Re: Modelling objects with variable number of properties in an RDBMS

Roy Hann wrote:

> ...
> Actually no, let's not. What you have re-invented here is the tired old
> entity-attribute-value (EAV) design (much beloved of medical research wonks
> for some reason).
 > ...

I have the same (limited) impression. I wonder why? Are they truly misguided or do they have some reason to think that the computer will discover relationships they can't pin down? Is finding those relationships the purpose of the research in the first place and are they just taking money for no good reason?

To put it another way, is using a conventional dbms for such aspects of research the wrong approach or is that that software is missing some ingredient/feature that other computer science approaches deal with more readily?

(I admit I know almost nothing about medical research except from a movie or two about isolated genius and jealous colleagues and the many ordinary people I see running for cures who are convinced that money is the solution.)

pc Received on Sun Oct 30 2005 - 14:11:55 CET