W3C home > Mailing lists > Public > www-tag@w3.org > November 2002

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

From: Julian Reschke <julian.reschke@gmx.de>
Date: Fri, 29 Nov 2002 20:32:36 +0100
To: "Richard Tobin" <richard@cogsci.ed.ac.uk>
Cc: <www-tag@w3.org>, <xml-names-editor@w3.org>
Message-ID: <JIEGINCHMLABHJBIGKBCCEFGFOAA.julian.reschke@gmx.de>

> From: www-tag-request@w3.org [mailto:www-tag-request@w3.org]On Behalf Of
> Richard Tobin
> Sent: Friday, November 29, 2002 1:38 AM
> To: julian.reschke@greenbytes.de
> Cc: www-tag@w3.org; xml-names-editor@w3.org
> Subject: RE: last call comments, usage of IRI rather than URI
>
>
>
> > It has been demonstrated that using IRI (refs) as namespace names breaks
> > existing XML applications (such as XML Schema, schemaLocation
> attribute) and
> > APIs (JAXP). I'd like to understand why this isn't considered a problem.
>
> IRIs that contain spaces can't be used with the schemaLocation
> attribute, but IRIs in general can, as far as I can see.  It seems to

Well. IRIs in general cannot be used in the schemaLocation, only a specific
subset can (yes, I know, this isn't a contradiction, but this statement is
equally true :-).

> me that IRIs with spaces have disadvantages in other contexts too (eg
> in plain text), and that we should either not allow spaces in IRIs or
> live with the problem, rather than not using IRIs at all.

The former. IMHO, IRIs should concentrate on their mandate (which is I18N,
right). Lots of problems would go away if they wouldn't allow *ASCII*
characters that are forbidden in URIs.

Julian

--
<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Friday, 29 November 2002 14:33:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:13 GMT