W3C home > Mailing lists > Public > public-vocabs@w3.org > April 2014

Re: Socialnetworks of a person or organization

From: Jarno van Driel <jarno@quantumspork.nl>
Date: Sat, 12 Apr 2014 17:46:00 +0200
Message-ID: <CAFQgrbZO7DcwmP-+tuG=Zuv6kjpyXBpoT-Rq4wiguoKrT5SXNA@mail.gmail.com>
To: Kingsley Idehen <kidehen@openlinksw.com>
Cc: Public Vocabs <public-vocabs@w3.org>
I'm so glad semantics can help me understand and express things. But which
of the below properties helps me say: "this is my Youtube page, here is
where I share the useless stuff I like"?

1. describedBy
2. profileAt
3. hasProfileDoc
4. identifiedBy
5. referencedBy
6. subjectOf

Or none of these and and do we need even more properties? And who is going
to explain this 'simple' construction to the same people who are already
struggling to create proper HTML? Or isn't that needed since it all is so

I apologize for my cynicism up front but it originates from the fact that
for me - the simple mark up guy in this group - this entire discussion is
becoming way more complex than is needed, or at least that's how I see it.

Please trust me when I say that the average webmaster who tries to markup a
Product already screws that up 7/10 times. I know because I clean their
mess up on a daily basis. When a developer looks at the schema.org list
he/she spends about 0-5 seconds looking at the description of a
property/Item, if you're lucky. Webmasters aren't interested in in looking
through a gazillion options. Let alone that they worry about which property
is more correct to laws of grammar and semantics. Really they couldn't care
less, although I completely understand most of you in this group look at
this differently.

But when a developer looks for a property to link an Organization or Person
to a social media (or inversed) he/she looks at the description and chooses
the one that says: social media/facebook.G+/whatever. Without having any
debate if the property is named correct. Heck you might even call it
'abcde', if the description is right, they'll use is. This really is the
everyday reality.

The way a webmaster looks at it:
@url > good description - let's use that for the website
@sameAs > good description - let's use that for Freebase and Wikipedia
@abcde > good description - let's use that for any social media account

Maybe add the inverse property of these as well, and from a developer's
standpoint we're done. The average John-Doe doesn't spend a second longer
on it, seriously. because they don't care, they have other important things
to do. (other things? Yes, other things)

Now I completely understand that for some situations this doesn't cut it.
Especially when you look at the large corporations, but sorry, the large
majority of sites are small simple and most of the time messy. Can we at
least get a quick and simple solution for those? Then you can discuss how
to have the 'special cases' fit in here as well, for as long as you want.
Please stop turning a mundane question into another 100+ post thread.

On Sat, Apr 12, 2014 at 3:43 PM, Kingsley Idehen <kidehen@openlinksw.com>wrote:

>  On 4/11/14 7:50 PM, Aaron Bradley wrote:
> I think Thad typifies the difference between the "regular web" and the
> "social web" (and it is indeed an important one for marketers) well when he
> says:
>  > The difference between the 2 is that one has the context of "allows a
> communication pathway to an Organization or Person"...versus those that are
> not constructed to really have communication to a Organization or Person".
>  Or - as I might have mentioned before - the "regular web" references
> resources *about* an entity, whereas the social web references resources
> that emanate *from* an entity.
> It is simpler, really.
> The Web provides foundation for many kinds of network abstractions
> [1][2][3]. Each of this abstractions is fundamentally driven by denotation
> relations.
>   In regard to the referenced entity the former is passive, the latter
> active (or at least  potentially so) - it's the difference between a third
> and person narrative.
>  Is the Wikipedia page *about *Monsanto in the same category as the
> Twitter account run *by* Monsanto?  I sure don't think so, and I think
> that its useful for data consumers to be able to distinguish between these
> two classes of identifiers when returning information about the entity in
> question.
> It just about which abstraction you are working with. We only get into
> trouble when we mangle the abstractions :-)
> BTW -- Dan's example provide good foundation for understanding the
> matters. It provides a bridge solution that ultimately aids in
> understanding important lines of demarcation. For instance, the role of a
> *pronoun* in natural language is subtle but powerful.
> Sticking with Dan's example, here are some additional relation name
> suggestions, as alternatives to anything with "sameAs" in it:
> 1. describedBy -- your Twitter Home page Identifies you since its
> comprised of a Description of you
> 2. profileAt -- your Twitter home page is a document at a location denoted
> by its HTTP(S) URL
> 3. hasProfileDoc -- variant of the above.
> The above are in addition to:
> 1. identifiedBy
> 2. referencedBy
> 3. subjectOf .
> All of these make Dan's example a little clearer, as alternatives to
> "sameAs" .
> Again, there is a subtlety in Dan's example that really needs to be picked
> up, it solves a major headache re. bridging the world of:
> 1. structured data
> 2. linked data
> 3. semantically enhanced linked data.
> Links:
> [1] http://bit.ly/1cjYwqN -- Internet abstraction (driven by DNS names
> for computers (actually their NICS) re. entity denotation and scope)
> [2] http://bit.ly/1e0pwdI -- Linked Document (or World Wide Web 1.0)
> abstraction (driven by HTTP names for Web Documents)
> [3] http://bit.ly/INv6ag -- Linked Data (or Data Web) abstraction driven
> by HTTP names for anything
> [4] http://bit.ly/1lHD31h -- Semantic Web abstraction (driven by names
> for anything combined with machine comprehensible relation semantics)
> On Fri, Apr 11, 2014 at 1:58 PM, Kingsley Idehen <kidehen@openlinksw.com>wrote:
>>  On 4/11/14 4:06 PM, Jarno van Driel wrote:
>> Being a non-illuminati I think simple. The description of sameAs mentions
>> about the item's identity. Now for me my 'identity' isn't defined by a
>> Youtube channel where I share random stuff I like on the web. I am no
>> @VideoGallery, I'm me, a real life person and not a collection of videos.
>>  Dan's example in HTML+Microdata (which by notation choice
>> **inadvertently** blurs visibility of the relation semantics in play) :
>> <div itemscope itemtype="http://schema.org/Person"<http://schema.org/Person>>
>> <span itemprop="name">Stephen Fry</span>
>>     (<a itemprop="url" href="http://www.stephenfry.com/"<http://www.stephenfry.com/>
>> >stephenfry.com</a>,
>>       <a itemprop="sameAs" href="http://twitter.com/stephenfry"<http://twitter.com/stephenfry>>twitter</a>,
>>  <a itemprop="sameAs"
>> href="http://en.wikipedia.org/wiki/Stephen_Fry"<http://en.wikipedia.org/wiki/Stephen_Fry>>wikipedia</a>)
>> </div>
>>  Turtle  translation:
>> <> <http://www.w3.org/ns/md#item> <http://www.w3.org/ns/md#item> [
>>        <http://www.w3.org/1999/02/22-rdf-syntax-ns#type><http://www.w3.org/1999/02/22-rdf-syntax-ns#type>
>> <http://schema.org/Person> <http://schema.org/Person>;
>>        <http://schema.org/name> <http://schema.org/name> "Stephen Fry";
>>        <http://schema.org/sameAs> <http://schema.org/sameAs>
>> <http://twitter.com/stephenfry> <http://twitter.com/stephenfry>,
>>        <http://en.wikipedia.org/wiki/Stephen_Fry><http://en.wikipedia.org/wiki/Stephen_Fry>;
>>        <http://schema.org/url> <http://schema.org/url>
>> <http://www.stephenfry.com/> <http://www.stephenfry.com/>
>>      ];
>> <http://www.w3.org/ns/rdfa#usesVocabulary><http://www.w3.org/ns/rdfa#usesVocabulary>
>> <http://schema.org/> <http://schema.org/> .
>> What does Dan's example demonstrate?
>> The function of a **pronoun** in a sentence or statement. Basically, the
>> example makes the following claim, using terms from <http://schema.org/><http://schema.org/>(a Vocabulary):
>> Someone or something has determined the existence of an entity that has
>> the following discernible attributes:
>> Name: "Stephen Fry"
>> Type: Person
>> referencedBy: <http://twitter.com/stephenfry><http://twitter.com/stephenfry>,
>> <http://twitter.com/stephenfry> <http://twitter.com/stephenfry>,
>> <http://www.stephenfry.com/> <http://www.stephenfry.com/> .
>> Personally, I wouldn't denote a relationship predicate/property for this
>> relation, in this manner, due to the **equivalence** intuition.
>> Alternatives inclued:
>> 1. referencedBy
>> 2. subjectOf
>> 3. identifiedBy -- this is my personal favorite .
>> --
>> Regards,
>> Kingsley Idehen	
>> Founder & CEO
>> OpenLink Software
>> Company Web: http://www.openlinksw.com
>> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
>> Twitter Profile: https://twitter.com/kidehen
>> Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
>> LinkedIn Profile: http://www.linkedin.com/in/kidehen
> --
> Regards,
> Kingsley Idehen	
> Founder & CEO
> OpenLink Software
> Company Web: http://www.openlinksw.com
> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
> Twitter Profile: https://twitter.com/kidehen
> Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
> LinkedIn Profile: http://www.linkedin.com/in/kidehen
Received on Saturday, 12 April 2014 15:46:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:29:39 UTC