- From: Murray Maloney <murray@muzmo.com>
- Date: Tue, 09 Jun 2009 14:58:30 -0500
- To: "Edward O'Connor" <hober0@gmail.com>
- Cc: Shelley Powers <shelleyp@burningbird.net>, "public-html@w3.org WG" <public-html@w3.org>, Sam Ruby <rubys@intertwingly.net>
At 10:58 AM 6/9/2009 -0700, Edward O'Connor wrote: > > With the new June 3rd addition to the HTML 5 specification that > > demands active non-support for elements/attributes not specifically > > mentioned in the spec, user agents are now told that they must not > > support non-existent (never existed, or previously existed but now > > obsolete) elements and attributes. According to the HTML 5 spec, FONT > > could then be non-conformant, which means, if I read the HTML 5 spec > > correctly, user agents _must not_ support the element. Are you saying: "User agents claiming conformance with HTML 5 must not process any markup is not specified by HTML 5?" Doesn't that mean that nobody will ever be allowed to create new markup? How then will new cow paths develop? >I've just examined all changes to the spec on June 3rd (svn diff >-r3180:3187 source), and I wasn't able to find a change like the one you >describe. Could you be more specific? > >Perhaps you're referring to r3183, which adds this text: "Authors must >not use elements, attributes, and attribute values that are not >permitted by this specification or other applicable specifications." >Note that this is an /author/ conformance requirement, not a /user >agent/ conformance requirement. I am confused. Surely an HTML specification has no authority over authors. If such a claim is being made, then on behalf of all authors, I reject it. Perhaps it could be rephrased to say that all documents claiming conformance with HTML 5 must only contain markup specified by HTML 5, and that all producers of documents claiming conformance must not produce document content that is not permitted by HTML 5 [or other applicable specifications]. Regards, Murray
Received on Tuesday, 9 June 2009 18:59:40 UTC