A few months ago Flickr, the photo sharing site, announced geo tags. Geo tags let the user specify the longitude and latitude of a picture (e.g, “geo:long=123.456”), resulting in the ability of displaying photos on a yahoo map.
As Dan Catt writes
“They are called “Machine Tags” because we expect them to generally be added by automated systems and later sucked up and processed by machines. “
What is then the difference with this and RDF?
For one thing while RDF relies on a specifying properties though URIs (e.g. http://xmlns.com/foaf/0.1/firstName abbrevated to foaf:firstName), they are formally described in a vocabulary (or ontology). Machine tags (the way I’ve understood it) however are not defined anywhere, but rather up to the community to define through use. It is however likely that communities will create useful tags-sets for their respective domains and interests. And most important it will bootstrap the Semantic Web as machine tags could quite easily be converted into RDF triples. Exiting stuff.