W3C home > Mailing lists > Public > public-ldp@w3.org > June 2013

Re: Proposal to close ISSUE-19: Adressing more error cases, as is

From: Kingsley Idehen <kidehen@openlinksw.com>
Date: Mon, 03 Jun 2013 19:43:22 -0400
Message-ID: <51AD2A1A.3050204@openlinksw.com>
To: public-ldp@w3.org
On 6/3/13 5:30 PM, Erik Wilde wrote:
> On Jun 3, 2013, at 14:11, Kingsley Idehen <kidehen@openlinksw.com> wrote:
>> Semantics and Media Types are two distinct things. Media Types don't determine Semantics. Semantics don't determine Media Types.
> for you they don't; for many others they do. both options are design choices. in the REST community, this debate took years.

The REST community isn't the point. RDF != REST or vice versa.

RDF based inked Data is a RESTful interaction pattern for data access 
that leverages RDF model theory -- entity relationship semantics that 
are comprehensible to humans and machines.

RDF based Linked Data is all about about a contemporary tweak to the 
age-old entity relationship model where the relationship semantics are 
explicit (rather than implicit) and discernible to machines and humans. 
The logic is baked into the data and that has nothing to do with Media 
Types.

>   it seems that nowadays, many favor exposing application/protocol+xml over just exposing application/xml, but i am sure you'll still find naked XML out there.
RDF has nothing to do with Media Types when the context of the 
conversation is model theory.

The greatest disservice anyone can do to RDF is speak about it as a format.

Formats come into play when transmitting data. The formats in question 
do not determine the semantics of the payload when said payload is an 
RDF graph.

>
> in the end, you have to have some way to reach agreement about what you're talking about.

You have logic. Things exist, they are denoted using identifiers and 
their existence is quantifiable. Nothing exists on its own, everything 
is related. The final question is just about relatedness i.e., what are 
the semantics of the relation that facilitates a relationship between 
two entities.

>   you can push this into your model or metamodel, or surface it in the service interface. if you want loose coupling, the latter is better, but like all design choices, it's a trade-off with pros and cons.

RDF offers a fundamental affordance: the ability to separate entity 
relationship graph representation from entity relationship semantics. 
Adding Linked Data URIs to the mix simply makes said affordance Webby 
[1][2].

>
> cheers,
>
> dret.
>
>
Links:

1. http://bit.ly/11xnQ36 -- RDF based Linked Data URI (it resolves to 
content that uses RDF model theory to describes the URI's referent)

2. http://bit.ly/15tk1Au -- same thing via hashless URIs.





-- 

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






Received on Monday, 3 June 2013 23:43:51 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:16:35 UTC