2006/5/10, Tim Bray <[EMAIL PROTECTED]>:
On May 10, 2006, at 12:01 PM, John Panzer wrote:
> There is a separate question lurking here: Whether or not we
> mandate multiple URIs, or allow them, or ban content negotiation...
> is there one resource hiding behind the URI(s) or two?
That is an unanswerable question; the Web architecture allows you to
say that two URIs that are character-for-character identical identify
the same resource, and in a few cases, that two URIs that are
slightly different can be shown to be "equivalent". But beyond that,
the Web gives no help. It turns out that http://www.tbray.org/
index.html and http://www.tbray.org/ always produce the same
representation, but do they identify the same resource? Whichever
answer you give, I can quickly make it wrong. See http://www.w3.org/
TR/webarch/#identifiers-comparison -Tim
That's about URI equivalence or identifier comparison, that's about
the concept/notion of resource.
Given what I wrote here:
http://www.imc.org/atom-protocol/mail-archive/msg05099.html, I believe
that there is a single resource hiding behind the URI(s) –I mean, in
the APP, wrt to "media entries" handling, there are two URIs at least,
but a single resource, with at least two representations (hence the
two URIs)–.
Now, is this ("media resource" and "media link entry" as two
representations of the same resource, or as two distinct, related
resources) impacting the interop of the protocol? [1] I can't tell.
I tend to think that there is no impact as long as you only use the
"core" APP, but that it's likely to prove wrong as soon as extensions
will come (extensions or "bindings" –WebDAV, HTML forms, etc.– but I
think relationship to bindings could be seen as implementation
details)
--
Thomas Broyer
