W3C home > Mailing lists > Public > public-html@w3.org > January 2010

Re: Suggestion for Microdata to RDF conversion

From: Benjamin Nowack <bnowack@semsol.com>
Date: Mon, 25 Jan 2010 10:58:33 +0100
To: "Philip Jägenstedt" <philipj@opera.com>
Cc: public-html@w3.org, Ian Hickson <ian@hixie.ch>
Message-ID: <PM-GA.20100125105833.597C2.2.1D@semsol.com>
>How does one identify an RDF class? Is there an RDF class in the below?
><div itemscope itemtype="http://microformats.org/profile/hcard">
>   My name is <span itemprop="fn">Alec Tronnick</span>
>   <img itemprop="photo" src="mypic.jpg" alt="">
What Tantek said[1]. And you spotted a bug in the current Microdata 
spec: "http://microformats.org/profile/hcard" is not the item*type*
for a vcard, it's the URI of the whole vocabulary. The itemtype value
should be "http://microformats.org/profile/hcard#vcard" to be in line
with the XMDP at "http://microformats.org/profile/hcard".


[1] http://lists.w3.org/Archives/Public/public-html/2010Jan/1142.html

>Unless you intend to whitelist common RDF classes or blacklist common  
>microdata item types, it seems to be that the above would generate
>_:n a <http://microformats.org/profile/hcard> .
>_:n <http://microformats.org/profile/fn> "Alec Tronnick" .
>_:n <http://microformats.org/profile/photo> <http://origin/mypic.jpg> .
>Which is most definitely not what you want and might be harmful if e.g.  
><http://microformats.org/profile/photo> should be used as a vocabulary  
>identifier in the future.
>Given that namespace URIs and item types alike are supposed to be opaque 
>identifiers (right?), the approach seems rather risky, and I'm surprised 
>that it could actually work with most (all?) RDF vocabularies.
>Philip Jägenstedt
>Core Developer
>Opera Software
Received on Monday, 25 January 2010 09:59:05 UTC

This archive was generated by hypermail 2.4.0 : Saturday, 9 October 2021 18:45:08 UTC