Re: [Axiom-developer] Nag libraries

[Mike Dewar]

Personally I wouldn't use nagd - it dates from a time before dynamic
linking was a practical approach and when a typical user's machine was
not fast enough for solving real-life numerical problems.  When we
reelased Axiom under Windows it used dynamic linking to load NAG
routines into the running image, and we planned to use the same approach
on Unix for Axiom 3 (which never got beyond the design stage).

Mike.

On Sat, Dec 09, 2006 at 02:05:25AM +0100, Waldek Hebisch wrote:
> I recently tried removing NAG related domains. After removing about 100
> domains Axiom build fine with no change in test results.  This means that
> large part of Axiom remains completly untested.
> 
> Looking at Nag tarball I see that it contins a 'nagd' subdirectory
> which at first glance looks like sources of Nag demon.  If we have
> right to re-distibute it we could try to get it working and
> substitute some other numeriacal library in place of the Nag library.
> ATM I do not know how much work is needed to get 'nagd' and 'naglink'
> running but hopefully this may be easy.
> 
> -- 
>                               Waldek Hebisch
> address@hidden 
> 
> 
> _______________________________________________
> Axiom-developer mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/axiom-developer
> 
> ________________________________________________________________________
> This e-mail has been scanned for all viruses by Star. The
> service is powered by MessageLabs. For more information on a proactive
> anti-virus service working around the clock, around the globe, visit:
> http://www.star.net.uk
> ________________________________________________________________________