W3C home > Mailing lists > Public > public-webid@w3.org > October 2012

Re: Browser UI & privacy - a discussion with Ben Laurie

From: Henry Story <henry.story@bblfish.net>
Date: Fri, 5 Oct 2012 14:18:42 +0200
Cc: Hannes Tschofenig <hannes.tschofenig@gmx.net>, Dick Hardt <dick.hardt@gmail.com>, Melvin Carvalho <melvincarvalho@gmail.com>, "public-webid@w3.org" <public-webid@w3.org>, public-identity@w3.org, "public-philoweb@w3.org" <public-philoweb@w3.org>, Ben Laurie <benl@google.com>
Message-Id: <186C3BD5-DB9D-492F-BF41-037DF99BFDD1@bblfish.net>
To: Kingsley Idehen <kidehen@openlinksw.com>

On 5 Oct 2012, at 13:58, Kingsley Idehen <kidehen@openlinksw.com> wrote:

> On 10/5/12 3:08 AM, Hannes Tschofenig wrote:
>> Dick,
>> 
>> you have to see this from a computer science perspective.
>> 
>> Think about a relational database system. You have one or more tables in each database with columns. The columns represent the attributes that are stored about you. Rows are specific instances (values) for these attributes.
>> 
>> Different systems (with their databases) store different data about you. Even though there is from a philosophical point of you only one Dick Hardt each of these system only see a small subset of the attributes of you.
>> 
>> The index, the unique key, is the identifier and is (of course) an attribute itself. This makes the crisp separation of identity and identifier somewhat difficult.
> 
> 
> Only because you are using an extensional system for your example.
> The identifier in your example is an extension of the data that makes the record. An intensional system wouldn't have such a limitation which is basically what graph / object model DBMS engines bring to the table. This is what Object Identity [1] in the object manifesto of yore was all about.

I am not sure how intensional and extensional come into play here? Can you point me to SemWeb documents on this?

But I can try to get it from first principles.

The extension of a one place predicate P in logic is the set of objects that satisfy it. So for Giraffe it 
is the set of Giraffes.  Hence if Sophie is a Giraffe, then Giraffe(Sophie) is true off Sophie is 
in the set of Giraffes.

For foaf:knows which is a relation (== a two place predicate) the extension is the
set of pairs of people who know each other.

The Intension is the definition of the term, or perhaps something like the way of 
knowing  that something  is in the set of things for which it stands. In the 
semantic web/Linked Data/Web space the  meaning of a URI is given by the document 
returned on dereferencing it. That gives the canonical meaning for the URI.

So yes, the Web has a way of finding the definition of a URI for a URI, and that is what
OpenID uses in fact when on dereferencing a webID document the Relying party finds a relation
such as:

<> openid:provider <http://some.big.provider.com/auth?>  

The openID <> is the document that defines what it is about. Hence it can define that it is
an openid profile and that it delegates authentication to the provider.

With WebID we have just removed the need for the provicer, by using Cryptography.

So yes, the intension/extensional issue does help make sense of this. But it's a bit concise
in your explanation above.


>>  Of course there may be more than one key that identifies you within that system. Hence, sharing the attributes with other parties may then allow them to uniquely identify you (with a certain probability) even though you never disclosed the unique key.
>> 
>> Hope that this makes sense.
> 
> Yes, but within confines of an extensional system. In an intensional system every record is a proposition. Basically, this is the model upon that drives the Semantic Web, Linked Data, and  WebID.
> 
> Links:
> 
> 1. http://www.cs.sfu.ca/CourseCentral/354/zaiane/material/notes/Chapter8/node8.html
> 2. http://www.cs.cmu.edu/afs/cs.cmu.edu/user/clamen/OODBMS/Manifesto/htManifesto/node4.html#SECTION00022000000000000000 .
> 
> Kingsley
>> 
>> Ciao
>> Hannes
>> 
>> 
>> On Oct 4, 2012, at 8:26 PM, Dick Hardt wrote:
>> 
>>>  a somewhat tangential comment.
>>> 
>>> On Oct 4, 2012, at 8:10 AM, Hannes Tschofenig <hannes.tschofenig@gmx.net> wrote:
>>> 
>>>>   $ Identifier:   A data object that represents a specific identity of
>>>>      a protocol entity or individual.  See [RFC4949].
>>>> 
>>>> Example: a NAI is an identifier
>>> Easy to  agree to.
>>> 
>>>>   $ Identity:   Any subset of an individual's attributes that
>>>>      identifies the individual within a given context.  Individuals
>>>>      usually have multiple identities for use in different contexts.
>>>> 
>>>> Example: the stuff you have at your Facebook account
>>> Not so easy to agree to.
>>> 
>>> I would argue that your identity is everything about you, your Facebook data being part of your identity. Saying I have multiple identities is confusing. There is only one of me. Any slicing becomes challenging as there are no sharp lines between who I am on Facebook, Twitter, Google+ or LinkedIn. There is lots of overlap.
>>> 
>>> -- Dick
>> 
>> 
>> 
> 
> 
> -- 
> 
> Regards,
> 
> Kingsley Idehen	
> Founder & CEO
> OpenLink Software
> Company Web: http://www.openlinksw.com
> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
> Twitter/Identi.ca handle: @kidehen
> Google+ Profile: https://plus.google.com/112399767740508618350/about
> LinkedIn Profile: http://www.linkedin.com/in/kidehen
> 
> 
> 
> 
> 

Social Web Architect
http://bblfish.net/



Received on Friday, 5 October 2012 12:19:17 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:54:37 UTC