althaugh I have some understanding of the Haskell basics and the most important monads, I feel that I have to see more well designed code in order to become a good Haskeller. Can somebody make suggestions what materials are best to work through in order to achieve this? Are there easy research papers about Haskell programming? Or should I try the Monad.Reader? I'm looking for topics that either can be used directly in many situations or that show some functional principles that boost my creativity and functional thinking. You may want to start with the Functional Pearls http://www.haskell.org/haskellwiki/Research_papers/Functional_pearls In particular, I recommend * Richard Bird. A program to solve Sudoku. * Graham Hutton. The countdown problem. * Martin Erwig and Steve Kollmansberger. Probabilistic functional programming in Haskell. * Conor McBride and Ross Paterson. Applicative Programming with Effects. Best regards, Heinrich Apfelmus -- http://apfelmus.nfshost.com ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Lifted Spine View
Hi there. I tried to follow the program of the paper Scrap your boilerpolate Revolutions. Unfortunately, I found the program in the section lifted spine view does not compile in my GHC, could anybody point out where I am wrong? Many Thanks My code is posted here http://hpaste.org/54357 You have to flip the two fields of (:-), i.e., the type has to be first and the annotated term has to be second. This is because pattern matching on GADTs and refinement is implicitly left-to-right in GHC. The paper presents it the other way round and remarks on the flipped order in a footnote near the beginning. Cheers, Andres ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Superset of Haddock and Markdown
On 20 Nov 2011, at 22:20, Ivan Lazar Miljenovic wrote: On 21 November 2011 03:19, David Fox dds...@gmail.com wrote: On Fri, Nov 18, 2011 at 1:10 AM, Ertugrul Soeylemez e...@ertes.de wrote: Ivan Lazar Miljenovic ivan.miljeno...@gmail.com wrote: Wasn't there talk at one stage of integrating pandoc into haddock? I wouldn't mind Haddock depending on Pandoc, at least optionally (-fmarkdown-comments). Taking this to its conclusion you could easily have syntax-highlighted code examples in Haddock documentations and allow alternative output formats. I'm not sure the pandoc license (GPL) is compatible with the GHC license. Do you mean because GHC ships with a Haddock binary? GHC also ships with a GPL'd gcc binary, on at least some platforms. Regards, Malcolm ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Decision procedure for foldr/foldl/foldl'?
I would say a good practice with folds (and maybe in Haskell in general) is that either all be strict or all be lazy. In the expression: foldXX f init list: Remember that foldr does: x `f` ( ... the accumulator ... ) and foldl: (... the accumulator ...) `f` x The accumulator has to match a non-strict argument of* f*, so that *f* will be able to return results *even if* the accumulator is not fully evaluated. More detailed: if *f* is lazy in its second argument, then use foldr. Everything is lazy, you build a very small thunk since nothing is evaluated. In the rare cases where* f *is (also) lazy in its first argument, you can use foldl. And of course, if *f* is strict in both its arguments, then use foldl'. Everything is then strict, you build no thunk. 2011/11/20 David Fox dds...@gmail.com Does anyone have a quick way to decide which of the fold functions to use in a given situation? There are times when I would like to find out which to use in the quickest way possible, rather than reading a long explanation of why each one behaves the way it does. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Decision procedure for foldr/foldl/foldl'?
Yves Parès: if *f* is lazy in its second argument, then use foldr. Everything is lazy, you build a very small thunk since nothing is evaluated. In the rare cases where*f *is (also) lazy in its first argument, you can use foldl. ... I have the impression that this is not the most useful advice possible. 1. Nothing is evaluated?? Look, foldr is designed to consume incrementally the list argument, and to produce, also incrementally the result ; it may stop in the middle if f is lazy, but if you organize badly your program, you will pollute your memory. You might wish to process the whole of the list as fast as possible, and then foldr may be dangerous. You may build a veeery big thunk before its reduction. 2. This is not only the issue: f x z versus f z x. foldl is iterative (tail-recursive) and the reduction proceeds until all the source list is consumed. foldl works better with strict functions, not lazy. (of course, unless I am mistaken...) == In general, sorry for the cynism, but when I read: There are times when I would like to find out which to use in the quickest way possible, rather than reading a long explanation of why each one behaves the way it does of David Fox, I compare it with a question of a young army officer, addressed to his elders: Tell me how to win the war in the quickest way possible, rather than boring me with the explanations behind all those complicated strategies. Jerzy Karczmarczuk Caen, Normandy, France (William the Conqueror, who lived here, had a reputation of a strategist who tried to understand his enemies. 350 years later, the French didn't try to understand anything, they just wanted to win the battle of Azincourt as quick as possible). ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
2011/11/21 Karol Samborski edv.ka...@gmail.com: Hi all, This is my sister's proposition: http://origami.bieszczady.pl/images/The_Lamb_Da.png What do you think? Second version: http://origami.bieszczady.pl/images/The_Lamb_Da2.png Best, Karol Samborski ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Drawing charts over a lot of data
Hi cafe, Are there any (possibly unfinished?) libraries dedicated to drawing charts over large amounts of data? I mean, such amounts that you don't want to store the whole set of input data in memory and instead you prefer to do one or a few passes over the input, calling the library's drawing functions for each data point. I'm currently using the excellent Chart package http://hackage.haskell.org/package/Chart , but it seems to require quite a bit of work to become usable in this mode (I haven't looked really deep yet). So, I'd be happy if someone pointed me to an existing package (I didn't find one on hackage) or if someone expressed interest in making Chart avoid having the whole input in memory. -- Eugene Kirpichov Principal Engineer, Mirantis Inc. http://www.mirantis.com/ Editor, http://fprog.ru/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Drawing charts over a lot of data
This doesn't directly solve your problem, but you may want to take a look at zoom-cache [1]. I've never used it myself, but it seems pretty nice. Cheers, [1] http://hackage.haskell.org/package/zoom-cache -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
I've read Martin Erwig and Steve Kollmansberger's *Probabilistic functional programming in Haskell*. Does someone know if the library they are talking about is available on hackage? 2011/11/21 Heinrich Apfelmus apfel...@quantentunnel.de Tim Baumgartner wrote: Thanks a lot! Althaugh I have some understanding of the Haskell basics and the most important monads, I feel that I have to see more well designed code in order to become a good Haskeller. Can somebody make suggestions what materials are best to work through in order to achieve this? Are there easy research papers about Haskell programming? Or should I try the Monad.Reader? I'm looking for topics that either can be used directly in many situations or that show some functional principles that boost my creativity and functional thinking. You may want to start with the Functional Pearls http://www.haskell.org/**haskellwiki/Research_papers/**Functional_pearlshttp://www.haskell.org/haskellwiki/Research_papers/Functional_pearls In particular, I recommend * Richard Bird. A program to solve Sudoku. * Graham Hutton. The countdown problem. * Martin Erwig and Steve Kollmansberger. Probabilistic functional programming in Haskell. * Conor McBride and Ross Paterson. Applicative Programming with Effects. Best regards, Heinrich Apfelmus -- http://apfelmus.nfshost.com __**_ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/**mailman/listinfo/haskell-cafehttp://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Decision procedure for foldr/foldl/foldl'?
On Mon, Nov 21, 2011 at 4:44 AM, Jerzy Karczmarczuk jerzy.karczmarc...@unicaen.fr wrote: In general, sorry for the cynism, but when I read: There are times when I would like to find out which to use in the quickest way possible, rather than reading a long explanation of why each one behaves the way it does of David Fox, I compare it with a question of a young army officer, addressed to his elders: Tell me how to win the war in the quickest way possible, rather than boring me with the explanations behind all those complicated strategies. I'm not trying to avoid learning the differences between the different folds, but I am looking for a mnemonic device that will allow me to proceed more quickly towards my goal. My ultimate goal is to write software, not to understand folds. Just as it is inappropriate for a young officer to even contemplate an overall strategy for winning the war, it would be inappropriate for a general to spend more time than necessary on the minute details of military tactics, as vital as they are. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Deduce problem.
Magicloud Magiclouds wrote:
So I think I got what you guys meant, I limited ClassB to only H.
Then how to archive my requirement, that from and to only return items
that instanced ClassB?
If you are willing to go beyond Haskell98 (or Haskell2010), you can use a
multi-parameter class. Enable the extension:
{-# LANGUAGE MultiParamTypeClasses #-}
An then, instead of
class (ClassA a) = ClassC a where
from :: (ClassB b) = a - [b]
to :: (ClassB c) = a - [c]
you say
class (ClassA a, ClassB b) = ClassC a b c where
from :: c - [b]
to :: c - [a]
This means that for each triple of concrete types (a,b,c) that you wish to
be an instance of ClassC, you must provide an instance declaration, e.g.
instance ClassC Test H H where
from = ...whatever...
to = ...whatever...
Now you have the fixed type H in the instance declaration and not a
universally quantified type variable.
Cheers
Ben
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Superset of Haddock and Markdown
On Sun, Nov 20, 2011 at 2:20 PM, Ivan Lazar Miljenovic ivan.miljeno...@gmail.com wrote: On 21 November 2011 03:19, David Fox dds...@gmail.com wrote: On Fri, Nov 18, 2011 at 1:10 AM, Ertugrul Soeylemez e...@ertes.de wrote: Ivan Lazar Miljenovic ivan.miljeno...@gmail.com wrote: Wasn't there talk at one stage of integrating pandoc into haddock? I wouldn't mind Haddock depending on Pandoc, at least optionally (-fmarkdown-comments). Taking this to its conclusion you could easily have syntax-highlighted code examples in Haddock documentations and allow alternative output formats. I'm not sure the pandoc license (GPL) is compatible with the GHC license. Do you mean because GHC ships with a Haddock binary? Haddock carries the same license as GHC. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Superset of Haddock and Markdown
On Mon, Nov 21, 2011 at 12:28, David Fox dds...@gmail.com wrote: On Sun, Nov 20, 2011 at 2:20 PM, Ivan Lazar Miljenovic ivan.miljeno...@gmail.com wrote: On 21 November 2011 03:19, David Fox dds...@gmail.com wrote: I'm not sure the pandoc license (GPL) is compatible with the GHC license. Do you mean because GHC ships with a Haddock binary? Haddock carries the same license as GHC. More to the point, Haddock uses ghc internals these days; it's not just a matter of bundling, and the licenses *must* be compatible. -- brandon s allbery allber...@gmail.com wandering unix systems administrator (available) (412) 475-9364 vm/sms ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
On 21 November 2011 14:48, Yves Parès limestr...@gmail.com wrote: I've read Martin Erwig and Steve Kollmansberger's Probabilistic functional programming in Haskell. Does someone know if the library they are talking about is available on hackage? Henning Thielemann has a batteries included version on Hackage: http://hackage.haskell.org/package/probability I'd expect the original (micro) library is still available from Martin Erwig's website. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
Free Monads. It's amazing to be confronted again with notions I learned more than ten years ago for groups. I have to admit that I'm probably not yet prepared for a deeper understanding of this, but hopefully I will return to it later ;-) Is Cont free as well? I guess so because I heard it's sometimes called the mother of all monads. Regards Tim 2011/11/21 David Menendez d...@zednenem.com On Sat, Nov 19, 2011 at 3:29 PM, Felipe Almeida Lessa felipe.le...@gmail.com wrote: On Sat, Nov 19, 2011 at 6:08 PM, Tim Baumgartner baumgartner@googlemail.com wrote: I have not yet gained a good understanding of the continuation monad, but I wonder if it could be used here. What would a clean solution look like? Perhaps there are other things that need to be changed as well? Your 'Interaction' data type is actually an instance of the more general operational monad (as named by Heinrich Apfelmus) or prompt monad (as named by Ryan Ingram). Both of which are just disguised free monads. For reference: data Free f a = Val a | Wrap (f (Free f a)) foldFree :: Functor f = (a - b) - (f b - b) - Free f a - b foldFree v w (Val a) = v a foldFree v w (Wrap t) = w $ fmap (foldFree v w) t instance Functor f = Monad (Free f) where return = Val m = f = foldFree f Wrap m To use Free, just find the signature functor for Interaction by replacing the recursive instances with a new type variable, data InteractionF a b x = ExitF b | OutputF b x | InputF (a - x) instance Functor (InteractionF a b) where fmap f (ExitF b) = ExitF b fmap f (OutputF b x) = OutputF b (f x) fmap f (InputF g)= InputF (f . g) roll :: InteractionF a b (Interaction a b) - Interaction a b roll (ExitF b) = Exit b roll (OutputF b x) = Output b x roll (InputF g)= Input g type InteractionM a b = Free (InteractionF a b) runM :: InteractionM a b b - Interaction a b runM = foldFree Exit roll exit :: b - InteractionM a b c exit b = Wrap (ExitF b) output :: b - InteractionM a b () output b = Wrap (OutputF b (Val ())) input :: InteractionM a b a input = Wrap (InputF Val) -- Dave Menendez d...@zednenem.com http://www.eyrie.org/~zednenem/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
Hi Heinrich, I read your article about the operational monad and found it really very enlightening. So I'm curious to work through the material you linked below. Thanks! Regards Tim 2011/11/21 Heinrich Apfelmus apfel...@quantentunnel.de Tim Baumgartner wrote: Thanks a lot! Althaugh I have some understanding of the Haskell basics and the most important monads, I feel that I have to see more well designed code in order to become a good Haskeller. Can somebody make suggestions what materials are best to work through in order to achieve this? Are there easy research papers about Haskell programming? Or should I try the Monad.Reader? I'm looking for topics that either can be used directly in many situations or that show some functional principles that boost my creativity and functional thinking. You may want to start with the Functional Pearls http://www.haskell.org/**haskellwiki/Research_papers/**Functional_pearlshttp://www.haskell.org/haskellwiki/Research_papers/Functional_pearls In particular, I recommend * Richard Bird. A program to solve Sudoku. * Graham Hutton. The countdown problem. * Martin Erwig and Steve Kollmansberger. Probabilistic functional programming in Haskell. * Conor McBride and Ross Paterson. Applicative Programming with Effects. Best regards, Heinrich Apfelmus -- http://apfelmus.nfshost.com __**_ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/**mailman/listinfo/haskell-cafehttp://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
Cute! I like it! On Mon, Nov 21, 2011 at 7:52 AM, Karol Samborski edv.ka...@gmail.comwrote: 2011/11/21 Karol Samborski edv.ka...@gmail.com: Hi all, This is my sister's proposition: http://origami.bieszczady.pl/images/The_Lamb_Da.png What do you think? Second version: http://origami.bieszczady.pl/images/The_Lamb_Da2.png Best, Karol Samborski ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe -- Heath Matlock +1 256 274 4225 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Lifted Spine View
Thanks, I don't see the footnote, but that works. :-) On Mon, Nov 21, 2011 at 5:15 AM, Andres Löh andres.l...@googlemail.comwrote: Hi there. I tried to follow the program of the paper Scrap your boilerpolate Revolutions. Unfortunately, I found the program in the section lifted spine view does not compile in my GHC, could anybody point out where I am wrong? Many Thanks My code is posted here http://hpaste.org/54357 You have to flip the two fields of (:-), i.e., the type has to be first and the annotated term has to be second. This is because pattern matching on GADTs and refinement is implicitly left-to-right in GHC. The paper presents it the other way round and remarks on the flipped order in a footnote near the beginning. Cheers, Andres -- Best, bob ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
On Mon, Nov 21, 2011 at 2:13 PM, Tim Baumgartner baumgartner@googlemail.com wrote: Free Monads. It's amazing to be confronted again with notions I learned more than ten years ago for groups. I have to admit that I'm probably not yet prepared for a deeper understanding of this, but hopefully I will return to it later ;-) Is Cont free as well? I guess so because I heard it's sometimes called the mother of all monads. As I understand it, Cont is not a free monad, but there is a connection between the ideas. Certainly, any free monad can be easily reimplemented using Cont. Here's how you might implement your monad using Cont, type InteractionM a b = Cont (Interaction a b) exit b = Cont $ \k - Exit b output b = Cont $ \k - Output b (k ()) input= Cont $ \k - Input k runM m = runCont m Exit That's very similar to the free monad's implementation, only with the continuation replacing Val. exit b = Wrap $ ExitF b output b = Wrap $ OutputF b (Val ()) input= Wrap $ InputF Val runM m = foldFree Exit roll m The Cont-based version has essentially taken the work performed in foldFree and distributed it to return and (=). This eliminates the intermediate data structures, resulting in a more efficient implementation. -- Dave Menendez d...@zednenem.com http://www.eyrie.org/~zednenem/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Superset of Haddock and Markdown
On 21 November 2011 17:34, Brandon Allbery allber...@gmail.com wrote: Haddock carries the same license as GHC. More to the point, Haddock uses ghc internals these days; it's not just a matter of bundling, and the licenses *must* be compatible. No. If the haddock library any program that links against that library (e.g., the haddock program) effectively becomes GPL licensed (roughly speaking). However, ghc and none of the core libraries link against haddock, so that's not a problem. You could run into issues if you wanted to write a closed-source program that links against the haddock library. I generally don't like GPL-licensed libraries for that reason, but in this case I think it would be fine. (Of course, the issue is that many tools start out as programs and then turn into libraries -- pandoc is a good example. At that point GPL becomes an issue again.) -- Push the envelope. Watch it bend. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
2011/11/21 David Menendez d...@zednenem.com Here's how you might implement your monad using Cont, type InteractionM a b = Cont (Interaction a b) exit b = Cont $ \k - Exit b output b = Cont $ \k - Output b (k ()) input= Cont $ \k - Input k runM m = runCont m Exit That's what I originally wanted to know. I guess I struggled with the definition of output. Oh, there's so much more to learn... Thanks, Tim ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Interpreter with Cont
You'll probably get answers from people who are more proficient with this, but here's what I learned over the years. Tim Baumgartner wrote: Is Cont free as well? No. In fact, free monads are quite a special case, many monads are not free, e.g. the list monad. I believe what David Menendez said was meant to mean 'modulo some equivalence relation' i.e. you can define/implement many monads as 'quotients' of a free monad. But you cannot do this with Cont (though I am not able to give a proof). I guess so because I heard it's sometimes called the mother of all monads. It is, in the sense that you can implement all monads in terms of Cont. Cheers Ben ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Job Opportunity at Parallel Scientific
Parallel Haskell Programmers Parallel Scientific, LLC is a Boulder, CO based early stage, but funded startup company working in the area of scalable parallelization for scientific and large data computing. We are implementing radically new software tools for the creation and optimization of parallel programs benefiting applications and leveraging modern systems architecture. We build on our mathematical knowledge, cutting edge programming languages and our understanding of systems software and hardware. We are currently working with the Haskell development team and major HPC laboratories world wide on libraries and compiler extensions for parallel programming. Parallel Scientific was founded by Peter Braam in 2010. Peter formerly taught mathematics and computer science at Oxford and Carnegie Mellon. Then he contributed file systems to Linux and invented Lustre (which provides storage to 9 of the top 10 systems in the world). He ran several successful startups, and Parallel Scientific is run by an very experienced management team and board. Successful candidates can in some cases work remotely and will work in a modern virtual environment. We provide training in advanced processes for software design and implementation and domain specific knowledge. Required skills: - Very strong background in computer science or mathematics - Experience with Haskell - Knowledge of systems programming and operating systems functionality - Knowledge of system architectures, such as high performance networking, memory architectures, multi and manycore CPUs and GPGPUs - Experience with performance tuning of parallel or concurrent algorithms - Experience designing and implementing concurrent or parallel programs Experience in one or more of the following areas is desirable: - Experience with Haskell compiler technology - In depth knowledge of core Haskell libraries for parallel programming (NDP, REPA etc) - Experience in the area of middleware algorithms for data flow programming, graphs, cloud based data analytics, or sparse matrices - Domain specific knowledge in scientific programming areas related to irregular and sparse problems, e.g. data analytics using graph analysis, genomics, tightly connected numerical analysis - Experience with performance tuning for parallel applications on multi core systems (e.g. with the Intel toolkit), for GPGPU's using Cuda/OpenCL or with MPI/OpenMP on clusters of SMPs To apply, please send a resume to j...@parsci.com. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
heathmatlock wrote: Cute! I like it! Yea, it's cute. I don't like the formula, though: \x - x + x is just too trivial and not very Haskellish. Something higher order is the minimum requirement, IMO. The original (lambda knights) formula was cool: the fixed point operator is directly related to recursion, which is reflected in the picture that contains itself; note also that defining this operator requires an untyped language, so this fits LISP quite well (but not Haskell). What about the formula for function composition (f . g) x = f (g x) maybe together with its type (or maybe only the type) (.) :: (b - c) - (a - b) - a - c Extremely cool are GADTs, such as data Eq a b where Refl :: Eq a a Or, if you'd like something more obscure but still at the center of what Haskell is about, take the mother of all monads m = f = \k - m (\a - (f a) k) This is a formula I can spend a day contemplating and still wonder if I have _really_ understood it. And doesn't that properly reflect the depth and richness of Haskell? Cheers Ben On Mon, Nov 21, 2011 at 7:52 AM, Karol Samborski edv.ka...@gmail.comwrote: 2011/11/21 Karol Samborski edv.ka...@gmail.com: Hi all, This is my sister's proposition: http://origami.bieszczady.pl/images/The_Lamb_Da.png What do you think? Second version: http://origami.bieszczady.pl/images/The_Lamb_Da2.png Best, Karol Samborski ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Drawing charts over a lot of data
On 21 November 2011 22:36, Felipe Almeida Lessa felipe.le...@gmail.com wrote: This doesn't directly solve your problem, but you may want to take a look at zoom-cache [1]. I've never used it myself, but it seems pretty nice. Cheers, [1] http://hackage.haskell.org/package/zoom-cache Hi, zoom-cache is useful for managing time-series data. There is a zoom-cache-gnuplot in development, and it would probably be useful to make a tool that uses Chart. I'm happy to help with that :) Conrad. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Drawing charts over a lot of data
On Mon, Nov 21, 2011 at 3:47 PM, Conrad Parker con...@metadecks.org wrote: zoom-cache is useful for managing time-series data. There is a zoom-cache-gnuplot in development, and it would probably be useful to make a tool that uses Chart. I'm happy to help with that :) Be aware that Chart is pretty slow on large data sets. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ST not strict enough?
On Fri, Nov 18, 2011 at 4:05 AM, Yves Parès limestr...@gmail.com wrote: (Sorry for the double mail) ...so there is no way to do that inside the function passed to modifySTRef? In other words, there is no way to ensure *inside* a function that its result will be evaluated strictly? modifySTRef looks like this (sugared up a bit): modifySTRef r f = do x - readSTRef r writeSTRef r (f x) Note that this writes the *thunk* (f x) to the STRef, and there is absolutely nothing you can do about it. f isn't demanded by this function, let alone f x. You can implement your own strict modifySTRef' easily though, as other people in this thread have shown. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
On 21/11/2011, at 9:22 PM, Karol Samborski wrote: Hi all, This is my sister's proposition: http://origami.bieszczady.pl/images/The_Lamb_Da.png What do you think? It looks like a skittle with a baby bonnet. C'est mignon, mais ce n'est pas la guerre as Pierre Bosquet almost said. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
Yea, it's cute. I don't like the formula, though: \x - x + x is just too trivial and not very Haskellish. Something higher order is the minimum requirement, IMO. The original (lambda knights) formula was cool: the fixed point operator is directly related to recursion, which is reflected in the picture that contains itself; note also that defining this operator requires an untyped language, so this fits LISP quite well (but not Haskell). I would go with something like ! forall A B. A - B, saying that type casting under the C-H isomorphism is a lie and therefore we all must avoid it :D, but I don't like the ! in front of it. Just my 2 cents. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
I think the artwork is nice, but I am not sure that a lamb is an appropriate mascot for Haskell. A mascot is supposed to represent characteristics, emotions, or desires that a particular group of people aspire to have, be like, etc. To outsiders, it provides a quick way to see if it might be a group they would like to belong to, and for insiders, it helps strengthen the bond and group identity by reminding them what they stand for. So far, the only justification I have noticed for why a lamb would represent Haskell users is that there is a pun about lambda's -- which only makes sense if you know English. Sheep are generally thought of as: - weak and needing protection - easily lead astray - being lead to the slaughter - dumb and easily lost Not sure those are traits that Haskeller's generally aspire to have. I think Haskeller's like Haskell because it is: - elegant - sophisticated - reliable - robust Haskeller's tend to be people who are curious. Pioneers who are willing to go off the beaten path in search of something better. People who are willing to evaluate something based on its merits rather than the mere approval of the mainstream. People who aspire to create elegant, beautiful code. People looking to better their skills, even if they don't use Haskell for most of their coding. And there is definitely a pragmatic aspect. Part of the appeal of Haskell is that it can actually be used for many real world applications and can often do the job better. The fact that you can use it to deliver more reliable and robust code in less time, is a very real and tangible benefit. Here are some suggestions of my own. I am not really excited about any of them either -- but they give some examples of how I think a mascot might work: - owl: traditionally thought of as 'wise'. Known for their keen (in)sight. Of course, some cultures believe they are a bad omen and a sign of impending death.. - honey badger - can't beat that for 'robust' and 'fearless', http://www.youtube.com/watch?v=wPKlryXwmXk - james bond - he's sophisticated, reliable, and he does it with 'class'. hahah, more silly puns :p Of course, he is also not public domain :) Plus, it is too male oriented. In summary, a mascot is supposed to elicit an emotional response from people and help create a bond. To do that, it needs to provide emotional leadership and say that, if you use Haskell, you can be like X. That doesn't it mean it can't be cute. People do tend to bond easily to cute things (like kittens!). But I don't think cute is enough. I also don't think that representing 'features' of Haskell, like 'laziness' or 'higher order' is the right core appeal either. That is too mental -- not enough emotion. Those things can, of course, be represented in the depiction of the mascot. Nothing wrong with cleverly hiding lamba's and _|_ in the picture. But, for example, saying that Haskell is 'lazy' so we should pick a sloth, is not really a good choice, IMO. - jeremy On Tue, Nov 15, 2011 at 7:01 PM, heathmatlock heathmatl...@gmail.com wrote: I liked Go's mascot, and I figure it couldn't hurt to have our own. I spent the past hour making this: http://i.imgur.com/Mib6Q.png What do you think? -- Heath Matlock +1 256 274 4225 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] A Mascot
On Tue, Nov 22, 2011 at 12:22 AM, Jeremy Shaw jer...@n-heptane.com wrote: - honey badger - can't beat that for 'robust' and 'fearless', http://www.youtube.com/watch?v=wPKlryXwmXk i think you were referring to this vid: http://www.youtube.com/watch?v=m7pGZudN8rE (nsfw... almost) i +1 a honey badger for haskell. i'm a newb and it rocks!! hex -- * my blog is cooler than yours: http://serialhex.github.com * The wise man said: Never argue with an idiot. They bring you down to their level and beat you with experience. * As a programmer, it is your job to put yourself out of business. What you do today can be automated tomorrow. ~Doug McIlroy --- CFO: “What happens if we train people and they leave?” CTO: “What if we don’t and they stay?” ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] FW: [Haskell] Job Opportunity at Parallel Scientific
I'm taking the liberty of forwarding this super-relevant advert to Haskell Cafe in case any readers read the Cafe but not the main Haskell mailing list. Simon From: haskell-boun...@haskell.org [mailto:haskell-boun...@haskell.org] On Behalf Of Peter Braam Sent: 21 November 2011 22:23 To: haskell Subject: [Haskell] Job Opportunity at Parallel Scientific Parallel Haskell Programmers Parallel Scientific, LLC is a Boulder, CO based early stage, but funded startup company working in the area of scalable parallelization for scientific and large data computing. We are implementing radically new software tools for the creation and optimization of parallel programs benefiting applications and leveraging modern systems architecture. We build on our mathematical knowledge, cutting edge programming languages and our understanding of systems software and hardware. We are currently working with the Haskell development team and major HPC laboratories world wide on libraries and compiler extensions for parallel programming. Parallel Scientific was founded by Dr Peter Braam in 2010. Peter formerly taught mathematics and computer science at Oxford and Carnegie Mellon. Then he contributed file systems to Linux and invented Lustre (which provides storage to 9 of the top 10 systems in the world). He ran several successful startups, and Parallel Scientific is run by an very experienced management team and board. Successful candidates can in some cases work remotely and will work in a modern virtual environment. We provide training in advanced processes for software design and implementation and domain specific knowledge. Required skills: * Very strong background in computer science or mathematics * Experience with Haskell * Knowledge of systems programming and operating systems functionality * Knowledge of system architectures, such as high performance networking, memory architectures, multi and manycore CPUs and GPGPUs * Experience with performance tuning of parallel or concurrent algorithms * Experience designing and implementing concurrent or parallel programs Experience in one or more of the following areas is desirable: * Experience with Haskell compiler technology * In depth knowledge of core Haskell libraries for parallel programming (NDP, REPA etc) * Experience in the area of middleware algorithms for data flow programming, graphs, cloud based data analytics, or sparse matrices * Domain specific knowledge in scientific programming areas related to irregular and sparse problems, e.g. data analytics using graph analysis, genomics, tightly connected numerical analysis * Experience with performance tuning for parallel applications on multi core systems (e.g. with the Intel toolkit), for GPGPU's using Cuda/OpenCL or with MPI/OpenMP on clusters of SMPs To apply, please send a resume to j...@parsci.commailto:j...@parsci.com. ___ Haskell mailing list hask...@haskell.org http://www.haskell.org/mailman/listinfo/haskell ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
