Re: ISSUE-66: LinkedDataT

On 8/5/14 12:44 PM, Markus Lanthaler wrote:
> On 5 Aug 2014 at 11:07, Ruben Verborgh wrote:
>> >Jumping in, as this is very relevant for the Linked Data Fragments spec
> [1].
>> >In fact, this issue appearing after I drafted an introductory section
> called
>> >     "What Linked Data is"
>> >might not be a coincidence. (And it's very good timing in any case.)
> Yeah, this is mostly about the Linked Data Fragments spec.
>
>
>> >Let me start out by saying I was totally oblivious of "non-RDF Linked
> Data".
>> >I.e., I had always assumed that Linked Data is in RDF;
>> >probably because Tim's original principles explicitly mention this [2].
> Funny fact: the "original principles" didn't explicitly mention it:
>
>   
> https://web.archive.org/web/20061115043657/http://www.w3.org/DesignIssues/Li
> nkedData.html
>
> It took 3 years till that was added.

Yes !!!

And the fact that recent narratives attempt to paper of this fact, 
poorly doesn't help matters at all. Even worse, the folks papering over 
this fact weren't anywhere close to the crafting of the original meme.

When WE launched Linked Data at the World Wide Web conference in Banff, 
2007, that was the meme presented to the enthusiastic audience [1] .

>
>
>> >Then again, we all know the principles are quite vague:
>> >- RDF* and SPARQL are mentioned between parentheses.
>> >   Did this mean "e.g., RDF*, SPARQL", or "i.e., RDF*, SPARQL"?
>> >   That's an important difference, and we'll likely never know.
>> >- Where is the asterisk after RDF ever resolved?
>> >Maybe I just missed the majority of the discussion;
>> >i.e., posts like [3] were written in 2009.
>> >
>> >That said, me being in the community for 4 years
>> >and never having heard about (or being selectively deaf towards)
>> >non-RDF Linked Data, means something at least.
> We had a very long (and heated) discussion about this when we standardized
> JSON-LD. It starts more or less here

Yes!

>   
> http://lists.w3.org/Archives/Public/public-rdf-comments/2013Jun/thread.html#
> msg28
>
> in case you want to waste a couple of hours (days?) reading more about this
> :-)
>
>
>> >I'd dare to say that the majority of people do assume
>> >that Linked Data is just done with RDF.
> That's obviously true for the Semantic Web community. Not so true for the
> rest of the world:-)  Hydra tries to bridge the gap between those two worlds
> (just as JSON-LD does).

Bridging the gap might even go beyond the wisdom of Solomon. Thus far, 
killing the baby (by proverbially splitting it in half) has always been 
the preferred route :-(

>
>
>> >So to what extent is it then necessary to clarify this?
> I think it is very important as our group is not a homogenous group of
> Semantic Web experts.

Yes!

>
>
>>>> >>>i was specifically trying not to get that discussion going. just asking
> I'd also like to urge to not get into that discussion.

Discussion are useful. They aren't homogeneous, forum and audience 
matters. JSON-LD made progress, the same SHOULD happen here too.

>
>
>>>> >>>whether there should be some definition/clarification of the term, just
>>>> >>>to let readers know what it means in the context of the spec/community.
>>>> >>>if you define a broad term to mean a narrow thing, then this might be
>>>> >>>helpful to avoid possible confusion.
> In my opinion we should define it.

Yes you SHOULD.

>   At the moment, I think it is clear that
> LDF are*exclusively*  for RDF. It would be interesting to see if it can be
> generalize in the future but till we get to that point I think we should
> simply be honest about it.

Yes!

You can provide a definition that stays true to the goal of bridging the 
artificial gaps.

RDF came after the Web. The Web has always been tied together using 
Triples based relations.

The issue here is the evolution from the implicit to the explicit in 
regards to the nature of the relations that facilitate what's known as 
the World Wide Web.

>
>
>> >What do you think about the current introduction
>> >to the triple pattern fragments spec [1]?
> It's quite nice but I think it could be further improved, especially for
> people without a lot of SemWeb background.

Yep! See my comments above, and in prior posts to this thread.

>
>
>> >Not knowing about this issue yet, I phrased it as:
>> >
>> >     By publishing Linked Data [LINKED-DATA],
>> >     we enable automated clients to consume information.
> Hmm... automated clients such as Google are quite happy consuming plain old
> HTML...

Plain Old Semantic HTML (POSH) is just another notation for relations 
represented as subject, predicate, object or entity, attribute, value 
based triples.
RDF didn't invent the triple. It simply introduced the use of IRIs to 
denote elements of the above.

> I know what you are trying to say but people who haven't spent a
> whole lot of time on this won't understand it, I think.
>
>
>> >     In practice, this information is available as RDF triples [.]
>> >
>> >So it leaves the question open whether non-RDF Linked Data exists;
>> >it just says that, in practice, it will be RDF. Good enough?
> Maybe it would be more straightforward to explain it the other way round:
>    - documents are in natural language

Yes, document content is crafted using natural language in the real-world.

>    - machines are bad in understanding natural language

Yes.

>    - machines prefer structured data using unambiguous identifiers

The can understand structured data. Even better, If you enhance the 
structured data with IRIs then you reduce ambiguity, which can trip them 
up en route to understanding the nature of a binary relationship that 
connects (links) two things. Finally, if you use HTTP URIs for 
denotation (naming) you end up with the identifiers that function like 
terms (rather than ambiguous words) [1].

>    - the Web uses URLs* as identifiers
No, it uses HTTP URIs as identifiers.

URLs are just a *kind* of HTTP URI that denotes a Web Resource i.e., a 
Web Document or Document Collection.

>    - RDF allows data to be expressed in a machine-processable way by
> leveraging URLs

RDF does this via IRIs.

Linked Data mandates doing this with HTTP URIs. That's an important 
difference.

>    (- RDF expresses data in the form of triples) -- could be omitted

Subject, Predicate, Object patterns should never be omitted. All you do 
is connect them to natural language [2], where sentences are constructed 
in the very same way.
>    - RDF can be serialized in various formats such as JSON-LD, HTML+RDFa, or
> Turtle

Serialization across-the-wire is only part of the story. You create RDF 
document content using a variety of notations. You serialize document 
across-the-wire using a variety of serialization formats. Again, a 
nuance that needs better treatment.

>
> * I think simply talking about URLs instead of URI or IRI is fine in this
> case

No, that's never a good choice. All you have to do is simply explain the 
fact that HTTP URLs are a kind of HTTP URI. Basically, a kind of HTTP 
URI that explicitly denotes (names) a document location. The content of 
the Document is where RDF really comes into play.

>
> I would also suggest to use a different term than "Linked Data document". Is
> it actually needed or could we also get rid of this concept?
>
>
>>> >>I think a definition could help.  I suggest copying the one from the
>>> >>W3C Linked Data Glossary verbatim (and referencing that document),
>>> >>rather than trying to craft a new one and risking another long debate
> about

That document is broken, to be blunt. I don't encourage you to do that 
if clarity is the goal.

>>> >>what it should be.
>> >
>> >Sadly, I think that definition is quite complicated.

It doesn't have to be so.

There is nothing complicated about RDF or Linked Data. That's 
misconception is being used to perpetuate dysfunctionality that impedes 
progress towards comprehension of this subject matter.

>> >Here it is at full length, copied from [4]:
>> >
>> >     Linked Data
>> >     
>> >     A pattern for hyperlinking machine-readable data sets to each other
>> >     using Semantic Web techniques, especially via the use of RDF and URIs.
>> >     Enables distributed SPARQL queries of the data sets and a browsing
>> >     or discovery approach to finding information (as compared to a search
> strategy).
>> >     Linked Data is intended for access by both humans and machines.
>> >     Linked Data uses the RDF family of standards for data interchange
>> >     (e.g., RDF/XML, RDFa, Turtle) and query (SPARQL).
>> >     If Linked Data is published on the public Web,
>> >     it is generally called Linked Open Data. See also [Linked Data
> Principles].
>> >
>> >It forces you to understand:
>> >- Semantic Web
>> >- RDF
>> >- URIs
>> >- SPARQL
>> >to make sense out of it.
> I fully agree. This definition is not going to help anyone.

Exactly !!!!

>
>
>> >On the technical level, nothing prohibits us from making Linked Data
>> >Fragments broader than RDF. We'd have to be very careful, however,
>> >that the concept would still be sufficiently meaningful; that it
>> >doesn't become hollow by broadening it.
> Yeah, I would like to explore that in the future. However, till we get
> there, we should make it clear that at least a mapping to RDF is required.

It is my hope, that like the JSON-LD effort, this group will end up with 
a much clearer, and less dogmatic view of RDF and Linked Data. Remember, 
Google adopted JSON-LD without a scintilla of the usual RDF virtues 
evangelism  :-)

Links:

[1] 
http://www.w3.org/wiki/SweoIG/TaskForces/CommunityProjects/LinkingOpenData/BanffGathering 
-- Inaugural Linked Open Data meetup at Banff 2007
[2] http://slidesha.re/QEqLZN -- RDF, Linked Data, and Natural Language
[3] http://www.jfsowa.com/pubs/fflogic.pdf -- Fad, Facts, and Fallacies 
about Logic
[4] http://www.wikihow.com/Differentiate-Between-a-Term-and-a-Word -- 
Word vs Term .

-- 
Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog 1: http://kidehen.blogspot.com
Personal Weblog 2: 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
Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this