Re: On Formal IS-A definition

> > I am unfamiliar with any restrictions on what goes in sets.

> Values are immutable.

If I may be so bold as to interject my two cents : I suspect you two are talking past one another.

It is true that mathematical set theory has no constraint on what type of thing can or cannot be member of a set. It is also true that the application of mathematical set theory in the context of relational data management, talks about sets of values exclusively, thus in a sense indeed limits sets to contain values exclusively.

I've never seen any theoretical stuff that tried, e.g., to define a program as a set of variables and then exploit the properties/axioms/ postulates of set theory to deduce therefrom some kind of "settheoretical  theory of programming". But of course that may just as well me due to my limited and unidirectional vision.

Exchanging arguments without agreeing on the premisses is not a fruitful debate. Regardless whether those arguments are detailed reasonings or dense and almost cryptic oneliners. Received on Sat May 08 2010 - 11:51:33 CEST