Re: review of current draft of 'RDFa syntax (January 25th, 2008)'


I have gone through and placed references to the diff-marked RDFa Last
Call editors draft to help you find the changes.

Ed Summers wrote:
> Generally I found the processing section much improved. I sympathize
> with the authors having to write code in natural language. I found
> myself wanting to read code instead of descriptive text at times--so
> perhaps pointing at the reference implementation would help people
> like me? I also found myself wondering whether bnode usage should be
> detailed so much.

There is no "official" reference implementation, the task force decided
that it was best to take a unit-test approach to assisting implementors,
but the W3C does not, in general, create reference implementations.
The test suite allows developers a great amount of flexibility when
implementing their parsers, it ensures that we don't accidentally
release buggy versions of the reference implementation, and it ensures
that there is a unit test suite to determine whether your parser is
conformant to the RDFa spec.

The test harness is available at:

This harness will probably be moved to a W3C URL after RDFa goes through
Last Call. The harness currently has 3 implementations and 2 SPARQL
endpoints that developers may choose from. Source code is available for
each implementation. For example, the librdfa source code can be
downloaded from the following URL:

git clone

> 3.8
> s/N-Triples/Turtle/

That has been corrected.

> 3.9
> """
> The aim of RDFa is to allow a single [RDF graph] to be carried in an
> XML document of any type, although this specification deals
> specifically with RDFa in XHTML.
> """
> Is non-XHTML RDFa discussed in any other documents that would be worth
> linking to here?

Diego raised a similar point. There are no formal non-XHTML document
types that use RDFa at the moment. That is why we don't reference them.

> 4.4
> Is it worthwhile mentioning the reference implementation?

There is no reference implementation, but we do have an outstanding item
to mention the test suite and the sample implementations.  It is now
referenced from the Status section.

> 5.2
> Is it worth mentioning that 'direction' needs to be captured in the
> list of incomplete triples?

Yes, it is. It is mentioned that the direction should be captured in the
list of incomplete triples.

See step #8:

> 5.4.3
> Just a style question, are the blue boxes that don't flow out to the
> right margins intended to draw attention to new changes temporarily?
> They are blocks with class = 'explanation'. IMHO they kind of break up
> the flow of things currently, and stand out a bit.

The blue boxes are a method of outlining algorithms or notes that are
pertinent to the processing rules in the syntax document. These blue
boxes are what is turned into code when the developer gets down to
implementation. This wasn't stated clearly in the previous document and
is now explicitly stated.

> s/initialised/initialized/ # twice

We corrected these - thanks for noticing!

-- manu

Manu Sporny
President/CEO - Digital Bazaar, Inc.
blog: RDFa Basics in 8 minutes (video)

Received on Sunday, 17 February 2008 23:26:28 UTC