- From: Hugh Glaser <hg@ecs.soton.ac.uk>
- Date: Mon, 14 Jul 2008 12:13:47 +0100
- To: Mark Birbeck <mark.birbeck@webbackplane.com>
- CC: "public-lod@w3.org" <public-lod@w3.org>, "semantic-web@w3.org" <semantic-web@w3.org>
On 14/07/2008 10:42, "Mark Birbeck" <mark.birbeck@webbackplane.com> wrote: [snip] > > >> And if we are talking about an RDF browser (as our pages are, albeit with a >> clean URI >> that doesn't have the browser URI in it), getting it to include the RDF as >> RDFa or whatever >> is even stranger; after all >> http://demo.openlinksw.com/rdfbrowser2/?uri%5B%5D=http%3A%2F%2Fos.rkbexplorer >> .com%2Fdescription%2Fosr7000000000017765 >> doesn't include the substantive RDF as RDFa, (or have a link rel to >> http://os.rkbexplorer.com/data/osr7000000000017765 for that matter) which >> would be the >> equivalent. > > I can't comment on that example, but ultimately there is no need for a > URL for an HTML+RDFa page to be any different to a normal one. > > (Although I might have missed your point, here....) Not sure. I think it relates to the question Tom is asking. Another way of putting it is that we don't expect Tabulator to include RDFa of the RDF we are currently browsing (I think - or maybe we should?). So I think it comes down the purpose of the exercise, and there is no one size fits all. If I have some web pages, then being able to simply embed RDF (by whatever means), or link to associated RDF resources, is really great. Good for convenience, maintenance. On the other hand, if I am trying to simply publish RDF, for example out of a DB, then things are a bit different. To help people who want to build agents that use it I might expect them to be able to visualise what I am publishing by using Tabulator or other tools to browse it generically. Or equivalently I might provide some human readable pages of the RDF to make my data more easily accessible. Because it is a bespoke browser, then I need to add rel link to the RDF URI. And then someone suggests I should add RDFa or something more to these pages. Before I know it, I have invested a lot of time providing (and maintaining) neatly formatted and RDF-friendly web pages to publish my DB, when all I wanted to do was join the Semantic Web by publishing my RDF. In fact, one of the options when Kingsley asked for the rel link was to simply remove all the html description pages, getting back to the core of what I want to do, which is publish RDF as Linked Data. Best Hugh
Received on Monday, 14 July 2008 11:14:58 UTC