Re: positive definite matrix predicate

Tim Daly <[email protected]> Mon, 28 Dec 2009 17:52:53 -0500
Newsgroups gmane.comp.mathematics.axiom.user
Message-ID <[email protected]>
Konstantin L. Metlov wrote:
>> It is consistent with the former. The domain Float does not include the
>> result %i so it is an error. The other domains include the result.
>>     
> Yes, but the Integer domain does not include complex unity either and in that 
> case the result is automatically converted to AlgebraicNumber, which can 
> represent it. Analogous behaviour in the first case could be to declare the 
> result of sqrt on Float as Complex(Float).
>
> Konstantin
>
>
> _______________________________________________
> Axiom-mail mailing list
> [email protected]
> http://lists.nongnu.org/mailman/listinfo/axiom-mail
>
>   
It could be argued either way. Using floats is generally done because you
want a numeric answer. A %i answer would by symbolic. I'd be willing to
accept either form but, at the moment, Axiom does not do the coercion for
floats automatically. You can get the behavior you want by an explicit call
(the so-called dollar call which specifies which domain to choose):

sqrt(-1.0)$Complex(Float) ==> %i

Tim