Re: [xml-dev] XML Schema as a data modeling tool

[Michael Kay <mike@saxonica.com>]

On 30 Sep 2013, at 21:49, Hans-Juergen Rennau wrote:

> Michael, you sound as if hierarchy is mainly the result of applying this or that view to a network, the mere results of transformations, without substance of its own. I disagree. 
> 

Yes, I think the hierarchies we play with are usually arbitrary projections of a network. They work well for documents or messages, they don't work well for models of things in the world. If you take any real-world domain - something like books, authors, publishers, and retailers, there's no natural hierarchy between them. You can force a hierarchy if the technology constrains you to do so, but it's artificial - and conceptual data models shouldn't be constrained by the technology you are using.

I'm NOT saying that a conceptual data model should be in 3rd normal form. Within these real-world entities like books and authors there is hierarchic attribute information (such as the brands or imprints belonging to the publisher, or the branches of the retailer) and normalizing these all only complicates the model. But any attempt to put books under authors or authors under books is essentially misguided.

Michael Kay
Saxonica