Re: By The Dawn's Normal Light

From: Dawn M. Wolthuis <dwolt_at_tincat-group.comREMOVE>
Date: Sun, 24 Oct 2004 11:54:57 -0500
Message-ID: <clgmpi$2b3$1_at_news.netins.net>

"Marshall Spight" <mspight_at_dnai.com> wrote in message news:RwGed.520800$8_6.383427_at_attbi_s04... > "Dawn M. Wolthuis" <dwolt_at_tincat-group.comREMOVE> wrote in message news:cleccb$q3d$1_at_news.netins.net...
> > "Marshall Spight" <mspight_at_dnai.com> wrote in message
> > news:tCxed.294876$MQ5.202563_at_attbi_s52...
> > > "Dawn M. Wolthuis" <dwolt_at_tincat-group.comREMOVE> wrote in message
> > news:cldvjl$hj9$1_at_news.netins.net...
> > > >
> > <snip> We should probably start talking about what the nested
> > > relational model would look like, and what operations it
> > > would support. I understand academia has been over
> > > this ground a lot....

> >

> > If we think of the type definition including operations and have an
> > extensible type system, then perhaps we could start with some generic
> > "thing" and call that type Object. Then we could extend that object,
add
> > some primitive types such as int, double, char, short, boolean, long,
and
> > float and then provide a standard library that users can extend as
needed.

>
> Actually, I'm not sure that it makes sense to mix subtyping into the
> conversation from first principles. I'm not clear that it's required,
> although in Java at least, subtyping is very useful. In fact, I'd say
> we have no immediate evidence that subtyping isn't superfluous
> in the face of a good nested relational system.
>
>

> > In fact, I think such systems are already out there. What would we lose
if
> > we used Java for defining types?

>
> The possibility of doing better?

Good answer. But I would also like to see the language of the database and the language of data that is not-persisted be the same; constraints used by the database and those used by the UI be the same; etc. Additionally, if we could pick an existing language that is widely used and available, such as Java which is promoted by both IBM and Sun, we could leap that hurdle faster.

I don't see Java as the ultimate computer language, but it is quite useful, with a good start at libraries, and it's portable. If it were at least one of the possible languages for encoding any metadata that is not handled as data and for defining types and constraints, that would make development all the easier, I suspect. Again, I'm not a Java-as-the-best-thing-since-sliced-bread person, but it has a good headstart on whatever we might come up with that we think is better. And, if we do think of something better, we might just be able to write java libraries that would give similar gains to the java developers.

Again, just typing perhaps without enough thinking, but it feels right to me to have at least a bit more consistency in our languages. That said, I'll go back to writing apps using java servlets that execute SQL commands along with jsp that includes html, not to mention the taglibs. Maybe I'm just looking to return to the easy COBOL environments of old where we had to know, well, COBOL and maybe CICS plus IMS or VSAM to get the job done, plus SPF and JCL -- OK, nevermind.
smiles. --dawn Received on Sun Oct 24 2004 - 18:54:57 CEST