[sc34wg3] Subject Locators

[Lars Marius Garshol]

* Jan Algermissen
| 
| Ah. That is a sound statement, compatible with HTTP. I like it.

Good. :)
 
| By definition what you get when dereferencing a URI is a
| representation. So the topic represents the (one) representation of
| the resource identified by the URI in the <resourceRef's> href.

Yes, this is what I tried to say at the beginning. It may not have
been very clear, though.
 
* Lars Marius Garshol
|
| which can't be Jim Smith, but has to be an information resource of
| some kind.
 
* Jan Algermissen
|
| If 'information resource' == representation. Fine. Must be since
| HTTP/ REST defines that what you get isa represenation.
| 
| Agree?

Almost. Every representation in the REST paradigm is an information
resource; not sure if every information resource is a representation
in the REST paradigm, but I think it's OK for the "information
resource" class to be more inclusive.
 
| And XTM/TMDM is deliberately ignoring that HTTP content negotiation
| (and time) might lead to different representations over time, yes?

Yes.

| Sometimes I will end up with an HTML document representing Jim Smith
| another time a JPEG image.  (no problem at all, just want to make
| sure we agree)

Well, it might be a problem, but given that this is a problem with
URIs rather than topic maps there's not that much we can do about
it. If someone gets into trouble over this they need to either not use
content negotiation and keep their resources stable or, uh, use
HyTime.

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >