Re: foundations of relational theory?

"Bob Badour" <bbadour_at_golden.net> wrote in message news:HPudnfvbBOkx4weiU-KYhw_at_golden.net...
> "Marshall Spight" <mspight_at_dnai.com> wrote in message
> news:MDomb.23527$HS4.91636_at_attbi_s01...
>
> > One thing this thread has gotten me thinking about is the agility
> question.
>
> Agility is just another word for logical independence. Pick lacks it.

Well, okay. But do existing DBMSs have complete logical independence? I think they don't. Views aren't as easy to work with as they need to be, for example.

> [lots of "you can use" statements snipped]
>
> > Relations look like the clear winner to me.
>
> One might use some as yet unidentified structure. Dawn wants to contribute
> something as important as any contribution by Aristotle, Boole or Goedel.
> She is welcome to try.

Indeed. I don't expect her, or anyone else for that matter, to make such a discovery. But that doesn't mean the exercise is without value. For example, Dawn has criticized the fact that in current practice, it may takes some time for an application programmer to get a DBA to make a needed schema change. This is a very real issue that affects me regularly; I'm interested to hear her take on the issue. Maybe she has useful ideas that will make this less of a problem.

Of course, this issue is one of practice and not one of foundations, but that's okay. If someone solves this issue while mistakenly believing it to be a foundational issue, it is still solved.

Although I don't believe Dawn will invent a new logical structure, I could imagine she might come up with something interesting about schema change management. And even if she doesn't, that will help me understand the degree of hardness of the problem, and the degree to which the problem is fundamental.

From her and others' posts, it sounds like Pick makes schema change management easy. I'd like to understand why: does it come at a cost of expressiveness, is it a well-designed tool, what?

Consider: the fact that Microsoft Access has a crappy unscalable SQL engine as its core doesn't change the fact that it has some pretty cool RAD tools.

Marshall Received on Sat Oct 25 2003 - 19:58:18 CEST