Haskell Support on Windows (Simon Peyton Jones)

Simon Peyton Jones simonpj at microsoft.com
Tue Apr 1 12:57:41 UTC 2014

A couple of poor assumptions are made here.

I'm not sure where "here" is.   In my email I specifically said exactly what you are saying (only I was not as eloquent as you).

What we need is some developers who are willing to give some love and support to the Windows version of GHC.  And they really are thin on the ground, unfortunately.


From: ghc-devs [mailto:ghc-devs-bounces at haskell.org] On Behalf Of Kyle Van Berendonck
Sent: 01 April 2014 13:54
To: ghc-devs at haskell.org
Subject: Re: Haskell Support on Windows (Simon Peyton Jones)

A couple of poor assumptions are made here.

The first is that the userbase of GHC on Windows is poor. This is false. In fact, the poor state of Windows is mentioned on #haskell more frequently as of late. In the last week I've talked almost every day to (different) people who have had Windows woes with using GHC. The hacker-base isn't here at the moment - I agree, but the user-base most certainly is. Could this be a problem with the demographic, or could it just be that Windows development isn't inviting? The build system and default test failures in-particular are a source of discouragement.
The second assumption is that GHC/Haskell will be worth a dime in the RealWorld without Windows as a primary tier platform. I'm going to put a crude guess that given the overwhelming magnitude of C#/F# coming up on my local careers portal that there's far more industry on Windows than on OSX or Linux. Those are all the businesses where the probability that they will ever touch Haskell goes from `probably not right now` to `impossible`. Let me also remind you that Windows still holds 89.2% of operating system market share ie the people that hackers and developers actually have to deploy their applications to.
Sorry to sound fumey, but there's all this suggesting that everyone would have a better day if we dropped Windows (let's be honest, if it wasn't a primary tier platform nobody would have fixed it for 7.8), but I doubt few of the people who think it's a "great idea" or sslt have actually thought through whether it's the best thing for Haskell or just the best thing to get 7.8 into their hands a couple weeks sooner.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.haskell.org/pipermail/ghc-devs/attachments/20140401/dcd50cde/attachment-0001.html>

More information about the ghc-devs mailing list