- From: Patrick Stickler <patrick.stickler@nokia.com>
- Date: Tue, 19 Mar 2002 14:16:30 +0200
- To: ext Norman Walsh <Norman.Walsh@sun.com>
- CC: ext Mark Baker <distobj@acm.org>, URI <uri@w3.org>
On 2002-03-15 22:47, "ext Norman Walsh" <Norman.Walsh@Sun.COM> wrote: > / Patrick Stickler <patrick.stickler@nokia.com> was heard to say: > | Sorry. I don't buy that. A cache is a context holding proxies for a > | locations. And in that context, the URI still denotes a location, > | but you are bypassing the direct retrieval from that location with > | the proxy location provided by the cache. > > Right. > > | The semantics of the URI is still a location. That "opaque string" is > > This is the tricky bit. I've had enough conversations about this to > achieve a kind of understanding of the "there's no difference between > names and locations" school of thought. Well, I think it's a kind of > understanding, I could just be confused in different ways. > > The sequence of characters "http://site.org/resource-i-want" can be > viewed as a name. Just like I could change my name to Cheyenne Wyoming > if I was so inclined. > > The problem is that naming something (other than the city) Cheyenne > Wyoming is just begging for trouble. You're going to have all sorts of > conversations like this one: > > Random Person(RP): What's your name? > Cheyenne Wyoming(CW): Cheyenne Wyoming > RP: No, I didn't ask where you live, I asked what's your name > CW: I told you, Cheyenne Wyoming > RP: Yeah, I got that, you live in Cheyenne. What's your name. > CW: No, you don't understand, my first name is Cheyenne and my last name is > Wyoming. > RP: Really? Man, that must confuse everybody. > CW: Yeah, just about. > > Similarly, the string 'http://site.org/resource-i-want' looks exactly like > an address. So why call it a name? Why confuse things so much? Exactly ;-) And the same kind of confusion arises when folks use e.g. mailto: URIs to denote people and http: URIs containing only the web authority portion to denote companies, etc. Folks clicking around in browsers may be able to live with a certain degree of confusion of that sort, but SW agents trying to infer things about the world will be woefully handicapped by such "sloppy knowledge". > I dunno. I don't subscribe to that school of thought. > > But the semantics of a URI is not always a location. I didn't say the semantics of every *URI* is always a location, I said that an 'http:' URL always is (or should be) the name of a location. There's a bit of a difference, I think, between the two claims ;-) Cheers, Patrick -- Patrick Stickler Phone: +358 50 483 9453 Senior Research Scientist Fax: +358 7180 35409 Nokia Research Center Email: patrick.stickler@nokia.com
Received on Tuesday, 19 March 2002 07:15:15 UTC