W3C home > Mailing lists > Public > public-grddl-wg@w3.org > October 2006

GRDDL Spec Review

From: Harry Halpin <hhalpin@ibiblio.org>
Date: Sun, 22 Oct 2006 03:36:45 +0100
Message-ID: <453AD93D.7090305@ibiblio.org>
To: Dan Connolly <connolly@w3.org>, public-grddl-wg <public-grddl-wg@w3.org>

I promised to review the GRDDL Spec before release, and so here it is.

Overall judgement: Go ahead and publish, and fix minor changes as listed
below unless there is an objection.

Minor corrections and suggestions:

1. I "think" this is right. Note that since Stephen King doesn't have a
URI and neither does the stand, we could just do this:

"@@RDF version" ->"

<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:dc="http://purl.org/dc/elements/1.1/">

<rdf:Description>
  <dc:title>The Stand</dc:title>
  <dc:creator>Stephen King</dc:creator>
  <dc:format>Book</dc:format>
</rdf:Description>

</rdf:RDF>
"

We could also add the ISBN of the Stand, a non-URI id:

  <dc:identifier>ISBN 0517219018</dc:identifier>

Or make up "www.example.org"  a URI for "The Stand" ala

<rdf:Description rdf:about="http://www.example.org/books/TheStand">


I'll leave this up to the editor if they want to go either route to get
an ID - since
a RDF statement with a URI is  not that useful.

2) So, did we resolve the RDF output format issue - cause it sure sounds
like we did in the spec in favor of RDF/XML.

"a source document, that preserves its meaning in an RDF/XML representation"

"the source document, that preserves its meaning in an RDF/XML 
representation."

"an RDF/XML rendition" (this one occurs twice)

In all these cases, just replace "RDF/XML" with "RDF".

3) "And how can software determine whether authors of each are in fact
the same person?" -> "How can software determine whether authors of each
are in fact the same person?"

I'm pretty sure it's good style not to begin a sentence with "and"

4) "A Primer[primer] <http://www.w3.org/2004/01/rdxh/spec#primer> is a
[progressive] tutorial on the GRDDL mechanism" ->
"A Primer[primer] <http://www.w3.org/2004/01/rdxh/spec#primer> is a
[step-by-step] tutorial on the GRDDL mechanism

"Progressive"? As in politically? Let's just say "step-by-step" to keep
it clear.

5) "GRDDL (Gleaning Resource Descriptions from Dialects of Languages), a
mechanism for getting RDF data out of XML documents and in particular
XHTML pages using explicitly associated transformation algorithms. These
use cases also illustrate how XML and XHTML documents can be decorated
with microformat, Embedded RDF or RDFa statements to support GRDDL
transformations in charge of extracting valuable data that can then be
used to automate a variety of tasks." ->"GRDDL. These use cases
illustrate how XML and XHTML documents can be decorated with
microformat, Embedded RDF or RDFa statements to support GRDDL
transformations in charge of extracting valuable data that can then be
used to automate a variety of tasks."

Just delete the re-definition of GRDDL. If the user has read the
introduction they already understand what GRDDL is :)

5) "XML DTD" -> "XML DTD or schema."

Just to keep RELAX NG and XML Schema folks happy.

6) "and to emphasize the data-centric focus of the RDF/XML view." ->
I would just delete, since I have no idea other than how using the word
"data" using the namespace prefix "data-transformation" emphasizes
"data-centric" focus, but regardless "RDF/XML view" -> "RDF view"

7)The question over what actually is supposed to go into a namespace
document is wide open, so maybe just add in " could be anything" :
"a namespace document may have an XML Schema representation or an RDF
Schema representation, or perhaps both, using content negotiation
<http://www.w3.org/TR/webarch/#def-coneg>." -> "a namespace document can
be a variety of documents, such as RDDL[CITE THIS]. A namespace document
may be or link to an XML Schema representation or an RDF Schema
representation, or perhaps both, using content negotiation."

8) This sentence seems a bit, well, too brief and therefore confusing
given the following formal bit:

"To associate a GRDDL transformation with a whole dialect, use the
|grddl:namespaceTransformation| property." ->  "To associate a GRDDL
transformation with a whole dialect, have the namspace document include
the  |grddl:namespaceTransformation | RDF property. The precise methods
for allowing various types of namespace documents to include this
property are detailed below, first formally and then by example."

9)  Isn't "the resource identified by ?NS " the namespace document? Then
why not introduce ?NSDOC there?

"if an information resource ?D has an XML representation whose root
element has a namespace name ?NS then any GRDDL result of the resource
identified by ?NS is a GRDDL result of ?D" ->"

"if an information resource ?D has an XML representation whose root
element has a namespace name ?NS then any GRDDL result of the resource
identified by ?NS (the namespace document ?NSDOC)  is a GRDDL result of
?D" "

So, just to check my first reading understanding:

"if an information resource ?D has an XML representation whose root
element has a namespace name ?NS then any GRDDL result of the resource
identified by ?NS is a GRDDL result of ?D"

"This  means that if the source document has a root element that has a
namespace, then the  result documents of the namespace document of the
namespace are also the result document of the source document. "

Perhaps to clarify "In detail, this means running the GRDDL
transformation on the namespace document, getting the result document,
and then merging that graph of that result document  with to the graphs
of the result documents of running any other GRDDL transformations (if
they exist) on ?D.

 This does *not mean* running any GRDDL transformation specified by the
namespace document on the result document - that's given next...right?

So, " if an information resource ?D has an XML representation whose root
element has a namespace name ?NSDOC** and ?D has a GRDDL result that
includes, for any ?TX, the RDF triple { ?NSDOC
<http://www.w3.org/2003/g/data-view#namespaceTransformation> ?TX } then
?TX is also a transformation of ?D"

This means that if the source document's root element has a namespace
whose namespace document that contains a GRDDL transformation given by the

 "http://www.w3.org/2003/g/data-view#namespaceTransformation" RDF
property. So then any GRDDL transformation specified by the namespace
document is also run on the source document in addition to any other
GRDDL transformations given by the source document."

Those are just my informal translations of what I think are DanC's
correct formalizations of what's going on with namespace docs and GRDDL.
If they are correct, it might be useful to add them in, since they use
the same informal vocabulary used in the GRDDL Primer and use-cases but
might be easier to understand at first glance than the formal spec.

I'll look at the N3 Rules tomorrow to see if I grok them.

10) "For example, consider this privacy policy written in P3Q, a
contrived analog to P3P[P3P] <http://www.w3.org/2004/01/rdxh/spec#P3P>:"
-> This isn't  a suggested change for the WD, but a note for  future
releases. It would be *much* better if we didn't have to use a
"contrived "analog" and could do using a deployed XML language.

11) Exactly what section is this referring to? "Using GRDDL with an XML
Schema namespace document"? If so, let's say it upfront:

"The Working Group is likely to add a section to the GRDDL primer much
like this subsection. Since this subsection has no novel normative
material, we're interested in feedback on whether it should remain part
of this specification once it is covered by the primer." -> "The Working
Group is likely to add a section to the GRDDL primer much like the
subsection "Using GRDDL with an XML Schema namespace document" given
below. Since this subsection has no novel normative material, we're
interested in feedback on whether it should remain part of this
specification once it is covered by the primer."

"Also you don't mean GRDDL Primer, or do you? Do you mean
"specification"? Just making sure. If you really do mean primer, perhaps
we should remove this from the spec and put it in the Primer before
releasing the GRDDL WD. It's just confusing to have a bit of the spec
that says "Well, this is interesting, it might appear in another
document later..." I mean, I assume once the press release works out
people will also look at the primer again.

12) Ditto for the next "likely to add a section" bit.

13) "value is a URI reference that refers to an executable script or
program which is expected to transform the source document into an
RDF/XML rendition." -> "RDF rendition" instead of "RDF/XML rendition"

14) This method is suitable for use with valid XHTML documents which are
constrained by an XML DTD."-> "XML DTD or schema" instead of just "XML
DTD".

15) "meta-data in RDF/XML in a way that preserves the meaning of the
document" -> "metadata in RDF in a way that preserves the meaning of the
document"

16) "In the figure below, the arrow labelled info relates a document to
an abstract notion of the information contained in the document. " ->
Note that this should be moved up in spec to the first point where we
use the "info" line, i.e. the possibly to be moved "Using GRDDL with an
RDF Namespace document" setion.

17) There's a singular/plural type here:  " a available representations
" -> "available representations." Just delete "a"

18)"define a an XHTML profile." -> "define an XHTML profile." Can't have
both "a" and "an" at once.

19) Is it okay to use "embedded RDF statements" in a W3C WD as they are,
well, non-standard? I'm not sure. I like Embedded RDF personally, but
just a note. Also, why is the entire "The GRDDL Vocabulary" section in
the Rec? Shouldn't it be in the profile/namespace doc for GRDDL and then
reference from the Spec???



















-- 
		-harry

Harry Halpin,  University of Edinburgh 
http://www.ibiblio.org/hhalpin 6B522426
Received on Sunday, 22 October 2006 02:36:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:11:46 GMT