W3C home > Mailing lists > Public > public-webarch-comments@w3.org > October to December 2004

[closed] non-authoritative syntaxes for fragment identifiers

From: Roy T. Fielding <fielding@gbiv.com>
Date: Mon, 25 Oct 2004 12:36:04 -0700
Message-Id: <1BED7574-26BD-11D9-BBD4-000393753936@gbiv.com>
Cc: public-webarch-comments@w3.org
To: Myriam Amielh <myriam.amielh@cisra.canon.com.au>

Hello Myriam,

On Sep 27, 2004, at 4:38 PM, Myriam Amielh wrote:
> Thank you for your clarification.  It makes much more sense for those 
> statements to apply to any URI, regardless of fragment.  I agree that 
> it is preferable to delete the two sentences from section 3.3.1 [Media 
> types and fragment identifier semantics] to remove the confusion.

This section has been updated to reflect your comments

   http://www.w3.org/2001/tag/2004/webarch-20041021/#media-type-fragid

> While we are on the topic of interpreting a fragment identifier, some 
> W3C recommendations and working drafts such as SVG and the XPointer 
> use scheme names in their fragment identifiers to assist the 
> interpretation of the fragment.  For instance, SVG uses 'svgView' 
> (e.g. http://www.example/file.svg#svgView(...)) and XPointer uses 
> 'xpointer' (e.g. http://www.example/file.svg#xpointer(...)). Has the 
> working group considered recommending such practice in AWWW, or do you 
> consider such recommendation as encouraging "the interpretation of 
> fragment identifier based on a syntactic analysis on part of a URI"?

We have not considered that, primarily because there is a preference
for media-agnostic identifiers, though you are certainly welcome to
bring up that discussion on www-tag@w3.org.


Cheers,

Roy T. Fielding                            <http://roy.gbiv.com/>
Chief Scientist, Day Software              <http://www.day.com/>
Received on Monday, 25 October 2004 19:36:40 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:26:48 UTC