Relational Database Normalisation Process, Jul 17, 2006
Commentary by Stephen Downes

In some senses, the Semantic Web is like a big distributed database. This applies equally well to less formal instances, such as the RSS network. As such, there are basic, well-known, principles for its design. One set of these falls under the heading of 'normalization'. The idea of normalization is that you don't needlessly duplicate data entries. That's why I wince every time I see an author name entered into an element as a string instead of as a reference (or link) to an author record. This article explains database normalization and offers some examples. It's a bit of work to get through, but if you design metadata schemas or systems, please have a look.
Views: 1 today, 223 total (since January 1, 2017).[Direct Link]