W3C home > Mailing lists > Public > xml-names-editor@w3.org > November 2002

Re: last call comments, usage of IRI rather than URI

From: Richard Tobin <richard@cogsci.ed.ac.uk>
Date: Thu, 28 Nov 2002 18:34:50 GMT
Message-Id: <200211281834.SAA20049@sorley.cogsci.ed.ac.uk>
To: julian.reschke@greenbytes.de
Cc: xml-names-editor@w3.org

This is a formal response from the XML Core WG to your comments on the
Namespaces in XML 1.1 last call working draft.

If we haven't heard from you by the end of Monday December 9th, we
will assume for the purposes of our planned CR request that you have
no objection to our resolution.

Commenter email address: julian.reschke@greenbytes.de

> Subject: last call comments, usage of IRI rather than URI
> 
> Looking at it -- I'm not convinced that allowing IRI references is a good
> idea. There is existing code out there that assumes that every namespace
> name in an XML document can be represented as a URI. With a IRI containing
> non-ASCII characters, this is not true anymore. Translating to a URI will
> not work because
> 
> - namespace identity is defined on a char-by-char basis, so the resulting
> URI does *not* identify the same namespace and
> - for a recipient of a "down-transformed" IRI there's no way to know it
> originally used IRI syntax (so there'll be problems round-tripping them).
> 
> This is *not* an academic argument -- our WebDAV server implementation
> relies on the fact that WebDAV properties are identified by URIs, and will
> reject those which do not conform to the URI syntax. So when at some point
> of time the WebDAV WG will have to decide about what do to with XML 1.1, it
> may have to restrict the set a legal namespace names to URIs.
> 
> To satisfy I18N requirements (which I *do* understand), wouldn't it make
> more sense to state that the characters in a namespace URI, when un-escaped,
> SHOULD map to valid UTF-8 octet sequences (possibly normalized)?

Summary: rejected

We decided that we should not make a decision on this ourselves, so we
consulted the Technical Architecture Group, and their view was that we
should use IRIs.  It is likely that an increasing numbers of
recommendations will specify the use of IRIs; Namespaces is just one
of the first to refer to them explicitly.

We will include a warning that authors should stick to URIs during
a transitional period.

-- Richard Tobin, Namespaces 1.1 editor
Received on Thursday, 28 November 2002 13:34:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:43 GMT