Re: math to apache commons was Re: [all] Separate email list for math development?

[Mark R. Diggory]

I do agree with much of what you say, but (2) There is significant 
number of us with an interest in seeing numerically sound 
implementations of various aspects of mathematics in java (no matter how 
wonderfully fast F77 implementations may be, Fortran will never be Java, 
nor would one want it to be). I could retort many reasons why this is an 
interest to myself and others in the group. Suffice it to say that java 
implementations do provide an elegant means to explore ideal Design 
Patterns for Mathematical packages and provide a foundation for 
exploring how to "bridge" java with other languages that may be more 
optimized to handle such computations. These are noble goals.

-Mark

Matt Cliff wrote:

my 2cents

  (1) I think it would be best to get a 1.0 release of commons-math as a 
jakarta-commons subproject before any movement (the good-old KISS acronym 
comes to mind)

  (2) I am not sure to what extent math may get developed, as was stated 
in a different thread previously Java is not the best platform for 
Numerical Analysis.  That being said, I dont forsee commons-math as a 
serious Numerical Anaylsis lib, rather a convenient and useful tool to 
explore 'simple' problems.  This could be significant in research, in my 
opinion you dont really need to 'crank' up the system in most cases, and 
have a handy library that can get some nice results quickly in a Java 
setting is very valueable.  There are plenty of high-scale number 
crunchers, and most of the best performing ones will always be in F77 due 
to its static nature.  It may very well be that some of the value out of 
the math project is as an SPI to these backend 'crunchers' which provides 
a nice Java interface to interact with the rest of the world.  (leading me 
back to the beginning of this rambling...)

  (3) I am thoroughly confused as to the pros-cons of what moving this 
the apache-commons level would do.  I like the idea of a 100% java 
framework at this point (leading me back to point #1).

In summary, I think it should stay where it is for now.

--
Mark Diggory
Software Developer
Harvard MIT Data Center
http://www.hmdc.harvard.edu
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]