Re: about"Peter Chen"

"Karel van der Walt" <karelvdwalt_at_webmail.co.za> wrote in message news:e743ac89.0409290218.29d91d97_at_posting.google.com...
> For me the novel thing was that ER model was originally proposed as
> yet another logical i.e. representational model. The use of ER as
> conceptual i.e. semantic model with mappings to the prevalent logical
> models of relational, hierarchical and network, only followed later.
>
> Another conceptual model would be ORM.
>
> hth
> k

I don't think so. All my early exposure to ER was as a conceptual model. In particular, I think it was Chen's intent to produce a model that served as a bridge between the network models and the relational model. And I've seen it work out that way in practice, too. I once read Chen's paper. I don't remember much, but I wasn't jolted into thinking he was trying for another logical model.

There was a guy at one customer site who had worked out an ER diagram of a gigantic network database, and he really had abstracted out the design decisions that were predicated on a network implementation. Taking any subset of that ER diagram, and turning it into a relational design, and an RDBMS implementation was an easy exercise. Received on Wed Sep 29 2004 - 13:55:20 CEST