sbp: "This email sketches class of documents I call RDF Stylesheets, of which one immediate application is making GRDDL easier to deploy via a small extension. This is a lightweight proposal with a big impact."
sbp: I mean specifying the styling in RDF, not styling RDF itself. Probably should've made the subject more specific—dang!
sbp: Anyway, this is something quite significant that's been brewing for a while. I discussed it with danja several weeks ago and he suggested I send it to the GRDDL WG (which I have done too), and this is the result of the discussion though taken on in leaps and bounds.
sbp: The summary of the benefit it'll provide to GRDDL is a) you don't have to use the @profile attribute, and b) it reduces the number of <link> elements you need to use, even, potentially to one. It makes mixing and matching transformation easier. It makes adding metadata about such unions of transformations easier. It is, in other words, an RDF Stylesheet Languages proposal!
sbp: It also plays nicely with non-XML, which I think might turn out to be a big plus potentially even in the GRDDL world.
sbp: READ THIS NOW! Get in on the ground floor! You're missing out by continuing to read even this sentence! :-)
sbp: I just realised that you can add s:stylesheet to this, to allow you to link from stylesheets within stylesheets, and also formalise the rel="stylesheet" link. It seems true, furthermore, that s:stylesheet owl:inverseOf s:input, which is neat. Note that s:input is a bit geared towards the imperative naming style, rather than the declarative naming style that's more common in RDF. Perhaps it could be called something else (better?)?
sbp: Furthermore, I suppose that grddl:transformation rdfs:subPropertyOf s:stylesheet. The sub property being necessary because it implies the chaining mechanism... gets a bit tricky around here.
sbp: Ah, how about s:styles rather than s:input?
sbp: Whoops, scratch a lot of that. The domain of grddl:transformation is an XPath node... need to work on this a bit.
sbp: Okay, I just sent a huge followup which considers some of the things I just jotted here in detail.
sbp: I mean specifying the styling in RDF, not styling RDF itself. Probably should've made the subject more specific—dang!
sbp: Anyway, this is something quite significant that's been brewing for a while. I discussed it with danja several weeks ago and he suggested I send it to the GRDDL WG (which I have done too), and this is the result of the discussion though taken on in leaps and bounds.
sbp: The summary of the benefit it'll provide to GRDDL is a) you don't have to use the @profile attribute, and b) it reduces the number of <link> elements you need to use, even, potentially to one. It makes mixing and matching transformation easier. It makes adding metadata about such unions of transformations easier. It is, in other words, an RDF Stylesheet Languages proposal!
sbp: It also plays nicely with non-XML, which I think might turn out to be a big plus potentially even in the GRDDL world.
sbp: READ THIS NOW! Get in on the ground floor! You're missing out by continuing to read even this sentence! :-)
sbp: I just realised that you can add s:stylesheet to this, to allow you to link from stylesheets within stylesheets, and also formalise the rel="stylesheet" link. It seems true, furthermore, that s:stylesheet owl:inverseOf s:input, which is neat. Note that s:input is a bit geared towards the imperative naming style, rather than the declarative naming style that's more common in RDF. Perhaps it could be called something else (better?)?
sbp: Furthermore, I suppose that grddl:transformation rdfs:subPropertyOf s:stylesheet. The sub property being necessary because it implies the chaining mechanism... gets a bit tricky around here.
sbp: Ah, how about s:styles rather than s:input?
sbp: Whoops, scratch a lot of that. The domain of grddl:transformation is an XPath node... need to work on this a bit.
sbp: Okay, I just sent a huge followup which considers some of the things I just jotted here in detail.
Wikier: the most important that it already exports reports in RDF
Wikier: using EARL
Wikier: see an example
DanC: ooh... EARL output from RDF schema best practice checker. sounds cool.
Wikier: thx, feedback is welcomed
Wikier: DanC: it internally works with a RDF model, and reports are exported to XHTML or RDF/XML
kidehen: DBpedia presentation
kidehen: From ISWC2007, Busan
Wikier: using EARL
Wikier: see an example
DanC: ooh... EARL output from RDF schema best practice checker. sounds cool.
Wikier: thx, feedback is welcomed
Wikier: DanC: it internally works with a RDF model, and reports are exported to XHTML or RDF/XML
kidehen: DBpedia presentation
kidehen: From ISWC2007, Busan