- From: Dominique Hazael-Massieux <dom@w3.org>
- Date: Tue, 05 Sep 2006 11:49:17 +0200
- To: Danny Ayers <danny.ayers@gmail.com>
- Cc: Ian Davis <ian.davis@talis.com>, public-grddl-wg@w3.org
Le mardi 05 septembre 2006 à 11:29 +0200, Danny Ayers a écrit : > On 9/5/06, Ian Davis <ian.davis@talis.com> wrote: > It also allows > > to describe profiles identified by fragments of the URI of the given > > document, in an elements with a class attribute set to subProfile, > > containing a link to the said fragment with a rel attribute set to > > profile, and another link with a profileTransformation rel as described > > above." > > Would I be right in thinking the intention here is to specify > different transformations for different sections of the doc? No; the idea is that if you a profile à la: http://example.org/foo#bar that you can put up a GRDDL-able HTML page at http://example.org/foo that links http://example.org/foo#bar to a profile transformation. To create such a link, you would have a section à la: <div id="bar">Link from <a rel="subProfile" href="#bar">a subprofile</a> to its <a href="/foobar-profile-transform" rel="profileTransformation">associated transformation</a></div> More to the point, this was needed to integrate the existing mechanism that linked the W3C home page to the DC vocabulary using the following profile: http://www.w3.org/2000/08/w3c-synd/#meta (as it appears, this isn't used correctly any more on the home page, but that's a different topic) I.e., using directly GRDDL doesn't allow you to bind a profile using a fragment URI to a profile transformation, so I came up with an additional abstraction layer to create that connection. But this abstraction layer is itself entirely covered by GRDDL, so it doesn't need to be defined by the specification at all (it certainly could, but it seems such a narrow use case that I think there is no value in making the spec more complex only for this) Dom
Received on Tuesday, 5 September 2006 09:51:15 UTC