"Publishing" means making a resource available for the use of others. A "resource" could be "anything" including data and services, identifiable over the Web. Publishing may be done on the basis of either the "characteristics" of the data or the data itself. When only some "characteristics" of a resource is published then some of the contents would naturally be left out. The "characteristics" include metadata and some keywords. This kind of publishing may be named as "limited contents" publishing or "publishing by metadata". One of the issues become then what characteristics to use to define the data. Or what what metadata definition to use. Another aspect of publish is "manual entry" and "automated collection". In the former publisher enters metadata while in the latter some harvesting mechanism collects metadata in an automated fashion. On the contrary, there is "unlimited contents publishing" where there is no limitation on the published contents. Open data publishing is in this class. In additon, some "additional semantics" may be subject of this type publishing through new relationships in the ontologies of publishing, which have not been explicit in the exisiting data model but are inherent in the data. And this last type is covered under the topic, "Publishing via a semantic definition of data."