W3C home > Mailing lists > Public > www-tag@w3.org > December 2003

Re: New URI scheme talk in RSS-land

From: <jon@hackcraft.net>
Date: Thu, 11 Dec 2003 12:30:40 +0000
Message-ID: <1071145840.3fd86370df6ae@82.195.128.192>
To: Tim Berners-Lee <timbl@w3.org>
Cc: Graham <dtcd@mac.com>, "www-tag@w3.org" <www-tag@w3.org>

> - Such a reference is much more reusable (you can bookmark it, email it 
> etc);
> - It corresponds to a "place" in the "navigation" of the hypertext, so 
> human beings can use their inherent mental navigation metaphors;
> - It respects that the user may want control of what to with the thing.

/me travels along his inherent mental navigation metaphors...

Things people have gone to the address "G.P.O., O'Connel Street, Dublin, 
Ireland" to do:

1. Post a letter.
2. Buy stamps, including relatively rare stamps.
2. Stand outside holding placards in protest about something.
3. Collect for charity.
4. Declare national independence and commence an armed uprising.
5. Visit a famous historical site.

Granted the fourth was only done there once, however those are all locally well-
known reasons for going to that particular place.

If we were to apply semantics to a postal address and make it an identifier 
(which arguably is done with addresses such as that example, 10 Downing Street, 
1600 Pennsylvania Avenue and 221b Baker Street - although that last is a 404) 
then the semantics potentially relate to all of the above.

If we are to use it as a locator (like normal people) then any of the above are 
possible along with things one can do at any physical location (just like there 
are things one can do with any URL, such as bookmarking it).

Yep, my inherent mental navigation metaphors will buy the point about place.

I'm not sure that URIs are always seen this way. <urn:isbn:0201889544> wouldn't 
be seen that way (but is it seen *any* way?...). Even 
<http://www.google.com/search?q=w3c> says little of place. These do still 
respect the other two points.

What of <http://www.google.com/search?hl=en&ie=UTF-8&q=w3c> though? Is it a bad 
URI that contains information that should be transmitted elsewhere? Is it a 
reasonable value for a Location header - returned from a process which decided 
on the language and encoding through headers, but not appropriate for 
interchange? Can something be a reasonable value for a Location header if it 
isn't also suitable for interchange? If it is received along with headers 
asking for Japanese in Shift-JIS should it be considered a specific and 
explicit override requesting English in UTF-8?

--
Jon Hanna                   | Toys and books
<http://www.hackcraft.net/> | for hospitals:
                            | <http://santa.boards.ie>
Received on Thursday, 11 December 2003 07:31:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:23 GMT