>> I think that only <link> should be used. All feeds linked by <a> >> should be ignored during the process of autodiscovery.
> Why? Autodiscovery should be limited to <head>...</head>. If an author wants his feeds to be discovered automatically by UAs, he should use <link>. Providing additional or same feed links using <a> is only for linking and does not affect the autodiscovery. Scanning whole document is not necessary and increases the complexity. Franklin Tse -----Original Message----- From: Lachlan Hunt [mailto:[EMAIL PROTECTED] Sent: Friday, November 24, 2006 15:45 To: Tse Shing Chi (Franklin/Whale) Cc: 'atom-syntax' Subject: Re: PaceResurrectAutodiscovery Tse Shing Chi (Franklin/Whale) wrote: >> However, since the Web Applications draft already covers all of >> these issues fairly well, I believe it is unnecessary for this >> draft to be resurrected. Instead, a few of the good ideas from >> this draft should be integrated into the WA1 spec. > > Web Applications 1.0 is new markup language based on HTML under > development. UAs that support feed autodiscovery are not necessary to > support Web Apps 1.0. Relying on the definitions in the specification > of Web Apps 1.0 is not appropriate. They don't need to support it fully, UAs can just implement the parts they need. The relationships can still be defined within it, just like many were defined in HTML4. > The reason of using "alternate" is that, "alternate" was already > defined the HTML 4.01 Specificiation and it is widely used by UAs. [ > http://www.w3.org/TR/html401/types.html#type-links ] I'm aware of the reason for choosing alternate originally and I accept that it needs to be supported like that for backwards compatibility, but that doesn't mean we can't fix the mistake. > "Authors may wish to define additional link types not described in > this specification. If they do so, they should use a profile to cite > the conventions used to define the link types. Please see the profile > attribute of the HEAD element for more details." The profile attribute isn't used in reality. Authors rarely set it, or it's set by their CMS to some default value and they don't bother to change it. Many of the tools that implement microformats don't even bother to check for the presence of the correct profile attribute, they just look for the values of the rel and class attributes. The profile attribute will most likely be dropped from HTML5. > On the other hand, the W3C Widgets 1.0 autodiscovery is also using > "alternate". [ http://www.w3.org/TR/widgets/#autodiscovery ] That's currently a first working draft and I think that's a mistake, rel="alternate" is not designed as a relationship for everything, it was designed for the specific purpose of linking to alternative representations. In the case of feeds, it sort of fits the definition, though not completely, but it certainly doesn't in the case of a widget. They should instead define something like rel="widget" because a widget is not necessarily an alternate representation of a document. > I think that only <link> should be used. All feeds linked by <a> > should be ignored during the process of autodiscovery. Why? -- Lachlan Hunt http://lachy.id.au/
