Re: Full Name as Composite Attribute
Date: Tue, 29 Mar 2005 13:13:01 GMT
Message-ID: <xvc2e.11159$S46.9808_at_newsread3.news.atl.earthlink.net>
"Troels Arvin" <troels_at_arvin.dk> wrote in message news:pan.2005.03.28.18.52.02.338788_at_arvin.dk...
> Isn't this the same problem as "what's an atomic value"? A word (a string)
> is composed of individual syllables and characters. So is the word atomic?
There is a lot of overlap, but I'd say there's a shade of difference.
In the "atomic values" discussion, the consensus was that the "relational engine" perceived character strings as "atomic values", while the "character string type engine" could perform subatomic operations on character strings.
The way I see it, "composite attributes" (attributes composed of other attributes) makes the composition visible to the data model itself, thus not requiring a separation of duties like the above. Received on Tue Mar 29 2005 - 15:13:01 CEST