- From: Tim Kindberg <timothy@hpl.hp.com>
- Date: Fri, 25 Jan 2002 08:13:37 -0800
- To: Patrick Stickler <patrick.stickler@nokia.com>, ext Tim Kindberg <timothy@hpl.hp.com>, URN <urn-ietf@lists.netsol.com>, URI <uri@w3.org>
At 12:25 PM 1/25/02 +0200, Patrick Stickler wrote: >Names may only be valid or interpretable within a given context, but >I do not agree that the same name in different contexts can correspond >to different "things". So /etc/passwd is guaranteed to be the same file on all UNIX computers? So we have to give up the independence of bindings of names like 192.168.0.*? If you want the Web to be different then you have to define the property of the Web that makes it so. I believe that we do need a way of getting a default binding -- a mechanism whereby software can automatically get the address of the resource bound by the name's minting authority. But the software that I have as my client should equally be capable of using alternative naming contexts to reach alternative resources. Let's have a market of naming contexts just as we have a market of web sites. I shouldn't have to have new software to take advantage of a new naming context. E.g. imagine that a film has a globally/temporally unique name; now imagine all the sites/naming-contexts you might want to get resources from, using that same identifier. Even if you insist on saying that those other resources are likely to be 'metadata' about the minting authority's resource, they're still separately managed resources. Tim. Tim Kindberg mobile systems and services lab hewlett-packard laboratories 1501 page mill road, ms 1u-17 palo alto ca 94304-1126 usa www.champignon.net/TimKindberg/ timothy@hpl.hp.com voice +1 650 857 5609 fax +1 650 857 2358
Received on Friday, 25 January 2002 11:06:50 UTC