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

Re: Socialnetworks of a person or organization

From: Kingsley Idehen <kidehen@openlinksw.com>
Date: Sun, 13 Apr 2014 22:06:13 -0400
Message-ID: <534B4295.9030207@openlinksw.com>
To: public-vocabs@w3.org
On 4/13/14 8:39 PM, Kingsley Idehen wrote:
> On 4/13/14 6:27 PM, Thad Guidry wrote:
>> Kingsley,
>>
>> Your not helping solve the problem we have at this moment.
>>
>> This is the problem that we are trying to solve.  We have a property 
>> for describing a Thing's alternative identity online. For that we use 
>> "sameAs", everyone knows that.  Stop this madness.
>>
>> And your probably misunderstanding the problem.  Let me try it this 
>> way...
>>
>> You are assigned a task to teach a computer which of the following 
>> URI's you can use to communicate with an entity, However, there are 
>> rules you must abide by, which are that you can only use 1 Schema.org 
>> Type and only 1 Property from that Type:
>>
>> A: http://www.freebase.com/user/thadguidry
>>
>> and
>>
>> B: https://plus.google.com/+ThadGuidry
>>
>> where A does not allow communication with me...and B does allow 
>> communication with me.
>>
>> The computer must know from a Schema.org property assignment of your 
>> choice that B allows for communication with me.
>>
>> Which Schema.org Type and Property would you use for B and describe 
>> why your answer would make simple sense to most Web Developers.
>
> Here is the simple answer: schema:webid .
>
> Why? Because "identity" oriented relations infer (as per this case) 
> that a sign signifies something. "sameAs" relations assert 
> equivalence. "You" are not the same as the sign that signifies "You" 
> .  You are != "Thad Guidry" that's a literal identifier that 
> "signifies" or "denotes" the entity "You" i.e., the entity "You" is 
> "denoted" or "signified" by the the literal sign "Thad Guidry" .
>
> Identification Documents like your passport, drivers license etc.., 
> "Identify" you based on the claims (statements or triples) that 
> constitute the aforementioned document in conjunction with 
> verification and authentication processes and protocols.
>
>
> Please keep the tone civil. I am happy to make points in a debate 
> towards clarity, but I really need it to be civil, please.
>

Seeing that schema:sameAs exist [1]. While owl:sameAs [2] also exists, 
there is nothing gained, but confusion if we stick with the notion that 
schema:sameAs is a relationship property defined as follows:

"URL of a reference Web page that unambiguously indicates the item's 
identity. E.g. the URL of the item's Wikipedia page, Freebase page, or 
official website."

That's the same as saying:

"This relationship property indicates an association between an item's 
identity and an Identity Card (e.g,  Driver's license, Passport, Social 
Security Card, Credit Card etc.)" .

Breaking down Identity, Identifier, Identification, and Identity 
Verification (Authentication):

1. Identity -- nebulous
2. Identifier -- a sign that signifies, denotes, indicates and identity 
etc.. (e.g., Passport Number, Driver's License Number etc...)
3. Identification -- a collection of attribute=value pairs that coalesce 
around an identifier for documenting identity claims (e.g., Name="Your 
Name" ; Address="Your Home Address" ; etc..
4. Identity Card -- a document comprised of identity claims
5. Authentication and Verification -- how Identity claims are verified 
or authenticated.

What about relationship properties for Account association?

schema:hasAccount should suffice. Here's an example the encompasses all 
the points I am trying to make about re. subtle tweak to Dan's example 
(i.e., not using "sameAs"), that ultimately achieves many goals without 
breaking anything or imposing any discomfort on any one i.e, the bridge 
nature of schema.org stays intact.

<#i>
<http://schema.org/name> "Kingsley Uyi Idehen" ;
<http://schema.org/hasAccount> <http://twitter.com/kidehen#account> ;
<http://schema.org/hasWebID> <http://twitter.com/kidehen#identity> ;
<http://schema.org/hasPublicKey> 
"MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEA3cxE/5GcvWYdDJ4+Px5QmTLbz5ar
WvgCylRV8MjZ35MIjpdi+/TZLRHDONbMnXFcpe7Cn1xW0j8Q+kEO4m0a6mcuN/Y0bzyHhgFc
bFpQHiyDq0NMn5vHniAwlF6Y2AzYBHlGOW+2h5YQJ/15a/B4juRpI38w/++KUKG9ZeRx9aK5
OE2r9ipq5ngnyyr0E5eVhuXzeDJMkgJKuO62olH9eXLY0bFhUVdU99B8trHpC6a+zMyWlfo9
pEygAOGO+BzHvs0GndSaRtj6FFULUb0j4S7BbbHb/sOgCeTSNd2qxTRH4gMkI95/KAOe4fXS
WyRORcUsR4wZ8hFOt5hkwKh3kQIDAQAB" ;
<http://schema.org/hasPublicKeyURL> 
<http://id.myopenlink.net/certgen/key/497> .


To conclude:

schema:webid or even schema:hasWebID is a relationship property name 
that provides a hint as to its function i.e., to associate an Identity 
with a sign that provides signification (denotation, naming, reference 
etc..).

Personally, the labeling of a relation doesn't matter one iota to me, I 
process relationship properties based on their relationship role 
semantics (not their labeling). I am simply concerned about how this 
affects the communities (and that goes beyond those imprecisely referred 
to "Web Developers" and "Web Masters") served by Schema.org.

Links:

[1] http://schema.org/sameAs -- schema:sameAs
[2] http://www.w3.org/TR/owl-ref/#sameAs-def -- owl:sameAs

-- 

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 Monday, 14 April 2014 02:06:36 UTC

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