Re: The wisdom of the object mentors (Was: Searching OO Associations with RDBMS Persistence Models)
Date: 2 Jun 2006 05:57:27 -0700
> drawbacks appear more troubling to me into handling better relational
> requirements are the fact that SQL neither correctly support domain
> definition, nor it implements any real coherence of what relational
> data types are.
> The main impact is that a better integrity preservation, a core issue,
> becomes very difficult to implement.
In my opinion SQL does a fabulos job even with poor system of datatypes. VARCHAR2(20) what a humpback! I also don't see how particular SQL syntax quirks impede an ability to write assertions (constraints). Complex assertions can even be implemented via basic constraints (CHECK, UNIQUE, FOREIGN KEY) on materialized views.
Again, overthrowing SQL would require a deaper insight. For example, what if we treat functions as relations? Shouldn't we write
select ename, sal2 from emp, (sal2=2*sal) f where f.sal = emp.sal
as a "pure" relationaly styled version of
select ename, sal*2 from emp
? Is allowing predicates in the "where" clause give us any benefits? Received on Fri Jun 02 2006 - 14:57:27 CEST