Re: [ublas] No == operator?

From: Paul C. Leopardi (leopardi_at_[hidden])
Date: 2006-03-05 08:55:52

On Sunday 05 March 2006 20:29, Dima Sorkin wrote:
> Quoting "Paul C. Leopardi" :
> > GluCat ( http://glucat.sf.net ) defines operator== for multivectors.
> > For matrix_multi this is implemented by:
> > 1) check the matrices to see if either has zero entries
> > 2) if so, subtract and use norm_inf
> > 3) if no, iterate over both matrices in lock step, comparing entries.
> > Best regards, Paul Leopardi
>
> http://lists.boost.org/MailArchives/ublas/2005/05/0261.php

Hi Dima, yes I saw that. I know that uBLAS only approximately models linear 
algebras. 

GluCat has a different philosophy to that of uBLAS. It tries to model the 
Clifford algebras as closely as possible within the contraints of floating 
point artihmetic. As equality is a fundamental algebraic relationship, GluCat 
provides operator==, knowing that however it is implemented, it will not be 
perfect. 

As the French philosopher once said, "Il meglio è l'inimico del bene." 
http://en.wikiquote.org/wiki/Voltaire

Best regards, Paul Leopardi