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

Re: Re-definition of Linked Data

From: Pat Hayes <phayes@ihmc.us>
Date: Sun, 16 Jun 2013 13:01:20 -0500
Cc: "'public-rdf-comments'" <public-rdf-comments@w3.org>
Message-Id: <8C5BD605-2BC3-4DC5-997B-A63DAE7B6ACF@ihmc.us>
To: "Markus Lanthaler" <markus.lanthaler@gmx.net>

On Jun 16, 2013, at 8:15 AM, Markus Lanthaler wrote:

> David Booth raised concerns that we are re-defining Linked Data in the
> JSON-LD spec. That's certainly not the intention. We just try give the
> reader an idea what we mean when we use the term throughout the document.
> IMO it would be very weird to not do so and thus I wouldn't like to remove
> that explanation completely. So let me try to make a proposal which sounds
> more like an explanation than a definition.
> 
> The current text reads as follows:
> 
>  Linked Data is a technique for creating a network of inter-connected
>  data across different documents and Web sites. In general, Linked Data
>  has four properties: 1) it uses IRIs to name things; 2) it uses HTTP
>  IRIs for those names; 3) the name IRIs, when dereferenced, provide
>  more information about the thing; and 4) the data expresses links to
>  data on other Web sites. These properties allow data published on the
>  Web to work much like Web pages do today. One can start at one piece
>  of Linked Data, and follow the links to other pieces of data that are
>  hosted on different sites across the Web.
> 
> Here's a very rough draft about how we could paraphrase this to not make it
> sound like a definition:
> 
>  The World Wide Web is a global information space consisting of
>  documents and other resources connected by hyperlinks. The same
>  principles that underpin the document-based Web can also be used for
>  data. In fact, identifying things, i.e. entities and their properties
>  with IRIs (Internationalized Resource Identifiers as described in
>  [RFC3987]; a generalized form of URLs) is often beneficial because it
>  means that those IRIs can be resolved to find more information about
>  those things. Such data is often called Linked Data. As Figure 1
>  illustrates, it has the form of directed graphs, meaning that every
>  property points from a node to another node or value.
> 
>  <<image showing the default graph of Figure 1, no named graphs yet>>
> 
>                      Figure 1: A Linked Data Graph
> 
>  JSON-LD is a lightweight syntax to serialize Linked Data in JSON
>  [RFC4627]...
> 
> 
> David, would that address your concerns? Could you (and of course also the
> rest of the RDF WG) live with something like this?
> 
> Pat, what about the editorial concerns you raised? Do you think this is
> enough clarify that the data model is based on directed graphs early enough
> in the document?

Well, it helps, but I think a paragraph that outlines the data model would be better, in addition to this. 

The problem as I see it is, one could (I hear the sound of ice cracking as I type this) have something that fulfilled the description above with IRIs as links to other data, etc.., and yet used a completely different data model. We could have implemented linked data using Excel tables as our local data model rather than graphs. The idea of using a node/graph model for storing data locally (the data model that RDF and JSON-LD have in common) is really orthogonal to the whole using-IRIs-as-links idea that is the heart of LD. So (quickly getting back onto firmer ground) I think the document needs to actually say that JSON-LD *chooses* to use this node/graph model. It doesn't need to say *why*, just quickly describe the basic model, no details, and use the terms graph, node, subject/property/value. 

Pat

PS I like the first text better than your replacement, I am afraid. It is clearer and more to the point. You could make it less definition-like by the substitution

> In general, Linked Data
>  has four properties:

//

Although there is no exact definition of Linked Data, it is typically has four important properties:

or some such wording. 

> 
> 
> Thanks,
> Markus
> 
> 
> 
> --
> Markus Lanthaler
> @markuslanthaler
> 
> 
> 

------------------------------------------------------------
IHMC                                     (850)434 8903 or (650)494 3973   
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Sunday, 16 June 2013 18:01:46 UTC

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