Levels of rights information

The Web is an almost inexhaustible source of resources. With an internet connection we are truly standing on the shoulders of giants. However, we cannot freely use any resource online, unless we know we are authorised to. For example:

  • Content published along with a CCZero waiver can be used for any purpose (commercial, etc.) with no restriction.
  • Content published with a license typically permits some uses
  • Content published with no rights information, cannot be used for anything besides seeing/hearing it

In the Semantic Web paradigm, we like computers to understand everything, including rights info. From this point of view, we may define four levels of quality:

No licensing terms

Using unlicensed resources in industrial settings usually is an unacceptable source of legal troubles.

Licensing terms as a text

The human has to read an understand the licensing terms. This might not be trivial (e.g. think of the text being in a foreign language you don't speak)

Licensing terms refer to a well-known URL

This is perhaps the best practice as of today for the majority of the cases.
Referring to well-known licenses (like Creative Commons' or Open Data Commons') minimizes unambiguity, and to some extent, permits an automatic parsing: machines can be trained to understand famous licenses.

Licensing terms as RDF

As these options are not mutually exclusive and the vast majority of people (including myself) dislike reading RDF, this rights declaration can be accompanied by one of the former.
In fact, if no machine-discovery is foreseen, describing the detailed licensing terms with RDF is overengineering-- hence a bad practice.

.: