W3C home > Mailing lists > Public > public-schemaorg@w3.org > June 2016

Re: publisher field for Blogposts and websites

From: Aaron Bradley <aaranged@gmail.com>
Date: Tue, 7 Jun 2016 14:27:20 -0700
Message-ID: <CAMbipBsp6AX0RV1CFUAvPyPOcrUC2MOqyWSuvRrOk5vudtR6ug@mail.gmail.com>
To: Richard Wallis <richard.wallis@dataliberate.com>
Cc: Elias Kaerle <elias.kaerle@sti2.at>, Bäck, Gerald <gerald@baeck.at>, "schema.org Mailing List" <public-schemaorg@w3.org>
"Broadening out the question of the possibility of a publisher being a
Person or an Organization, to any CreativeWork, that does in this age of
self-publishing have something [i]n its favour."

+1 to this.  The requirements of specific data consumers entirely aside,
one more than one occasion having Organization as the sole expected type
has either struck me as limiting, or *has *been limiting.  To cite the most
obvious use case, the publisher of a single-author blog is almost always
the Person who is that single author, and it's limiting not to be able to
declare that without either reverting to a text string or using an
unexpected type.


On Tue, Jun 7, 2016 at 6:43 AM, Richard Wallis <
richard.wallis@dataliberate.com> wrote:

> From a Schema.org vocabulary point of view no properties are deemed to be
> required.
>
> In the case of the Google SDTT complaining about missing fields it is
> advising you on *their* requirements for displaying information about
> organisations (e.g.. asking for a logo) etc. Questions regarding the needs
> should be addressed to their developer mailing lists.
>
> This list is inly for discussions regarding the vocabulary itself.
>
> In the particular circumstance you describe, I would probably not have
> applied a publisher to individual BlogPostings for which an author would
> suffice.  However I would have associated each post as being ‘partOf’ a
> Blog which optionally would have a ‘publisher’ reference.
>
> Broadening out the question of the possibility of a publisher being a
> Person or an Organization, to any CreativeWork, that does in this age of
> self-publishing have something n its favour.
>
> ~Richard.
>
>
>
> Richard Wallis
> Founder, Data Liberate
> http://dataliberate.com
> Linkedin: http://www.linkedin.com/in/richardwallis
> Twitter: @rjw
>
> On 7 June 2016 at 11:44, Elias Kaerle <elias.kaerle@sti2.at> wrote:
>
>> Hi Gerald,
>>
>> I agree. Another solution could be to simply accept schema:Person and
>> schema:Organization as publisher.
>>
>> Maybe one of the people maintaining schema.org can comment on that
>> issue!?
>>
>> Best, Elias
>>
>>
>> On 07.06.2016 10:59, Bäck, Gerald wrote:
>> > Hi Elias,
>> >
>> > the interesting thing is, if you put a logo field into the person
>> entity,
>> > google validator claims that a logo field is not valid within the person
>> > entity:) The conclusion is that persons cannot be publishers, which is
>> > simply wrong.
>> >
>> > My proposal is to get rid of the publisher entity as a requirement,
>> because
>> > blogposts and websites still need an author which should be enough for
>> > private run blogs.
>> >
>> > best wishes, Gerald
>> >
>> >
>> >
>> >
>> > ----
>> > DI Gerald Bäck | fb <https://facebook.com/geraldbaeck> | blog
>> > <http://www.baeck.at/> | devblog <http://dev.baeck.at> | fitblog
>> > <http://fitness.baeck.at> | +43 664 5107761 <+436645107761>
>> >
>> > On Tue, Jun 7, 2016 at 9:44 AM, Elias Kaerle <elias.kaerle@sti2.at>
>> wrote:
>> >
>> >> Hi Gerald,
>> >>
>> >> this is indeed a strange behaviour. I would blame it on the way
>> Google's
>> >> structured data testing tool works: it does, as far as i know, not
>> >> necessarily validate/verify annotations strictly the way schema.org
>> >> defines them, but more in a way they need the annotations for feeding
>> >> their Rich Snippets and Rich Cards.
>> >>
>> >> So I would assume Google doesn't care about having a schema:Person as a
>> >> publisher, but requires a logo (or some kind of picture) to process a
>> >> beautiful Rich Snippet/Rich Card out of it.
>> >>
>> >> Best, Elias
>> >>
>> >> On 07.06.2016 08:36, Bäck, Gerald wrote:
>> >>> Hi,
>> >>>
>> >>> I am currently doing my first steps with JSON-LD and try it on my
>> private
>> >>> blog. As far as I understand Blogposts do require a publisher field,
>> >> which
>> >>> can only be an organisation. But I think it should be possible for
>> >> persons
>> >>> to be publishers too, but I also would like to question, that
>> blogposts
>> >> or
>> >>> even Websites do need a publisher field at all.
>> >>>
>> >>> I tested my blog with Google's Structured Data Testing Tool.
>> >> Interestingly
>> >>> enough the tool did not complain about the publisher being a Person,
>> but
>> >>> that the publisher entitiy had no logo, which on the other hand is not
>> >>> allowed as a field for a person.
>> >>>
>> >>>
>> >>
>> https://search.google.com/structured-data/testing-tool#url=http%3A%2F%2Fwww.baeck.at%2Fblog%2F2016%2F05%2F30%2FWahlmanipulationen%2F
>> >>>
>> >>> I also tested the root of my blog, which is defined as website, also
>> with
>> >>> myself as a publisher person. This time the tool was fine with it.
>> >>>
>> >>>
>> >>
>> https://search.google.com/structured-data/testing-tool#url=http%3A%2F%2Fwww.baeck.at%2F
>> >>>
>> >>> thx, Gerald
>> >>>
>> >>>
>> >>>
>> >>> ----
>> >>> DI Gerald Bäck | fb <https://facebook.com/geraldbaeck> | blog
>> >>> <http://www.baeck.at/> | devblog <http://dev.baeck.at> | fitblog
>> >>> <http://fitness.baeck.at> | +43 664 5107761 <+436645107761>
>> >>>
>> >>
>> >> --
>> >> Elias Kärle, MSc
>> >> Semantic Technology Institute
>> >> University of Innsbruck
>> >>
>> >> ICT - Technologie Park Innsbruck
>> >> 2nd Floor, Room 3S02
>> >> Technikerstrasse, 21a
>> >> 6020 Innsbruck
>> >> Austria
>> >>
>> >> Tel.: (+43) 512 507 53738
>> >> Skype: elias.kaerle
>> >>
>> >>
>> >
>>
>> --
>> Elias Kärle, MSc
>> Semantic Technology Institute
>> University of Innsbruck
>>
>> ICT - Technologie Park Innsbruck
>> 2nd Floor, Room 3S02
>> Technikerstrasse, 21a
>> 6020 Innsbruck
>> Austria
>>
>> Tel.: (+43) 512 507 53738
>> Skype: elias.kaerle
>>
>>
>
Received on Tuesday, 7 June 2016 21:27:50 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 7 June 2016 21:27:51 UTC