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

Re: Official response to RDF-ISSUE-132: JSON-LD/RDF Alignment

From: David Booth <david@dbooth.org>
Date: Fri, 07 Jun 2013 20:22:28 -0400
Message-ID: <51B27944.5030003@dbooth.org>
To: Kingsley Idehen <kidehen@openlinksw.com>
CC: public-rdf-comments@w3.org
On 06/06/2013 10:15 PM, Kingsley Idehen wrote:
> On 6/6/13 7:55 PM, David Booth wrote:
>> The term "Linked Data" has a well-established meaning within semantic
>> web community.  The JSON-LD group would be *misleading* the public by
>> stating or implying that Linked Data is not necessarily based on RDF.
>>
>> As I pointed out to Kingsley a few weeks ago:
>> http://lists.w3.org/Archives/Public/public-rdf-comments/2013Apr/0086.html
>> [[
>>>   - Of the top 10 hits from in a google search for "Linked Data",
>>> **every one of them stated or implied that Linked Data is based on
>>> RDF.**
>>>
>>>   - Of the top 10 sites listed in a google search for '"Linked Data"
>>> is', **every one of them stated or implied that Linked Data is based on
>>> RDF.**
>>>
>>>   - Of the top 10 sites listed in a google search for '"Linked Data"
>>> definition', **every one of them stated or implied that Linked Data is
>>> based on RDF.**
>>>
>>> How much evidence do you need?  Shall we check the top 100 hits?  Or the
>>> top 1000 hits?  Shall we try other search engines?   If you search hard
>>> enough you might find a tiny fraction that supports your claim. But the
>>> vast majority of the evidence does not.
>>>
>>> The vast majority of the evidence indicates that in established usage,
>>> the term "Linked Data" implies the use of RDF.  If you wish to propose a
>>> new definition that is contrary to this established usage, you are
>>> obviously free to do so.  But please do *not* make the patently false
>>> claim that your proposed new definition reflects accepted usage.  It
>>> very clearly does NOT.
>> ]]
>>
>
> And I told you I disagree with your conclusions strongly.

It seems that no amount of evidence is going to convince you.  You can 
disagree with reality all you want, but it doesn't help.

> You are quite inaccurate in those claims.
>
> Moving forward, hopefully, what's to the problem with RDF based Linked
> Data?
>
> Did you ever read the original Linked Data meme?
>
> I assume you are aware that TimBL actually tweaked his original meme,
> and that simple tweak (which I complained about vehemently at the time)
> has lead these kinds of threads.
>
> When TimBL designed the Web [1] the letters "RDF" did not exist, the
> concept of Linked Data already existed.

No, the generic concept of "linking data" existed, i.e., the concept of 
connecting data.  The concept of "Linked Data" -- the concept defined by 
TimBL in talking about the semantic web -- did *not* exist.

> Now don't tell me for one second
> you truly believe that the concept of Linked Data is an RDF invention,
> that's utterly inaccurate.

The term "Linked Data" certainly was coined by TimBL, and it has a 
specific, well established meaning in the semantic web community, and 
that includes being based on RDF.

If one is talking in a database context about a "foreign key", that term 
has specific, well-established meaning in that context.  It does *not* 
refer to an imported piece of metal that is used to unlock a door. 
Similarly, "Linked Data" has a specific, well-established meaning in 
this community -- it does *not* refer to the concept of creating paper 
chains out of core dump printouts.

>
> When TimBL designed the Web the following where firmly established:
>
> 1. denote entities using URIs
> 2. describe entities using Relations
> 3. Relations are sets of Relationships
> 4. Relationships can be represented using 3-tuple based
> statements/proposition/claims.
>
> Items 1-4 describe Linked Data.
>
> You can have Linked Data silo-ed at the following levels:
>
> 1. operating system
> 2. programming language runtime
> 3. DBMS system
> 4. Web APIs .
>
> Where does RDF add value?
>
> Its ability to make the relationship semantics *explicit* rather than
> *implicit*. Basically, RDF enables you to publish very smart and
> inherently extensible Linked Data. You can tweak the Relations via
> vocabularies and ontologies.
>
>
>> Furthermore, the official charter for the W3C Linked Data Platform
>> Working Group states explicitly that: "RDF is the basis for Linked
>> Data and the Semantic Web".
>> http://www.w3.org/2012/ldp/charter
>
> Really?

Yes.  Please dereference the URI and read it.

> Have you had the pleasure of digesting some of the threads on
> that group? If you are concerned about JSON-LD, wait to you see what's
> going on over there (see links section).
>
> There is a simple solution to this problem:
>
> RDF based Linked Data is how we should refer to Linked Data that's based
> on RDF. Otherwise, we are simply fighting a justifiably losing battle
> trying to pack the generic phrase "Linked Data" into the already
> over-conflated bucket called "RDF".

AFAICT, the only conflating that's going on here is that you are 
conflating the generic (non-web) concept of "linking data" (i.e., 
connecting data) with the specific concept of "Linked Data", i.e., the 
term coined by TimBL as a better way to talk about the semantic web.

David

>
> Overloading RDF has never helped the course. Each item added to the
> bucket, in non negotiable manner, simply lays the foundation for more
> political distractions that simply undermine RDF.
>
> RDF is great technology. All participants in its development should be
> proud. Thus, there's no reason on earth to be insecure about RDF virtues
> by eternally trying to pack everything into it.
>
> Making entity relationship semantics human- and machine-comprehensible
> and extensible is quite a technical feat! Let's put our energy into
> coherent messaging about RDF.
>
> Links:
>
> [1] http://www.w3.org/DesignIssues/diagrams/history/proposal-fig1.gif --
> note the "describe" relation in that diagram (there was no RDF then, but
> the fundamental concept already existed)
> [2] http://bit.ly/YrAz5L -- Overview by Issue Creators
> [3] http://bit.ly/Xuyd6Z -- Overview by Issue Status .
> [4] http://bit.ly/VBoRsH -- Steve's Issues Report .
> [5] http://bit.ly/YrBrr9 -- Erik's Issues Report (using URIBurner data
> space) .
> [6] http://bit.ly/15llZU0 -- Issues by ID
> [7]
> http://linkeddata.uriburner.com/proxy-iri/7cc90e546b8b148bc97985a7735edefb97e1374e
> -- sample Linked Data URI for an issue
> [8] http://lists.w3.org/Archives/Public/public-ldp-wg/2013Jun/ -- the
> list (enjoy!) .
>
> --
>
> 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 Saturday, 8 June 2013 00:22:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:29:57 UTC