20.06.2012, 01:05, "Stéphane Corlosquet" <scorlosquet@gmail.com>:

On Tue, Jun 19, 2012 at 9:53 PM, Egor Antonov <elderos@yandex-team.ru> wrote:
Well, validators developers will not be very happy with this change.
However, if it happens, I prefer 'type' name, because 'additionalType' implies there are other, non-additional ones.
I disagree. 'type' would be misleading and imply that all types are set by this property, when it is not the case.
Why not? I don't see any semantic difference. Syntactic - yes.
Also I don't see any reason, why this
<div itemscope>
    <link itemprop="type" content="http://mytype.type">
must differ from
<div itemscope itemtype="http://mytype.type"/>
It can be some differences while disambiguating property names, but schema.org property names are unique.
When a single type is specified, there is no problem at all.
Microdata and RDFa have simpler ways of setting the types of a data item. This new property should be used sparingly and for cases where the default microdata itemtype attribute is not sufficient. Note that it does not make much sense to be used in RDFa since mixing vocabularies for types is natively supported.
And on semantic level there must be no difference between object types.
19.06.2012, 22:37, "Martin Hepp" <martin.hepp@ebusiness-unibw.org>:

Hi Dan,
Thanks for summarizing this. Two suggestions:

1. Instead of samePropertyAs: rdf:type, use owl:equivalentProperty if you want to express how the additional type information can be mapped into an RDF world.

2. Add a link to the thread in the W3C archive so that the full details of the discussions are easy to reach:

By the way, the original proposal now being favored was already described in my post


and presented at the schema.org Workshop back in September 2011 (slide 16):




On Jun 19, 2012, at 7:24 PM, Dan Brickley wrote:

 Thanks everyone. Lots of mail!

 I have tried to make a brief summary of some of the points in the Web
 Schemas wiki, just a sketch of individual positions really rather than
 a summary of the whole debate. I also started there to write up
 details of a concrete proposal for 'additionalType'.

 http://www.w3.org/wiki/WebSchemas/additionalTypeProposal ...

 Since nobody has volunteered to lead an effort to get Microdata syntax
 changed to support multiple types from different vocabularies, and on
 balance after reading thru all the debate, I think we should go for
 the new property approach.

 I'd like to make sure that we capture all the concerns people have in
 the Wiki and in the resulting property definition, and to give some
 thought to how validators and checkers ought to behave.

 Peter, Egor, others, ... can you live with a new property here? (one
 of 'additionalType' or 'type') Any preferences on name?



martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp@ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
* Project Main Page: http://purl.org/goodrelations/

Egor Antonov