[Haskell-cafe] Re: Hoogle and Network.Socket

[John A. De Goes]

I don't think it's that black and white.

At the lower end, when the language is controlled by a few, there's  
not much innovation poured into the language or libraries, and there  
are no tools to support development. As the community grows, you see  
much more innovation in language and libraries, and maybe a few  
primitive tools. With much greater, the community demands backward  
compatibility, so the language itself may only evolve in highly  
constrained ways (ways that are usually detrimental to consistency),  
but the library space explodes with innovation, and the tools become  
extremely powerful.

Personally, I'd be happy to see that explosion of innovation in the  
library and tool spaces, even if it means the language itself stops  
evolving (for the most part). It will make it a lot easier do use  
Haskell commercially, and the innovators in the language space will  
find or invent a new target to keep themselves occupied.

Regards,

John A. De Goes
N-BRAIN, Inc.
The Evolution of Collaboration

http://www.n-brain.net    |    877-376-2724 x 101

On Feb 25, 2009, at 5:52 PM, Jonathan Cast wrote:

> On Wed, 2009-02-25 at 17:54 -0700, John A. De Goes wrote:
>> It's a chicken-egg thing. A Linux or OS X developer tries Haskell and
>> finds he can write useful programs right away, with a minimum of  
>> fuss.
>> But a Windows user tries Haskell and finds he has access to very few
>> of the really good libraries, and even the cross-platform libraries
>> won't build without substantial effort. As a result, I bet it's  
>> easier
>> for a Linux or OS X developer to like Haskell than a Windows  
>> developer.
>>
>> I use OS X exclusively myself, but I'll ensure my first published
>> Haskell library is cross-platform compatible, because I think it's
>> good for the community. The more people using Haskell, the more
>> libraries that will be written, the more bugs that will be fixed, the
>> more creativity that will be poured into development of libraries and
>> the language itself.
>
> I don't think this is founded in experience.  The experience of the  
> last
> 5 years is that the more people use Haskell, the more important
> backward-compatibility concerns become, and the harder it becomes for
> Haskell to continue evolving.
>
> Creativity being poured into a language doesn't do much good if the
> result is the language moving sideways, still less the language  
> growing
> sideways.
>
> jcc
>
>