[tm-pubsubj-comment] Supporting the semantic web

[Lars Marius Garshol <larsga@garshol.priv.no>]

I had an interesting discussion with Eric Miller yesterday on IRC
about published subjects and their possible use in a semantic web
context. He hoped published subjects could be used there as well, but
listed three specific concerns that he thought were important for the
success of PSs:

  (1) That the organizations currently responsible for making
      identifying codes for subjects/resources should themselves take
      over the responsibility for maintaining the published subject
      sets based on their code sets.

  (2) The issue of what published subject identifiers resolve to, and
      whether that is machine-readable or not.

  (3) That stability of published subject identifiers could be
      achieved and, as far as possible, guaranteed through the right
      technical and political measures.

We discussed these issues back and forth a little, and basically
agreed that they were all key issues, and I said that I thought Eric
and the TC would be likely to agree on (1) and (3), but that (2) might
be harder.

I managed to persuade him to join the list, so what I would most like
to see right now would be for Eric to explain in more detail what he
sees as the requirements for (2) and (3). For example, what
information does Eric think needs to be at the end of PS URI? Also,
what are the actual requirements behind (3)?

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
ISO SC34/WG3, OASIS GeoLang TC        <URL: http://www.garshol.priv.no >