Re: General semantics

From: Erwin <e.smout_at_myonline.be>
Date: Fri, 21 May 2010 13:35:17 -0700 (PDT)
Message-ID: <79a73036-d7bb-42fa-bbca-92f2ebbbd220_at_j9g2000vbp.googlegroups.com>



On 21 mei, 17:58, Clifford Heath <n..._at_spam.please.net> wrote:
> paul c wrote:
> > Nilone wrote:
> >> As such, I see no real argument between
> >> the models; in fact, the little I know of ORM (mostly from looking at
> >> Halpin's page a while back) seemed to match up with the relational
> >> model well.
>
> Indeed. It has the potential to help people understand
> the relational model better,

Nope. it can be useful to help people understand information schemas better. While possibly valuable, that's a totally different thing.

> and to implement it more purely than has yet been done.

Nope. Because it does not seem to contain anything similar to or capable of replacing the relational algebra.

> > Tell us why we need more 'correspondences'.
>
> Because it does contain additional insight, for those
> who are open to it. Few here though, it seems. They
> refer to parade their partial knowledge instead.

The discussion about "correspondences" was about correspondences between RM and OO objects, not between RM and ORM objects.

> But if you're not in that class, I dare you to at least
> take a glance at some of the example models in ORM2 and
> CQL on this page and honestly tell me you think there's
> nothing interesting happening there:
> <http://dataconstellation.com/ActiveFacts/examples/>
> This stuff is pure relational, but at a level that anyone
> can understand... and that makes a difference.

I'll take a look and may respond later. Received on Fri May 21 2010 - 15:35:17 CDT

Original text of this message