Re: Another fragment issue

Aaron,

I re-read TBL's Design Issues (DI) last week,
and found some interesting opinions about the fragment issue.

Some document in DI claims that there is no fundamental difference btw
URIs and URI refs.

I also read that 'fragment' must be read in a quite general meaning :
a fragment identifiers identifies in fact a "property" of the resource
(must be a property of the *representation* of the resource)

About RDF, "fragments" of an RDF document may identify virtually
anything :
   http://my.server.org/metadata.rdf#Freedom
identifies
  "the thing called Freedom in metadata.rdf"
Quite handy...

I guess it brings some answer elements,
anyway, I'm like you: this bothers me a bit.

  Pierre-Antoine