[om] Specfun CD's (was: library)

[J H Davenport]

On Fri, 10 Aug 2001, Andrew Solomon wrote:
> I found p.7 of James' paper a bit disturbing.
Well, I wouldn't be doing my job if nothing was disturbing but ...

> I really think that core OM CDs should stick to expressions, not commands.
But what is an expression.
> Therefore "factor" should be a predicate factor(a,b) which expresses
> that a is a factor of b, not a command "go factor this polynomial".
There is a 'factorof' symbol,
> One could also have factors(a) which refers to the list/set of
> factors of a polynomial a.
OK - I called it 'factor' rather than 'factors' - the name could certainly
be changed. What about the example I quote Fateman:
!> In a mathematics paper, one could well write (using <> to denote
!> OpenMath objects in text)
!> <factor(x^2+1)=factored(x^2+1)> over <Z>, but over <Z[i]>,
!> <factor(x^2+1)=factored(x+i,x-i)>

> I don't think that actions/behaviours should be encoded in these
I don;t think they are actions, though CA systems may interpret them as
such: see the exampe above.
> "standard" CDs.  Such specifications are only relevant to a particular
> application of OM -  computer algebra.   Behaviours could be encoded
> in special "computer algebra" CDs or even in some other format
> like an LDAP style directory for describing computer algebra services and
> package capabilities.
Yes, but one wold still need a way, IN OPENMATH, of saying that this LDAP
entry produced a factorisation, so one needs to be able to talk about
factorisation, which is what the OpenMath 'factor' (which has been o poly
CDs of one form or another) for over four years is intended to express.
I'd be happy to hear of a better wording, though.

James

--
om at openmath.org  -  general discussion on OpenMath
Post public announcements to om-announce at openmath.org
Automatic list maintenance software at majordomo at openmath.org
Mail om-owner at openmath.org for assistance with any problems