Mark Nottingham wrote: > FYI; this would affect the interpretation of <link> in Atom as well...
How would it affect Atom? Are you suggesting that some RFC update RFC 4287 to change the syntax of links in Atom? > > The alternative is to say that the 'stylesheet alternate' > > combination isn't specific to how it's serialised, but is tied to > > the occurrence of the links. I.e., when both relations are present > > in links between the same resources, these special semantics take > > affect. However, this does seem to directly conflict with the HTML4 > > language (see link above), so I don't think doing so is viable. Atom cannot express "stylesheet alternate" without defining a new, explicitly-named link relation, since atom:link/@rel cannot contain any spaces. I don't see how that limitation of Atom can removed without breaking many existing Atom processors. This was one of the issues I brought up when arguing against one of the early Link header drafts. Although HTML, Atom, HTTP, and other languages have very similar linking syntax and semantics, there are many important differences that make it impractical to unify them. - Brian
