just yesterday i was writing about the OAI-ORE data model and how it replicates previous efforts in defining formats for linking resources on the web. today, i ran into draft-snell-atompub-link-extensions (which of course had to be added to my Atom Landscape Overview, even though it is outdated and expired in July 2006), and now i am wondering how that one compares to the linking approaches mentioned yesterday, which were XLink, METS, DIDL, and OAI-ORE.
while i do think that exposing some information about resource relationships in Atom is a good thing, i am a bit worried about the fact that this again introduces a new model
of what linking resources
means. and this model is not defined anywhere (as far as i can tell), it is implicit in the elements and attributes defined by the proposed extensions. at least, the draft does reuse some existing concepts, such as the media attribute for identifying for which media type a resource is intended.
from my first impression (i still need to read through the draft in detail), it looks to me as if it is trying to solve a real problem, i.e. how to put entries in an atom feed into context. what i still need to find out is how well the draft connects with previous approaches to solve the same problem.
Comments