Re: The Practical Benefits of the Relational Model

> What burdens programming languages is lack of conceptual integrity, not
> their ambition.

As I said, those who are impure cannot see the holy grail. Lack of concpetual integrity is simply another way of putting it. I differ about whether ambition is irrelevant. The more ambitious the goal, the more lack of conceptual integrity gets in the way. It's a trade off.

> If it's trivial, why has it not been done? Or has it?

A real good question, Paul. AFAIK no one has built these capabilities into a language. What comes close are some of the "in memory" databases, or so it seems to me. I'm looking forward to hearing from people who have used them, or who know that "it" has been done.

I'm going to add the proviso that, while extending the language to allow relational operators to be expressed is trivial, implementing the relational operators is anything but trivial. A crude implementation is likely to be terribly slow, or not general enough, or both. And implementing them the right way is likely to involve a large engineering effort, if the example of engineering relational DBMS products is any indicator. Received on Tue Sep 17 2002 - 23:16:15 CEST