Flat RDF Documents

From OpenOrg
Revision as of 13:30, 11 January 2011 by WikiSysop (talk | contribs) (Boilerplate)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

If you have very limited resources & skills then your options are limited.

The most simple way to publish RDF is as a single RDF+XML document somewhere on your website. This document could be produced either by hand, or with a tool.

Other options would be to use a web based service (when they start existing), or to embed RDFa in your HTML pages (although this also requires either a tool or some training/skills)

Writing RDF+XML by hand

While this isn't an ideal solution, here's a guide to the absolute minimum you need to know.

URIs

These are the strings that uniquely identify your things (like places, people etc.). Ideally they should look something like one of these:

 http://id.example.org/building/23
 http://example.org/id/person/ddf

However, these both require you to do some clever stuff with your webserver (or have a tool which does). If you are just creating a flat plain single RDF document you can just use ones something like this:

 http://example.org/rdf-documents/buildings.rdf#building-23
 http://example.org/rdf-documents/people.rdf#person-ddf

...It's not quite as slick as the first examples, but it'll work and you don't have to do anything clever with your configuration.

This means that any time some tool looks up the identifier "http://example.org/rdf-documents/buildings.rdf#building-23" then it will get the contents of "http://example.org/rdf-documents/buildings.rdf" and therefore find out some facts about Building 23.

Boilerplate

Even if you are creating RDF files by hand, you should still make sure each one contains at least a little metadata about the RDF document itself. At the very least a license to let people know what they are allowed to do with this data.

  • See Boilerplate for much more information and examples.