Re: Does FOAF provide sufficient ontological support for WebID & WWW Identity AUTH REQ's?

On 6/10/14 8:12 AM, Tim Holborn wrote:
>
> On 10 Jun 2014, at 10:12 pm, Kingsley Idehen <kidehen@openlinksw.com 
> <mailto:kidehen@openlinksw.com>> wrote:
>
>> On 6/9/14 10:43 PM, Tim Holborn wrote:
>>>
>>> On 10 Jun 2014, at 10:36 am, Melvin Carvalho 
>>> <melvincarvalho@gmail.com <mailto:melvincarvalho@gmail.com>> wrote:
>>>
>>>>
>>>>
>>>>
>>>> On 10 June 2014 01:49, Timothy Holborn <timothy.holborn@gmail.com 
>>>> <mailto:timothy.holborn@gmail.com>> wrote:
>>>>
>>>>
>>>>
>>>>     Sent from my iPad
>>>>
>>>>     On 10 Jun 2014, at 2:26 am, Melvin Carvalho
>>>>     <melvincarvalho@gmail.com <mailto:melvincarvalho@gmail.com>> wrote:
>>>>
>>>>>
>>>>>
>>>>>
>>>>>     On 9 June 2014 17:30, Timothy Holborn
>>>>>     <timothy.holborn@gmail.com <mailto:timothy.holborn@gmail.com>>
>>>>>     wrote:
>>>>>
>>>>>         Hi Kingsley,
>>>>>
>>>>>         thanks for the response. Yes. Current format of WebID
>>>>>         (specifically) is simply FOAF.  Calling it WebID when it
>>>>>         means FOAF is well...  FOAF.
>>>>>
>>>>>
>>>>>     WebID is not coupled to FOAF.  Is FOAF ever mentioned in the spec?
>>>>     http://www.w3.org/2005/Incubator/webid/spec/identity/
>>>>
>>>>
>>>> Sorry, my mistake.  FOAF is indeed mentioned in the spec.  But it 
>>>> is not coupled.
>>>
>>> Ok.  My mistake - i thought it was coupled.
>>>>
>>>> "WebIDs can be used to build a Web of trust using vocabularies such 
>>>> as FOAF [FOAF 
>>>> <http://www.w3.org/2005/Incubator/webid/spec/identity/#bib-FOAF>]"
>>>>
>>> Perhaps the documentation could include / notate other vocab used in 
>>> connection to a WebID-TLS certificate; for example,
>>>
>>> http://xmlns.com/wot/0.1/
>>> http://www.w3.org/TR/vocab-org/
>>> http://usefulinc.com/ns/doap
>>> http://rdfs.org/sioc/ns
>>>
>>> (Which all appear to also use FOAF)
>>>
>>> I also found: http://www.w3.org/TR/2014/REC-vocab-data-cube-20140116/
>>>
>>>
>>>> But the point is that webid is not coupled to FOAF.
>>>
>>> Cool.
>>>
>>>
>>> Cheers.
>>>
>>
>> ACID test question:
>>
>> Can I make a document comprised of Identity claims, represented using 
>> RDF statements, that uses relations to assert possession of a 
>> keypair, where none of the relations use terms from the FOAF vocabulary?
>>
>> Answer: Yes!
>>
>> The <http://www.w3.org/ns/auth/cert#cert> relation isn't FOAF 
>> vocabulary based.
>
> Cool.  Perhaps you could show how something like Cimba gets your name 
> and storage location without using FOAF?

The issue with Cimba is UX i.e., it doesn't indicate to the user that 
following successful WebID-TLS authentication it also needs the following:

1. a relation to determine data storage location -- 
<http://www.w3.org/ns/pim/space#storage> isn't described by FOAF
2. the location above (object of 
<http://www.w3.org/ns/pim/space#storage> relation)  has to be provided 
by a service that supports LDP for read-write operations.

In regards to FOAF usage by Cimba (and other services) what's wrong with 
exploiting objects of relations such as: foaf:name, foaf:mbox etc.?

Let's say you put FOAF aside for something else, that alternative will 
still be a collection of relationship properties that are basically 
renditions of what FOAF describes.

FOAF isn't the problem. Understanding RDF based Linked Data remains the 
eternal problem. It's power (delivered via relations) is still generally 
misunderstood, across the board :-(


-- 

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