[Haskell-cafe] Haskell Propeganda

[Brandon S. Allbery KF8NH]

On 2008 Aug 24, at 4:00, Thomas Davie wrote:
> On 24 Aug 2008, at 05:04, Albert Y. C. Lai wrote:
>
>> "Dear friends, Haskell prevents more errors and earlier." This is  
>> honest, relevant, good advocacy.
>>
>> "Dear friends, segfaults are type errors, not logical errors." Why  
>> would you indulge in this? It's even less relevant than bikeshed  
>> colours.
>
> Is it?  when I write C I spend a lot of my time sat in gdb trying to  
> figure out where the error that the Haskell type system would have  
> caught for me is.  This is *very* relevant, it's right at the bottom  
> line of whether I'm more productive in Haskell or in C.


Half agreed.  It occurs to me this morning that one thing to consider  
is that often segfaults are more the kind of error that requires  
dependent types, which are Hard in Haskell (see, for example, the  
discussion of type-level naturals elsethread).  (Example:  copying a  
buffer when the copy length is negative due to a bug.)

-- 
brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery at kf8nh.com
system administrator [openafs,heimdal,too many hats] allbery at ece.cmu.edu
electrical and computer engineering, carnegie mellon university    KF8NH