- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] Increasing Haskell modularity
by Gershom B 01 Oct '14

01 Oct '14

It seems to me the far easier choice, in ditching the feature “not worth the problem” is simply to eliminate orphan instances as in “Extension 1” described in Edward Yang’s blog post (http://blog.ezyang.com/2014/09/open-type-families-are-not-modular/) I’d like to open up a further discussion on this at some future point, as I’m convinced that there is no “legitimate” use of orphans in the Haskell ecosystem that cannot be substituted for with other safer and more straightforward techniques. -g On October 1, 2014 at 9:26:45 AM, Gesh (gesh(a)gesh.uni.cx) wrote: > Dear Café, > Edward Yang recently wrote about the non-modularity of the current semantics > for typeclasses[0] and open type families[1]. > Given these problems, we can either try to work around the problem, or > decide > whether these features are worth the problems. > The specific pair of proposals I am making are the following: > * Give up on global uniqueness of typeclass instances. > Oleg discusses adding local typclass instances without losing > coherence in his > reflection paper[2]. If we adopt his proposal, and weaken global > uniqueness of > instances to local uniqueness, we can then say that typeclass > instances in a > module are considered more local than imported instances, and therefore > override them. > Since instances will be locally unique, we don't have to pass around > dictionaries. Data types that expect globally unique instances, such > as Set, > will have to store dictionaries, however. > * Give up on non-associated open type families. > This I have researched less well, but it seems to me that these have > no use > case. Since one can make whatever instances one wishes for these type > families, one can't do anything useful with values of their instances. > In other words, values of type F a are morally of type (forall a. a). > Thus, the only way I can see these being useful is if anyone creating an > instance of the family must also provide functions that can work with > these > instances, which brings us back to the case of associated types. > > As far as I can see, these two features are useless and impeding the > drive to > make Haskell more modular. Therefore, my questions for you are twofold: > * Is there a reason to keep these features? > * If not, why don't we drop them? > > Awaiting your criticism and ire, > Gesh > > [0] - > http://blog.ezyang.com/2014/07/type-classes-confluence-coherence-global-uni… > [1] - http://blog.ezyang.com/2014/09/open-type-families-are-not-modular/ > [2] - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.112.7819 > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe >

3 2

Increasing Haskell modularity
by Gesh 01 Oct '14

01 Oct '14

Dear Café, Edward Yang recently wrote about the non-modularity of the current semantics for typeclasses[0] and open type families[1]. Given these problems, we can either try to work around the problem, or decide whether these features are worth the problems. The specific pair of proposals I am making are the following: * Give up on global uniqueness of typeclass instances. Oleg discusses adding local typclass instances without losing coherence in his reflection paper[2]. If we adopt his proposal, and weaken global uniqueness of instances to local uniqueness, we can then say that typeclass instances in a module are considered more local than imported instances, and therefore override them. Since instances will be locally unique, we don't have to pass around dictionaries. Data types that expect globally unique instances, such as Set, will have to store dictionaries, however. * Give up on non-associated open type families. This I have researched less well, but it seems to me that these have no use case. Since one can make whatever instances one wishes for these type families, one can't do anything useful with values of their instances. In other words, values of type F a are morally of type (forall a. a). Thus, the only way I can see these being useful is if anyone creating an instance of the family must also provide functions that can work with these instances, which brings us back to the case of associated types. As far as I can see, these two features are useless and impeding the drive to make Haskell more modular. Therefore, my questions for you are twofold: * Is there a reason to keep these features? * If not, why don't we drop them? Awaiting your criticism and ire, Gesh [0] - http://blog.ezyang.com/2014/07/type-classes-confluence-coherence-global-uni… [1] - http://blog.ezyang.com/2014/09/open-type-families-are-not-modular/ [2] - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.112.7819

2 1

Re: [Haskell-cafe] ANN: Nomyx 0.7 beta, the only game where you can change the rules
by Corentin Dupont 01 Oct '14

01 Oct '14

Hi again, we started the game, please join, don't be shy :) Some interresting rules have already been proposed by players: - we now have a list of friends. - we have bank accounts in a currency called the "Bling". - we'll soon have planets, and maybe spaceships. A lot of things can happen now (what friends allow to do? How to win money? how to trade with planets? etc.). www.nomyx.net Cheers! C On Tue, Sep 30, 2014 at 12:31 PM, Corentin Dupont <corentin.dupont(a)gmail.com > wrote: > > Hello everybody! > I released Nomyx <http://www.nomyx.net> V0.7, the only game where You can > change the rules! > Here is the website: www.nomyx.net > Here is a video introduction of the game: http://vimeo.com/58265498 > I created a tutorial to learn how to play: > http://www.corentindupont.info/blog/posts/2014-09-23-first-Nomyx-tutorial.h… > > Let's start a new game! I propose to call it "The Space Merchants". Dear > rulers, this part of the Galaxy is yours... > Please login to the game here: www.nomyx.net:8000/Nomyx. > The corresponding forum thread is here: > http://www.nomyx.net/forum/viewtopic.php?f=4&t=1525&p=1739 > Please also register on the mailing list to follow the game: > nomyx-game(a)googlegroups.com > > Some background: this is an implementation of a Nomic [1] game in Haskell > (I believe the first complete implementation of a Nomic game on a > computer). In a Nomyx game you can change the rules of the game itself > while playing it. The players can submit new rules or modify existing ones, > thus completely changing the behaviour of the game through time. The rules > are managed and interpreted by the computer. They must be written in the > Nomyx language <http://hackage.haskell.org/package/Nomyx-Language>, which > is a subset of Haskell. > > At the beginning, the initial rules are describing: > - how to add new rules and change existing ones. For example a unanimity > vote is necessary to have a new rule accepted. > - how to win the game. For example you win the game if you have 5 rules > accepted. > But of course even that can be changed! > > New in this beta release: > - the Nomyx language is now based on functional reactive programming > - the language also embeds a mechanism to separate effectful instructions > from effectless ones > - ability to fork an existing game into a new one > - ability to type-check rules before submitting them > - ability to create new games > > If you wanna help developing Nomyx, join us on the dev mailing-list: > dev-game(a)googlegroups.com <nomyx-game(a)googlegroups.com> > > > Cheers, > Corentin > > [1] www.nomic.net > >

1 0

context deduction in GHC
by Doug McIlroy 30 Sep '14

30 Sep '14

Why does ghci 7.6.3 not like this example, which section 4.3.2 of the language definition says is valid Haskell? class Foo a class Foo a => Bar a instance (Eq a, Show a) => Foo [a] instance Num a => Bar [a] Ghci complains, "Could not deduce (Show a) arising from the superclasses of an instance declaration from the context (Num a)". Yet Show is a superclass of Num. Hugs accepts the code. Doug McIlroy

3 2

ANN: Nomyx 0.7 beta, the only game where you can change the rules
by Corentin Dupont 30 Sep '14

30 Sep '14

Hello everybody! I released Nomyx <http://www.nomyx.net> V0.7, the only game where You can change the rules! Here is the website: www.nomyx.net Here is a video introduction of the game: http://vimeo.com/58265498 I created a tutorial to learn how to play: http://www.corentindupont.info/blog/posts/2014-09-23-first-Nomyx-tutorial.h… Let's start a new game! I propose to call it "The Space Merchants". Dear rulers, this part of the Galaxy is yours... Please login to the game here: www.nomyx.net:8000/Nomyx. The corresponding forum thread is here: http://www.nomyx.net/forum/viewtopic.php?f=4&t=1525&p=1739 Please also register on the mailing list to follow the game: nomyx-game(a)googlegroups.com Some background: this is an implementation of a Nomic [1] game in Haskell (I believe the first complete implementation of a Nomic game on a computer). In a Nomyx game you can change the rules of the game itself while playing it. The players can submit new rules or modify existing ones, thus completely changing the behaviour of the game through time. The rules are managed and interpreted by the computer. They must be written in the Nomyx language <http://hackage.haskell.org/package/Nomyx-Language>, which is a subset of Haskell. At the beginning, the initial rules are describing: - how to add new rules and change existing ones. For example a unanimity vote is necessary to have a new rule accepted. - how to win the game. For example you win the game if you have 5 rules accepted. But of course even that can be changed! New in this beta release: - the Nomyx language is now based on functional reactive programming - the language also embeds a mechanism to separate effectful instructions from effectless ones - ability to fork an existing game into a new one - ability to type-check rules before submitting them - ability to create new games If you wanna help developing Nomyx, join us on the dev mailing-list: dev-game(a)googlegroups.com <nomyx-game(a)googlegroups.com> Cheers, Corentin [1] www.nomic.net

2 1

Cabal Testing Pain
by Mike Craig 30 Sep '14

30 Sep '14

Hey all, I'm trying to setup an executable cabal project, the options for cabal-based testing seem poor. I poked around for discussions of this but didn't find much. My apologies if I'm beating a dead horse. The current options, as I understand them, are 1. Split code into library, executable, and test-suite. This gets you single compilation but you have to list every source file in the library's exposed/other-modules sections. 2. Split code into executable and test-suite. You don't have to list the source files, but you get double compilation and you have to declare every dependency (and version!) twice. Am I missing a third (better) option, or is this the state of things? -- Mike Craig

5 4

the complex package is hidden
by briand＠aracnet.com 28 Sep '14

28 Sep '14

Hi, surprisingly i can't seem to google an answer to this: Could not find module `Complex' It is a member of the hidden package `haskell98-2.0.0.2'. Use -v to see a list of the files searched for. it's been quite a while since I've used ghci, and I'm just about positive I've solved this problem before, but can't seem to figure it out. Thanks for any help.

2 8

Deploying bash completion with cabal?
by Jacob Stanley 28 Sep '14

28 Sep '14

Hi All I'm looking for a good way to deploy a bash completion script somewhere appropriate for the given operating system. Has anyone tackled this before? I was considering packaging the script in to the install using cabal data files, then having an 'install--bash-completion' option that tries to detect the right place to install it depending on the OS. Cheers Jacob

2 1

Separate compilation
by Lars Hupel 28 Sep '14

28 Sep '14

Hi all, I'm trying to compile a couple of file separately. The main purpose for that is that we're checking student-submitted exercises, and we want to compile them with -XSafe. The test suite doesn't require -XSafe. There are basically three files: * Exercise.hs -- uploaded by the student * Exercise_Interface.hs -- imports the 'Exercise' module and creates new bindings with appropriate types * Exercise_Test.hs -- imports the interface module and contains the test code and 'main' Here's how I try to compile these files: TEMP="$(mktemp -d -p /tmp ghcXXX)" ghc -c -XSafe -outputdir "$TEMP" Exercise.hs ghc -c -outputdir "$TEMP" -i"$TEMP" Exercise_Interface.hs ghc -o "$TEMP/runner" -outputdir "$TEMP" -i"$TEMP" Exercise_Test.hs This works except for the last step: GHC complains that it can't find the module 'Exercise_Interface'. Upon closer inspection with '-v', it turns out that it's actually looking for the .hs file, which I find odd. If I add '-c' to the last invocation, it works, but I don't get an executable. What am I doing wrong? Cheers Lars

2 2

Arrow key functionality for REPL written in Haskell
by PENG, BO YA 28 Sep '14

28 Sep '14

Hi, I'm trying to implement the arrow key functionality for a systemf REPL to recall previously entered lines. I think I should save the previous lines in a list and then traverse through the list when an arrow key is pressed, but I'm no sure how to detect an arrow key and implement this functionality. Can anyone give a brief description about the implementation or point me to some other resources? Thanks! Emma

4 3