Re: Dreaming About Redesigning SQL

"Alfredo Novoa" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> "Bob Badour" <[EMAIL PROTECTED]> wrote in message
news:<[EMAIL PROTECTED]>...
> > "Alfredo Novoa" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]
> > > "Bob Badour" <[EMAIL PROTECTED]> wrote in message
> >  news:<[EMAIL PROTECTED]>...
> > >
> > > > > But decent database languages are not imperative. Python is a very
> > > > > inappropriate language for database management.
> > > >
> > > > Alfredo, D is imperative.
> > >
> > > Well, the declarativeness of a language is relative.
> > >
> > > Tutorial D has an algebra and a calculus version. The calculus version
> > > is declarative and the algebra version is imperative, but both
> > > versions are very declarative compared to procedural languages.
> >
> > Both are imperative as they both involve assigning values to variables.
>
> If this is your criteria then both are imperative

This is the criterion for imperative languages. If you use other criteria,
you misspeak. You may verify this using any number of sources including the
ISO/IEC standard vocabularies.


>, but then ML is also
> an imperative language because it has an assignment operator.

Then ML is indeed imperative.


> BTW we could create a functional D without problem, thus D is not
> neccessarily imperative even if you use a purist criteria.

Alfredo,

Are you suggesting one can get rid of global relvars and still have a valid
D? I suggest one can make a non-procedural D by creating a functional D, but
I have difficulty understanding how one can make a non-imperative D while
conforming to the relational model.

Bob