Re: Impossible Database Design?

Cheaper? No doubt. Better? I'm not sure.

Snodgrass and Jensen and Christianson and Ben Ziv (I'm apologize if the names aren't correct) and others have made immense contributions to the field of temporal data.

But have you looked at the solutions presented in the Snodgrass book? Yes, it's a significant achievement and, yes, it gets things done using SQL, and yes, it may well work. But the complexity! The redundancy! The work-arounds! The number of situations which result in 'RAISE...ERROR' with no way out! It's an absolute horror. Sure, convincing someone to implement such a system will keep plently of arcane SQL coders pounding keys for quite a long while -- but the chances of actually delivering something significant (Say, a design that results in more than 100 bi-temporal tables.) and correct are slim.

Date's approach seems completely reasonable to me. After all, we are dealing with computers, right? The best we can hope for are acceptable representations of continuous systems. I mean, is there really a need to handle a time interval as an infinite number of instants?

Yes, he presented a language that isn't implemented. Judging from what I've seen in SQL-implementations of semi-temporal and bi-temporal data stores, it should be! Received on Wed May 17 2006 - 17:58:04 CEST