W3C home > Mailing lists > Public > public-vocabs@w3.org > September 2012

RE: new itemscope or not?

From: Michael Hopwood <michael@editeur.org>
Date: Mon, 10 Sep 2012 11:31:20 +0100
To: Cord Wiljes <cwiljes@cit-ec.uni-bielefeld.de>, "public-vocabs@w3.org" <public-vocabs@w3.org>
Message-ID: <F61A8945B05715448AF2221FB6080925072E85CEC0@EX27MAIL03.msghub.com>

This is the URI / URL / URN discussion again.

>>> Or in other words: "url" means something rather general: "There is a web document related to the resource that can be retrieved at this url."

The opposite, it means something very specific: "this is the Web location of" _____ (where ______ is some network addressable digital file).

A URI is just a Web syntax used to identify stuff because the HTTP syntax conveniently allows you

a) control over the lookup so you can put some description there (as you said);

b) uniqueness also because of a) and the way the HTTP is set up.

HTTP URI is paradoxically a very general protocol with very limited application ;)

What is lacking is the infrastructure to ensure persistence, quality of lookup and a lot of other things. Most URNs in practical application have those but without the nice HTTP aspect, but in principle they can get it as an added service (e.g. ISBN-A).

DOI is an attempt to add all those features together an generalise to the entire Web.



-----Original Message-----
From: Cord Wiljes [mailto:cwiljes@cit-ec.uni-bielefeld.de] 
Sent: 10 September 2012 11:17
To: public-vocabs@w3.org
Subject: Re: new itemscope or not?

The schema.org specification seems to support Jeff's interpretation of the property "url" as "the WWW-address where an electronic copy of the thing that s described can be downloaded". From

Property: url
Expected Type: URL
Description: URL of the item.

Only something that can be downloaded (an information resource) can have a URL. So schema.org's property "url" should only be available for "CreativeWork", not for "Thing" as it is right now. A person for example can't have a url. A person can have a website (which is an information
resource) and this website has an url. But then I cannot find any property like "website" or "homepage" for any of schema.org's classes. 
Combined with the fact that "url" is avalable for class "Thing" (i.e. 
for everything) I suppose that "url" is in fact used ambiguosly:

A book can have a url where you can download the book's text.
A person can have a url where you find information about this person.

Or in other words: "url" means something rather general: "There is a web document related to the resource that can be retrieved at this url." 
Essentially its just a "see also" to a document on the web.


Am 08.09.2012 04:14, schrieb Young,Jeff (OR):
> If I was Godz, I would NOT assume they are the same thing. I would use 
> schema:url thusly for those decreasingly rare situations where 
> somebody (especially a remote observer) wants to describe something 
> that is honest-to-godz located on the Web. For example:
> @prefix observer: <http://example.org/observer/> .
> observer:12345 a <http://purl.org/library/Thesis>;
> 	schema:name "Architectural Styles and the Design of Network-based 
> Software Architectures";
> 	schema:author <http://viaf.org/viaf/26681119>;
> 	schema:url
> <http://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm>.
> As a matter of principle, Roy's (HTML) thesis COULD be upgraded to be 
> self-describing with some hidden markup (either RDFa 1.1 or Microdata) 
> and a trivial Apache rewrite (303 redirect) upgrade to www.ics.ici.edu 
> to replace the observer URI.
> OTOH, if somebody decides that schema:url should be treated the same 
> as "itemid" (Microdata), "resource" (RDFa Lite 1.1), "rdf:about" 
> (RDF/XML), etc. then schema:url is a wasted opportunity and we (i.e. 
> the pedantic observers of reality) would need to find a new vocabulary 
> term fill this void.
> Jeff
>> -----Original Message-----
>> From: Jeni Tennison [mailto:jeni@jenitennison.com]
>> Sent: Friday, September 07, 2012 3:15 PM
>> To: Ed Summers
>> Cc: Dawson, Laura; Thad Guidry; public-vocabs@w3.org
>> Subject: Re: new itemscope or not?
>> On 7 Sep 2012, at 20:03, Ed Summers wrote:
>>> It would be interesting to know if the HTML spec allowed multiple 
>>> identifiers, similar to how other HTML attributes work:
>> "The itemid attribute, if specified, must have a value that is a 
>> valid URL potentially surrounded by spaces."
>> http://www.w3.org/TR/microdata/#attr-itemid
>> So that would be 'no', not according to spec.
>> I've often wondered whether the schema.org 'url' property is meant to 
>> be synonymous with itemid. I'm not sure what happens in schema.org 
>> interpreters when you specify one/other/both/multiple urls...
>> Jeni
>> --
>> Jeni Tennison
>> http://www.jenitennison.com

Cord Wiljes
Semantic Computing Group
Excellence Cluster - Cognitive Interaction Technology (CITEC) Bielefeld University

Phone: +49 521 106 12036
Mail: cwiljes@cit-ec.uni-bielefeld.de
WWW: http://www.sc.cit-ec.uni-bielefeld.de/people/wiljes

Room H-123
Morgenbreede 39
33615 Bielefeld
Received on Monday, 10 September 2012 10:32:14 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:29:25 UTC