Re: [Swarm-Modelling] Floating point arithmetic

[Marcus G. Daniels]

Thanks for posting this, Gary.   I remember seeing a document like this 
several years ago, but it appears to be evolving.. ;-)

> For Java to be taken seriously as a language for scientific
> programming, it  needs to treat floating point arithmetic with the respect it 
> deserves
Sun is working on a new language for scientific computing.   I think it 
sounds pretty interesting.  I guess their answer to Java's criticisms 
boils down to "you are using the wrong language".   Check it out:

http://www.sun.com/presents/minds/2005-0302
http://lambda-the-ultimate.org/node/view/673

Also of interest is this language from Cray and the University of 
Washington:

http://chapel.cs.washington.edu

Both languages have explicit support for parallelism, generics (a.k.a. 
templates), and many other things. 

Perhaps one of these languages would make a good foundation for a next 
generation agent simulation system?