Relational Theory & the New Math

A colleage of mine was talking today about how the New Math of the 1960's and following grew out of the efforts of Bertrand Russel and others at the start of the last century to derive mathematics from logic. This is when set theory started getting introduced in a big way in schools. Russell's approach has been largely abandoned as a basis for mathematics (IIRC).

The 1960's is also when Codd started working with sets as a foundation for databases. I don't know the status of the "New Math" in our schools and I do definitely have an appreciation for set theory. But it is NOT everything-- there are other mathematical models that can shed light on our discipline and help us get our jobs done. With relational theory we are trying to base everything on set theory, but when that doesn't have what we need, then we add other functions as needed. If we start out recognizing that there are a variety of functions required -- those that aggregate data, as well as those that extract sub-values from data, we might not tie ourselves unnecessarily to set theory exclusively.

I think I said that poorly, but thought it was at least worth noting. I know I still owe more info on non-1NF, but I don't have it in good enough shape yet. Cheers! --dawn Received on Thu Oct 14 2004 - 20:10:18 CDT