- From: Martin Duerst <duerst@w3.org>
- Date: Wed, 30 Apr 2003 11:36:14 -0400
- To: Bjoern Hoehrmann <derhoermi@gmx.net>, public-iri@w3.org
Hello Bjoern, Many thanks for this and other comments on the IRI spec. At 05:23 03/04/29 +0200, Bjoern Hoehrmann wrote: >Hi, > >Section 6.3: > > "Note: Some formats already accommodate IRIs, although they use > different terminology. HTML 4.0 [HTML4] defines the conversion from > IRIs to URIs as error-avoiding behavior." > >HTML 4.0 does not accomodate IRIs, HTML 4.0 defines the value of the >href, src, etc. attributes as URI References as per RFC 2396, a document >using IRIs that are not legal URI References is thus invalid. HTML does >indeed recommend an error recovery behaivour, but nothing beyond that. I have noted this as issue http://www.w3.org/International/iri-edit#htmlIRI-12 I think the term 'accommodate' is rather loose, and the text points out that for HTML 4.0, this is just error behavior. So I don't thing that there is anything wrong with the current text. If you think that the current text needs some changes, can you please propose actual wording? Also, please note that formally (in the DTD), all the relevant fields are defined as CDATA. This means that validation doesn't catch IRIs, and documents containing IRIs, while not conforming to the HTML 4.0 spec, are therefore not invalid. Regards, Martin.
Received on Wednesday, 30 April 2003 12:11:47 UTC