Re: Official response to RDF-ISSUE-132: JSON-LD/RDF Alignment -- Sub-issue on the re-definition of Linked Data

In general, this seems like a specific and reasonable request. Certainly if it ends this perma-thread. IMO, some of the importance given to de-emphasising RDF from the intro may not be as important now, due to it's being picked up by large companies; this gives a reason for developers to be interested in it on it's own merits, and not because it may be related to a term that may have had undesirable connotations.

A couple of points below:

On Jun 9, 2013, at 2:28 PM, David Booth <> wrote:

> I have re-read the first sections of the current editor's draft:
> Overall, it is very good and needs very little change to fix the problem that it currently redefines the term "Linked Data" in a misleading way (by omitting "RDF").
> This message makes a specific proposal for resolving this (sub-)issue about the redefinition of "Linked Data".  It attempts to achieve the dual objectives of: (a) avoiding the term "RDF" in the intro; and (b) not re-defining "Linked Data" in a misleading way.  It does this by removing the (re-)definition of Linked Data and instead including a reference to TimBL's Linked Data document.
> Specific wording changes suggested:
> 1. Add TimBL's Linked Data document to the list of references, with a short name of [LINKED_DATA]:

+1. This is also consistent with the way it's introduced in the Linked Data Platform [1]

> 2. In section 1 ("Introduction"), change the first occurrence of "Linked Data" in two ways: (a) change the font to be a plain, non-bold, non-italic font; and (b) add the citation "[LINKED_DATA]" after it.

This is actually not a stylistic choice, but a side-effect of it's being a definition within the document (note the <dfn> around it). This is how other uses of the term relate back to the internal definition and use used widely for other similar terms that have a specific meaning in the document.

> 3. Also in section 1 ("Introduction"), change:
>  "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"
> to:
>  "It allows"

Typically re-stating things from a referenced document may be appropriate, but is not necessary. If the concern is that this re-states the properties, then IMO, this seems like splitting hairs. I find the properties to be essentially equivalent, and think that the wording we found in the JSON-LD introduction is arguably better, however they specifically don't reference RDF and SPARQL. All things considered, given the explicit reference to [Linked Data], I think that these properties form a manifesto that probably can't be repeated often enough. I would be a -0.1 on removing the redefinition.

> Thanks,
> David



Received on Sunday, 9 June 2013 22:01:25 UTC