- From: Aaron M Leventhal <aleventh@us.ibm.com>
- Date: Mon, 24 Sep 2007 10:25:21 -0400
- To: "Simon Pieters" <simonp@opera.com>
- Cc: "Anne van Kesteren" <annevk@opera.com>, public-html@w3.org, wai-xtech@w3.org
- Message-ID: <OF280CB277.E35AC850-ON85257360.004E702F-85257360.004F32E1@us.ibm.com>
> Do you mean only for HTML elements or elements in the > http://www.w3.org/1999/xhtml namespace? (I don't see any reason why role > attributes in no namespace on elements in the http://www.w3.org/1999/xhtml > namespace should not work in XUL or SVG host documents.) Right. > The XHTML role attribute module seems to be a bit confused on this point, > but AFAICT the intent of the spec is to have the role attribute be in no > namespace for elements in the http://www.w3.org/1999/xhtml namespace (e.g. > the attribute is in no namespace in all examples in the spec). Can you > point to any existing content that relies on this to be supported? We don't have access to all of the examples out there. There are companies working with ARIA now, but we don't have the privelege (or punishment) of looking at their source. Originally role was an xhtml2 namespaced attribute. After the role attribute came along to bring it into XHTML 1.1 (modules are technically not supported for XHTML 1.0 afaict). Thus, I felt that technically we needed to continue to support role as a namespaced attribute. If you think it is worth it, we can attempt to contact the companies and projects we know that are working on ARIA, and see how they are currently using the role attribute, and if they are willing to change it. But for me that's a much smaller issue than this main fix of allowing ARIA to be easily used in text/html. I'm willing to persue it, if you think it is important. - Aaron > > > 3. http://www.w3.org/TR/xhtml2 > > > > We go from least obscure to most obscure in our priority. > > Ok. > > > [...] > > -- > Simon Pieters > Opera Software
Received on Monday, 24 September 2007 14:25:45 UTC