Re: Attribution for modeling data with relations

David Cressey wrote:
> "dawn" <dawnwolthuis_at_gmail.com> wrote in message
> news:1141232675.124053.15680_at_j33g2000cwa.googlegroups.com...
>
> > Does anyone know to whom should we attribute the modeling of data with
> > relations?
>
> I had mentioned a year or more ago that some people had used relational math
> as a basis for "inference engines" before Codd's 1970 paper (and before the
> 1969 paper that I previously didn't know about, also).

I recalled that someone had mentioned something about it. Is the 1969 paper internal to IBM? Do you have a URL?

> Just starting with the footnotes in Codd's paper, and doing a little
> searching led me pretty quickly to this:
>
> www.rand.org/pubs/research_memoranda/RM5947
>
> I think this work, from 1967, predates Codd's vision of the RDM. What is
> new with codd is the application of the RDM to large shared data banks.

And normalization.

> The RDM could have been, and was, implemented on a files system, as I
> think the above citation shows.

Good point.

> A DBMS could have been, and was, built on some model other than the RDM.
> Witness IMS.

Certainly. (BTW, I wonder if IBM makes more dollars on DB2 than on IMS yet)

Cheers! --dawn Received on Wed Mar 01 2006 - 21:48:12 CST