Martin, all, I wrote a blog post summarizing the challenges of using multiple vocabularies in microdata, which includes Martin's snippet as well as another snippet I was given on #whatwg earlier last week.
http://openspring.net/blog/2011/06/10/microdata-multiple-vocabularies Steph. On Fri, Jun 10, 2011 at 6:57 PM, Stéphane Corlosquet <scorlosq...@gmail.com>wrote: > Hi Martin, > > On Fri, Jun 10, 2011 at 6:08 PM, Martin Hepp < > martin.h...@ebusiness-unibw.org> wrote: > >> Dear all: >> >> Until today, I had assumed that one limitation of Microdata was that it >> did not support more than one class per item, e.g. that you could not state >> that something was e.g. the intersection of >> >> http://www.productontology.org/id/Hammer >> and >> http://purl.org/goodrelations/v1#Individual >> >> which, in the context of GoodRelations, means that it is an actual hammer >> (like in describing antiques or other unique items). >> >> Now, reconsidering the issue, I am no longer convinced that this is valid >> criticism, because you could use the full URI for rdf:type with itemprop: >> >> <div itemscope itemtype="http://www.foo.com/Type1" itemid=" >> http://acme.org/things#1> >> <a itemprop="http://www.w3.org/1999/02/22-rdf-syntax-ns#type" href=" >> http://www.foo.com/Type2"></a> >> </div> >> >> >> This should result in >> >> <http://acme.org/things#1> a <http://www.foo.com/Type1>, < >> http://www.foo.com/Type2> . >> >> or am I mistaken? >> > > I believe you are correct, and according to the Live Microdata tool, it > yields the expected results. here is the link (click on the turtle tab): > http://j.mp/iUH2FS > > Steph. > > >> Best >> Martin Hepp >> >> >> >