Re: Hierarchal vs Non-Hierarchal Interfaces to Biological Taxonomy

"Larry Coon" <lmcoon_nospam_at_cox.net> wrote in message news:458EB4E4.115C_at_cox.net...

> I don't think there is a universal definition for
> species. Remember, the taxonomies are organizational
> conveniences, and as someone else said, nature is too
> subtle to fit cleanly within our systems.

I'd like to take the above comment and extend it to cover not only biological taxonomies, but also any part of the universe that might be the universe of discourse for some proposed database.

From time to time here in c.d.t. some contributors have objected to ER modeling on the grounds that "entities" are too subjective to have any place in a clean logical model. To some extent, entities are more subjective than objective. The division of the universe of discourse into entities and the discovery of relationships between entities are abstractions that we humans use to deal with complexity, including the complexity of the universe of discourse. To some extent, these are a matter of convention, rather than revealed truth or verifiable hypothesis.

The next step, to abstract attributes of entitites or of relationships that can be described by data values is, also to some extent, a matter of convention.

But that's the way it is, folks. Databases are about values, and the meaning of values are inextricably tied up with conventional meaning. You can make up a database that only you can understand and decipher. But don't expect other people to value such a database. Received on Sun Dec 24 2006 - 22:56:49 CET