Re: Issue-37: Ontological Modelling

On 1/22/13 12:35 PM, Wilde, Erik wrote:
> On 2013-01-22 18:28 , "Kingsley Idehen" <kidehen@openlinksw.com> wrote:
>> On 1/22/13 12:16 PM, Wilde, Erik wrote:
>>> yes i did. we simply have to fill in that blank and can learn from
>>> people
>>> who have done that already.
>> What does that mean? Who are these people and what are their credentials?
> people who have implemented collection management web services and decided
> that they need protocol support to create collections. what specifically
> do you want to know about their credentials? it's just implementation work
> that happened over the last couple of years, and something to look at and
> learn from, if we want to.
>
> cheers,
>
> dret.
>
>
>

You seem to assume that Atom is unknown to many on this list. I make 
that claim on the basis of your constant references to Atom and what you 
perceive as its completeness relative to what LDP is supposed to be 
pursuing. In addition, your comments (in my eyes) presume that there are 
no LDP participants that have implemented Atom.

FWIW -- we (OpenLink Software) have one of the earliest implementations 
of Atom in Virtuoso (we even use the Atom Ontology that Henry worked on 
re. RDF transformation). We also support WebDAV, and we also develop an 
deploy a widely used Linked Data deployment and management platform.

Henry too (as per my comments above) is no stranger to Atom or the 
problem at hand. He's provided a very clean mechanism for understanding 
the issues, unambiguously, and I don't understand your push-back or its 
basis, bearing in mind the following:

1. Linked Data Platform -- is by definition about RDF based Linked Data 
(this is a W3C effort so RDF is an integral part of the Linked Data 
definition, for good reason).
2. Linked Data Principles as espoused by TimBL are inherently aligned to 
HTTP and other parts of Web Architecture.
3. LDP is about Linked Data principles exploitation -- basically the 
pursuit of a solution that standardizes Web-scale Create, Read, Update 
(PATCH), Delete operations via RESTful interaction patterns.

If Atom was adequate LDP wouldn't need to even exist. Atom was all about 
XMLifying HTTP (as Henry's already articulated). LDP needs to be about a 
more intelligent augmentation of HTTP via exploitation of RDF's data 
model. Basically, HTTP combined with self-describing resources -- 
comprised of content bearing *explicit* machine and human comprehensible 
entity relationship semantics.

RDF's data model is what makes the entity relationship semantics 
(expressed in structured content) machine accessible and comprehensible.
The statement above has nothing to do with a specific RDF notation 
(Turtle, RDF/XML, RDFa, etc..) for expressing RDF's 
subject-predicate-object based syntax.

-- 

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