Re: Looking for a discussion about generic datamodels

<schreurs_roel_at_hotmail.com> wrote in message news:1125653973.496115.14570_at_g43g2000cwa.googlegroups.com...
> Every now and then, I come across a IT-project that stores its data in
> a generic data model. Such data models basically consist of 4 tables:
> Entities, Relations, Attributes and Values.

Me too. This brainstorm appears to be frequently re-discovered by the medical research community in particular. (There is an extensive literature, in refereed journals no less.)

There are innumerable reasons to object to it, and I look forward to seeing some of them rehearsed here, for convenient reference. I will get the ball rolling with my personal favourite: take the hardest query you ever wrote against a "properly" designed database, and try re-writing it for one of these designs. If that's too hard, try a simpler query. Keep relaxing the difficulty until you can discover one you actually can rewrite. That is will be the limit of what you can do with this approach.

Roy Received on Fri Sep 02 2005 - 13:50:01 CEST