Re: [Math-atlas-devel] multiple platform management

[Ed A. S. <ea...@be...>]

In message <110...@fr...> (on 1 December
2004 13:57:04 -0800), ste...@cs... (Stephen Walton) wrote:
>I am making a comment to one sentence of a long message by Dr. Whaley,
>most of which I have no quibble with.  As long as that's clear!
>
>On Wed, 2004-12-01 at 15:02 -0500, R Clint Whaley wrote:
>
>> I've often wondered why people are so wild about dynamic linking.
>
>It is a matter of platform management.  I know from the other posts that
>I'm not the only one running a shop which has a central file server on
>which users' home directories are stored plus a bunch of workstations,
>not all of which are the same architecture.

Same here, plus in ou case, ATLAS will be being mainly used for interfacing
with Perl's PDL (note that Perl is used widely in bioinformatics, plus PDL
was originally created by astronomers for handling massive datasets), and
statically linking ATLAS libraries into Perl is _not_ an option,
realistically speaking; it needs to be dynamically loadable.

	      -Allen

-- 
Allen Smith			http://cesario.rutgers.edu/easmith/
September 11, 2001		A Day That Shall Live In Infamy II
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety." - Benjamin Franklin