Ted, Please correct me if I get any of this incorrect, but for the sake of the discussion I wanted to summarize the HTML5 [1][2] definitions here:
The following three links are equivalent to one another and specify that the linked feed is an alternate representation of the page. <link rel="alternate" type="application/atom+xml" href="..." /> <link rel="alternate feed" type="application/atom+xml" href="..." /> <link rel="feed alternate" type="application/atom+xml" href="..." /> This means, for instance, if the links appear on a blog home page that lists the 10 most recent entries, the feed will likely also be a listing of the 10 most recent entries. However, if the link appears on a page that shows a single entry along with a listing of comments that have been received, the link will likely point to an Atom feed listing that entry and it's various comments. Is that correct? HTML 5 defines the "feed" relation as pointing to a feed that is not necessarily an alternative representation of the page where it is found. This relation can, for instance, be used on a blog home page to point to a comments feed or category feed. <link rel="feed" type="application/atom+xml" href="..." /> What I did not see in the HTML5 spec is any indication of whether the order of link relations is significant. I'm assuming that means that it is not. I'm also assuming that means that all "alternate feed" link relations, with the same type attribute value, appearing anywhere within the document are considered to be equivalent? - James [1] http://whatwg.org/specs/web-apps/current-work/#alternate0 [2] http://whatwg.org/specs/web-apps/current-work/#feed0 Edward O'Connor wrote: > Robert Sayre wrote: >> Don't move forward with the autodiscovery draft. > [...] >> At this point there seems to be no reason for the autodiscovery draft >> to exist, since the WHAT-WG has ably covered the subject in Web >> Applications 1.0. > > I am worried that there are three simultaneous efforts to spec out feed > autodiscovery: WA1, the RSS board's recent spec, and this draft. Ideally > this stuff would get specced just once. WHAT WG seems like a neutral > ground, syndication-format wise, so perhaps they're best positioned to > spec feed autodiscovery in a way that makes everybody happy. > > > Ted >
