- From: Ennals, Robert <robert.ennals@intel.com>
- Date: Mon, 22 Mar 2010 21:19:45 +0000
- To: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
- CC: "public-html@w3.org WG" <public-html@w3.org>, "Carr, Wayne" <wayne.carr@intel.com>
Leif Halvard Silli wrote: > Ennals, Robert, Mon, 22 Mar 2010 18:22:32 +0000: [snip] > I also think it should be possible to declare prefixes for attributes > inside HTML5 elements. Outside, <script>, to avoid that they are used > on element level, one could require that the prefix starts with a "_" > underscore. > > Your proposal doesn't solve the RDFa use case. > > > What do people think? > > > > As with my last proposal, if it looks like this will get support then > > I'll write it up. If it looks like there are serious problems then I > > won't. > > Much good. It looks to solve much of the element extensibility use > case. But little w.r.t. attributes. Yeah. This was intended to be purely a proposal for doing element extensibility, rather than attributes. If we want both, then it may be best to combine this with the way attributes are handled by my "proposal X". That is: * Element names cannot be prefixed – their namespace is determined by the innermost default namespaces or the innermost element that has an implicit default namespace (such as mathml or svg). * Attribute names can be prefixed, but the prefix must be registered, and the xmlns decl for the prefix must match what was registered. User agents do not look at this registry. It is used only to help guide spec writers, and for validator messages. * XMLNS SHOULD be declared for every attribute prefix used, but a user agent is allowed to give appropriate semantics to an element with a prefix it knows if the namespace decl is missing If I write this up then I think I'll probably make it work this way. -Rob > [1] http://www.amplesdk.com/examples/aml/databinding/ > [2] http://www.amplesdk.com/examples/svg/tiger/ > -- > leif halvard silli
Received on Monday, 22 March 2010 21:20:27 UTC