W3C home > Mailing lists > Public > public-iri@w3.org > May 2004

Re: Migration of HTTP to the use of IRIs [altdesign-17]

From: Martin Duerst <duerst@w3.org>
Date: Sun, 09 May 2004 16:04:04 +0900
Message-Id: <4.2.0.58.J.20040507151438.05acaaa8@localhost>
To: "Chris Haynes" <chris@harvington.org.uk>, <public-iri@w3.org>

Hello Chris,

I have changed this to be about issue altdesign-17.

At 12:49 04/05/06 +0100, Chris Haynes wrote:

>Martin,
>
>Thanks for this response.
>
>Actually, my original core concern has now been covered in your section 
>1.2.a -
>Applicability, where you make it clear that "the intent is not to 
>introduce IRIs
>into contexts that are not defined to accept them".
>
>This now makes it clear that new schemas will be required to replace http: ,
>https: etc. These will need to be self-identifying in some way, so that
>receiving equipment will know that an IRI is being presented.

I'm very surprised that you interpret this in this way. The point
discussing applicability of IRIs with respect to various URI schemes
is section 1.2.c. This is the only applicability item where URI
schemes are discussed. [I hope you don't confuse XML Schema and
URI schemes.]

What the spec means by 'contexts' is slots in a protocol or format
where URIs or IRIs may go. As an example, the first line in a
HTTP protocol exchange usually looks like:

GET /foo/bar/baz.html HTTP/1.1

The "GET" is the request method. The "HTTP/1.1" is the protocol
version. The "/foo/bar/baz.html" is an URI, absolute (for proxies)
or relative to the server in question. The HTTP spec designates
this as an URI, and section 1.2.a says that you are not supposed
to suddenly use an IRI in such a context, which means that if you
have an IRI that you want to resolve, you have to convert it to
an URI as described in the IRI spec.


>So, as I commented last June, I await with interest the recognition among 
>those
>responsible for the HTTP schema that new schemas with new names are required
>before IRIs can be used.

There is no such recognition, and there is no need to create new schemas.
While creating new schemas might in one way or another lead to a cleaner
solution, it would be overkill because it would be very difficult to
deploy.


Regards,    Martin.



>Returning to the logged issue...
>
>Your new paragraph in 7.8 is helpful, but not, I fear, strictly accurate.
>
>The phrase "returned query URIs will use UTF-8 as an encoding" is accurate 
>only
>if the browser's user has not manually changed the page encoding via the menu
>commands available to her (e.g. with MSIE the  "View - Encoding" menu 
>sequence).
>It can easily be demonstrated that this user selection of the encoding 
>overrides
>the encoding declared in the HTML text or associated HTTP header when requests
>are formulated.
>
>'will use' is therefore too strong.
>
>Changing the phrase to "returned query URIs will, by default, use UTF-8 as an
>encoding" is an accurate statement - it just leaves open the question of what
>'by default' means.
>
>Chris
>.
Received on Sunday, 9 May 2004 08:37:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 30 April 2012 19:51:53 GMT