On 4 February 2015 at 10:13, Ralf Hemmecke <[email protected]> wrote:
> On 02/04/2015 03:52 PM, Bill Page wrote:
>
> Whatever can be done with something of type Symbol depends on
> the library functions that have Symbol in there argument types.
>
Yes, that is trivially true. I was thinking more about the
connotation of the use of the name "symbol".
> The interpreter (in my eyes) is just a shell around the library
No, I do not agree that the FriCAS interpreter is just a shell. It
obviously tries to be much more than that. Perhaps what you are
saying is that in your opinion it should be just a shell?
> and blurs types and even the distinction between symbols and
> identifiers. But all CAS I know do this.
Yes, that is really my point. Compared to some other CAS, FriCAS does
not go far enough in blurring the distinction. In spite of Waldek's
sceptism, I do think it would be useful in the interpreter to allow
scripted identifiers and relate them implicitly to scripted symbols.
I do not see any problem with runtime resolution of names in the
interpreter but of course the SPAD language is a separate issue.
>
> If you restrict to library programming (i.e. just writing .spad
> programs) I don't see this confusion between identifier and Symbol.
>
Agreed.
>> Somewhere between Symbol and identifiers we also have something
>> called Variable.
>
> Well, Variable('x) ...
> If you grep the sources, you'll find that its main use is to give 'x a
> domain where only that symbol lives and nothing else.
Why do we need that?
> If all our polynomials and series were without variable names
> (i.e. something like SparseUnivariatePolynomial), then it would
> probably make sense to think about whether Variable as a type is
> really needed.
>
It is technically possible to do without symbols but in most
mathematical notations they are essential.
Still I am not entirely convinced that the domain Variable is really
needed right now.
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
