W3C home > Mailing lists > Public > public-vocabs@w3.org > November 2011

Re: Canonicalizing relative URLs seen in URL type properties?

From: Philip Jägenstedt <philipj@opera.com>
Date: Tue, 01 Nov 2011 22:21:28 +0100
To: public-vocabs@w3.org
Message-ID: <op.v4adl2wqsr6mfa@localhost.localdomain>
On Tue, 01 Nov 2011 18:25:50 +0100, Martin Hepp  
<martin.hepp@ebusiness-unibw.org> wrote:

> Hi Philip:
> afaik,
>    http://schema.org/InStock
> is not an itemtype but the identifier of an individual representing a  
> value of the type
>    http://schema.org/ItemAvailability
> So I am unsure on how to interpret your statement:
>> Note of caution: the following is about URLs in itemtype, while the  
>> original thread is about URLs in itemprop. They are not the same.
> http://schema.org/InStock will only be used in patterns like
> <link itemprop="availability" href="http://schema.org/InStock"/>

Huh, that looks like bad design on the part of schema.org, as they could  
have just defined it to be <meta itemprop="availability" href="instock"/>,  
thereby saving both typing and the confusion of http://schema.org/InStock  
looking like a type but actually being an enumeration value.

> You also said that any client doing any kind of lax handling of this  
> identifier would be "non-compliant"; however, I am sure that at least  
> some major search engines will eventually tolerate
> <link itemprop="availability" href="http://schema.org/instock"/>
> You can see in Google's testing tool that they internally normalize all  
> itemprop URIs to lower case.

It is completely up to the vocabulary how to interpret the property  
values, so this is not violation of microdata. However, it is a huge bug  
in the schema.org vocabulary that it doesn't define the processing for  
consumers, and this is just one example of that.

Philip Jägenstedt
Core Developer
Opera Software
Received on Tuesday, 1 November 2011 21:22:08 UTC

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