Re: Functional Programming in the Larger or Functional Oriented Software Engineering

On 15 Mar 2007, at 05:49, Doug Landauer wrote:

> On Mar 14, 2007, at 6:45 PM, Vadim Nasardinov wrote:
>> Clearly, FP people will tend to pick something with a well-defined
>> simple semantics like Z over something with an ill-defined complex
>> semantics like UML.
>
> Meanwhile, the vast majority of programmers out there will
> continue to do their modeling on napkins, whiteboards [*], and
> (for formal occasions) in Powerpoint or Visio.

The fact that UML is ill-defined doesn't matter much. Its usefulness  
comes from making it possible to communicate rough sketches, not from  
detailing them. (I regard attempts to formally specify the  
"semantics" of UML as misguided.)

There is always a stage where it not clear what the software to be  
written should look like or should do in the first place. UML is one  
of the possible tools to get a better understanding, not more, not less.

Martin Fowler puts the "UML as sketch" idea very well. See for  
example http://www.martinfowler.com/articles/designDead.html#IDA2H4VB  
or http://www.martinfowler.com/bliki/UmlAsSketch.html


Pascal

-- 
Pascal Costanza, mailto:[email protected], http://p-cos.net
Vrije Universiteit Brussel, Programming Technology Lab
Pleinlaan 2, B-1050 Brussel, Belgium