-1 on both. Leaving them out might encourage people to use "pure foo" or the idiom QuasiQuoter. It's not obvious [to me at least] which of the 3 styles is best, so I'd prefer to keep the additional syntax out of the Prelude. -Greg On Tue, Feb 24, 2015 at 7:43 AM, Alec wrote:

+1 on both

On Tue Feb 24 2015 at 10:42:41 AM Michael Snoyman wrote:

...

+1 on <$> +1 on <$

On Tue Feb 24 2015 at 5:39:56 PM Edward Kmett wrote:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have found some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

A hearty +1 for both. John

It is clearly sad that we reached that stage before realizing the problem, but OTOH it would be worst to not fix it now that we still have a (very) last chance. +1 on both from me. Unlike Greg I don't think encouraging people to use the QuasiQuoter is a good idea, it's for me obvious that the <$> syntax is widely used, I see it in mostly every open source code I read. On 24 February 2015 at 17:03, Richard Eisenberg wrote:

This is a tough call for me.

This change will be very annoying, I think. When updating for 7.10, I was quite surprised that I still needed to import Control.Applicative for <$>, and put in the CPP to do so. Now I'll have to remove that from quite a few modules. Given that we plan only a week for RC3, there are a lot of modules in the ecosystem that will have to be updated within that week!

Of course, an author who doesn't update (by removing the now-redundant import of <$>) just gets a warning, but it's still annoying.

In my opinion, little changes like this, made right before the deadline, make Haskell feel more like a little research language than something meant to support industrial work.

All that said, I desperately miss having <$> in the Prelude in 7.10. Hence the tough call.

0 from me, then.

Richard

On Feb 24, 2015, at 10:47 AM, Christopher Reichert wrote:

...

+1 on <$>. Indifferent on <$.

-Christopher

On Tue, Feb 24 2015, Edward Kmett wrote:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have

found

...

...

some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

-- Christopher Reichert irc: creichert gpg: C81D 18C8 862A 3618 1376 FFA5 6BFC A992 9955 929B _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

-- *Λ\ois* http://twitter.com/aloiscochard http://github.com/aloiscochard

This one draws me out of silence. Strong +1 on (<$>). 0 on (<$). — Sent from Mailbox On Tue, Feb 24, 2015 at 11:08 AM, Alois Cochard wrote:

It is clearly sad that we reached that stage before realizing the problem, but OTOH it would be worst to not fix it now that we still have a (very) last chance. +1 on both from me. Unlike Greg I don't think encouraging people to use the QuasiQuoter is a good idea, it's for me obvious that the <$> syntax is widely used, I see it in mostly every open source code I read. On 24 February 2015 at 17:03, Richard Eisenberg wrote:

...

This is a tough call for me.

This change will be very annoying, I think. When updating for 7.10, I was quite surprised that I still needed to import Control.Applicative for <$>, and put in the CPP to do so. Now I'll have to remove that from quite a few modules. Given that we plan only a week for RC3, there are a lot of modules in the ecosystem that will have to be updated within that week!

Of course, an author who doesn't update (by removing the now-redundant import of <$>) just gets a warning, but it's still annoying.

In my opinion, little changes like this, made right before the deadline, make Haskell feel more like a little research language than something meant to support industrial work.

All that said, I desperately miss having <$> in the Prelude in 7.10. Hence the tough call.

0 from me, then.

Richard

On Feb 24, 2015, at 10:47 AM, Christopher Reichert wrote:

...

+1 on <$>. Indifferent on <$.

-Christopher

On Tue, Feb 24 2015, Edward Kmett wrote:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have

found

...

...

some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

-- Christopher Reichert irc: creichert gpg: C81D 18C8 862A 3618 1376 FFA5 6BFC A992 9955 929B _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

-- *Λ\ois* http://twitter.com/aloiscochard http://github.com/aloiscochard

+1 for me on both. Besides convenience, symmetry is important so we should export <$> if we export <*> and <$ if <*. On Feb 24, 2015 9:33 AM, "Oliver Charles" wrote:

If <*> is in Prelude (I'm unclear) then I'm +1 on getting <$> there too - <*> is practically useless without <$>!

<$ is handy and I do tend to use it, but I don't feel so strongly about having that in the Prelude (so I'm not going to vote for or against that).

- Ollie

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

+1 for <$>, 0 on <$, I rarely use it. On Tue, Feb 24, 2015 at 7:25 PM, Conor McBride wrote:

On 24 Feb 2015, at 15:39, Edward Kmett wrote:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have found some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

For more on this idiom (indeed), let me shamelessly plug

http://stackoverflow.com/a/15124278/828361

and advise that the two travel together.

I won’t vote. While we are still in the situation that progress breaks code which has been engineered to work around the prior lack of progress, making progress is simultaneously a good and bad idea. We should think about how to adapt our language of imports to accept the only too obvious reality that libraries change over time, so that we might be allowed to program defensively against sudden outbreaks of common sense.

Apologetically

Conor

...

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1]

http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/

...

[2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

+1 to <$>. To me, Applicative means pure/<$>/<*>. Having 2/3rds of that in the Prelude seems weird. If we hadn't put things like *> in the Prelude, then I'd vote -1 for <$. But given the Prelude operator creep has begun, I can see why <$ lives on the other side of the line, given if I'm not particularly in favor of where the line lies. Thanks, Neil On Tue, Feb 24, 2015 at 7:06 PM, Milan Straka wrote:

Hi all,

+1 on both <$> and <$ .

Cheers, Milan Straka

...

-----Original message----- From: Edward Kmett Sent: 24 Feb 2015, 10:39

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have found some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp

...

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Just to show my ignorance about the internals of the GHC development process. Suppose I might want to take a look at the new Prelude in order to form an opinion and probably express it, where can I find it? I see a lot of talking about it, but I do not see a link anywhere. Doaitse

On 25 Feb 2015, at 11:22 , Phil Ruffwind wrote:

(+2) on both _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

Hi Doaitse, Here you go: https://github.com/ghc/ghc/blob/master/libraries/base/Prelude.hs On 25/02/15 17:29, S. Doaitse Swierstra wrote:

Just to show my ignorance about the internals of the GHC development process.

Suppose I might want to take a look at the new Prelude in order to form an opinion and probably express it, where can I find it? I see a lot of talking about it, but I do not see a link anywhere.

Doaitse

Hi, Am Mittwoch, den 25.02.2015, 16:29 +0100 schrieb S. Doaitse Swierstra:

Just to show my ignorance about the internals of the GHC development process.

Suppose I might want to take a look at the new Prelude in order to form an opinion and probably express it, where can I find it? I see a lot of talking about it, but I do not see a link anywhere.

if you prefer haddock, see http://haskell.inf.elte.hu/docs/latest/html/libraries/base-4.8.0.0/Prelude.h... for the state of GHC HEAD (which might deviate from 7.10 though) Greetings, Joachim -- Joachim “nomeata” Breitner mail@joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata@joachim-breitner.de • GPG-Key: 0xF0FBF51F Debian Developer: nomeata@debian.org

On 25/02/15 11:46 AM, Conal Elliott wrote:

(<$>): +1 (<$): +1

Same here.

Those operators are for Power Users - why put them in the Prelude? I mean, they are so money, I can't find words to describe them. Below are a few attempts. I start with a description in English, then implement using Functor's bling operators, and then again using the Prelude as-is. In each case, the code is shorter with the bling operators, but if you try to verbalize it, it's nonsense and the syntax is ungoogleable. import Data.Functor((<$>), (<$), ($>)) -- Map (+1) on a computation moneyCentric, moneyCentric2 :: IO Int moneyCentric = (+1) <$> return 1 moneyCentric2 = fmap (+1) $ return 1 -- Map (const 2) on a computation lessMoney, lessMoney2 :: IO Int lessMoney = 2 <$ return 1 lessMoney2 = fmap (const 2) $ return 1 -- Run a computation, but then return 2. moneyGreaterThan, moneyGreaterThan2 :: IO Int moneyGreaterThan = return 1 $> 2 moneyGreaterThan2 = return 1 `butReturn` 2 where m `butReturn` x = m >> return x And check this out: Prelude> 2 <$ return 1 <interactive>:2:3: Not in scope: ‘<$’ Perhaps you meant one of these: ‘<’ (imported from Prelude), ‘<=’ (imported from Prelude), ‘$’ (imported from Prelude) The syntax '<$' is close enough to '<=' and '$' that ghci guesses it was a typo. Shouldn't we assume this syntax would make a beginner's head spin? If you really want to use those operators, import Data.Functor. Why must it be in the Prelude? -Greg

Dne út 24. 2. 2015 16:39 uživatel Edward Kmett napsal:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have found some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

They definitely should be in both, since the default implementations of <$> and <$ f <$> x = pure f <*> x f <$ x = const f <*> x may be less efficient than a special implementation having more knowledge about applicative. Doaitse PS: for that reason they are for almost 20 years now member of the class IsParser in http://hackage.haskell.org/package/uulib-0.9.16/docs/UU-Parsing-Interface.ht... http://hackage.haskell.org/package/uulib-0.9.16/docs/UU-Parsing-Interface.ht... , which inspired the introduction of Applicative and Alternative in the first place.

On 27 Feb 2015, at 5:45 , Edward Kmett wrote:

On Thu, Feb 26, 2015 at 6:37 PM, Greg Fitzgerald mailto:garious@gmail.com> wrote: Those operators are for Power Users - why put them in the Prelude?

.. because without them that whole Applicative thing that we're bringing in as a superclass of Monad with the AMP is cut off at the knees.

Applicative sugar is typically used like

both f (x,y) = (,) <$> x <*> y

or

traverse f (x:xs) = (:) <$> f x <*> traverse f xs traverse f [] = pure []

Without (<$>), (<*>) from Applicative is quite difficult to use.

both f (x,y) = pure (,) <*> x <*> y

is much less efficient and

both f (x,y) = fmap (,) x <*> y

is painfully far from idiomatic.

-Edward

-Greg

...

Dne út 24. 2. 2015 16:39 uživatel Edward Kmett mailto:ekmett@gmail.com> napsal:

...

We have a couple of weeks until the third release candidate for GHC 7.10 goes out the door.

Along the way with the last couple of release candidates folks have found some problems with the way we implemented the AMP. [1][2]

Most notably, we failed to include (<$>) in the Prelude, so the standard idiom of

foo <$> bar <*> baz <*> quux

doesn't work out of the box!

I'd like to include (<$>) in the Prelude in RC3.

I'd also like to invite discussion about whether folks believe we should include (<$) out of the box.

(<$) has been a member of Functor for a long time, which is only visible if you import it from Data.Functor or bring in Control.Applicative. There is an idiom that you use (<*) and (<$) to point to the parts of the structure that you want to keep the answers from when building longer such Applicative chains.

Discussion Period: 2 weeks

Thank you, -Edward Kmett

[1] http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ http://www.reddit.com/r/haskell/comments/2wzixa/shouldnt_be_in_prelude/ [2] https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp https://plus.google.com/115504368969270249241/posts/URzeDWd7qMp _______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org mailto:Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries

On Thu, 26 Feb 2015, Edward Kmett wrote:

both f (x,y) = pure (,) <*> x <*> y

I find this variant more natural, because every operand is preceded by a (<*>). This way it is easier to reorder arguments in an editor. It is related to the common separator vs. terminator debate.

That is not and was never the motivation of FTP. FTP was about not wanting to have to hide a zillion functions from Prelude, Data.List, Control.Applicative, and Control.Monad. Functors are utterly ubiquitous in Haskell programming, so adding an operator to make them easier to use seems quite sensible. On Feb 27, 2015 12:19 PM, "Henning Thielemann" < lemming@henning-thielemann.de> wrote:

On Thu, 26 Feb 2015, Greg Fitzgerald wrote:

If you really want to use those operators, import Data.Functor. Why

...

must it be in the Prelude?

It seems that many Haskell programmers don't like to write "import". That's the whole motivation for FTP. I can't understand it, too. _______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries