Re: Overview and Abstract Data Model - new document

>>>Graham Klyne said:
> 
> A new draft of
>    Resource Description Framework (RDF): Overview and Abstract Data Model
> is available at:
>    http://www.ninebynine.org/wip/RDF-basics/2002-07-25/Overview.htm

Please can you copy to www-archive, thanks.

> 
> This has been updated to reflect WG decisions from the 2002-07-12.

OK, here goes.

Title:
  The group decided this is not an RDF overview document
  in the 2002-07-12 meeting:

  [[
  Decided to drop the implication that this is a top document
  and update it.  Change the title to something else
  (suggestion RDF Abstract Data Model, but can come back to that)
  ]] -- Minutes of 2002-07-12 telcon
     http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jul/0081.html

  The suggestion in the meeting is better.


Abstract
  typo: eacyh
  Doesn't describe that the document does 'meaning', design goals,
  rationale and graph syntax.

Status
  This is not a working draft yet.  Change all bits in the
  non-italics about 'This is a ... working draft' to say something
  like "... Editor's Draft".

TOC
  Rather deep TOC, do you need it to this level?  3.2.2.1 stands out
  as an oddity.

Section 1
  Para 3 Says section 2 is background, design goals and rationale.
  Rename section 2 title to represent that.  Maybe add motivation and
  design goals to earlier five bullets.

Section 2
  rename title to 'Design goals and rationale'.

  References citing should be [sowa][cg]... after W3c doc style and
  not in ()s

2.2
  [W3C] XML schema datatypes were never in the original RDF design,
  need to clarify somehow.

  Cite something about W3C XML Schemas.

2.2.1
  more references style change

2.2.4
  cite the normative rdf syntax WD here not stripedrdf

2.3 Meaning
  somebody is surely going to hate that title ;)

2.3.2
  section 3.7 link is broken.  Point to the mime types section of
  the syntax document.

2.4.1
  directed edge-labeled graph ?

  Introduce the term triple more formally?

2.4.2
  URIrefs appears in brackets.  Describe as "URI references" and then
  abbreviate. Make it appear more of a defn?.  Cite the section in the URI RFC.

  I note later on that this is defined in 3.1, maybe add a forward reference?

2.4.3
  [XML-AS-RDF] - no reference, not sure what you want to cite here.

  N-Triples rather than notation3?  Reverse setence to emphasis
  the normative syntax :)

2.5
  It would be nice to see something about XML namespaces, RDF schema
  document location and so on, how they have typically been used.

  I noticed CC/PP says things about this, beyond what RDF core does.

2.5.1
  This section has an out of date and wrong list - it confuses
  vocabulary terms with mere syntax construction things.  It
  duplicates a section in the syntax document updated after many WG
  decisions.  Please delete this list and add a pointer to the syntax
  doc and the RDF Vocab/Schema document for the definitive lists.

  The discussion of errors is also incomplete, the XML syntax
  document describes that in detail with test cases, please point at
  that.

2.5.2
  Deprecated terms - should again point to the syntax and schema
  documents rather than duplicate it, where what this mean can be
  explained.

3

(Syntax doc will point to 3.1*, 3.2* and remove the appropriate sections)

3.1
  RDF URIrefs are compatible with, but aren't based on, W3C XML
  Schema (WXS) anyURI.  If we define RDF in terms of WXS, that adds new
  normative references and we haven't done that yet.

3.2.2
  Explain where <tag> came from.  Something like, this is not a new
  element but used to show how an XML literal is any legal XML
  element content and can be used "as if" enclosed in a
  <tag>...</tag>

3.2.2.1
  odd numbering, only sub-section.  Maybe add 3.2.3 Equality of
  Literals and add equality of strings(+lang), xml(+lang)
  sub-sections of that/

3.4 Triples
  need to point here from earlier where triple was first mentioned

3.5 RDF graph
  "collection of RDF triples"?   Must be something more precise than
  that?

  The Note: maybe should be in the body text and a <ul> list?

3.7
  This list is correct at the moment, but I would ask it point
  to a syntax doc section or list, since the list may change
  (for example the proposed rdf:node)

4

4.2
  syntax doc and mime registration document should be updated to
  point to here

4.3
  qname reference should point to the definition of the term 'qname'
  in XML-NS.

  delete parsers/writers.  The syntax doc never uses these terms but
  only speaks of a documents and mappings.  Add pointers to
  the references to serializing in xml syntax doc.  Add links
  to the sections.

5
  I propose that the current/former WG member lists for RDF Core and
  earlier WGs be added to all our WDs.

  (might have to label this section as informational, but that seems
  rather pedantic.  I think the W3C style guide says how to do this.)

6
  To be completed.

  We aren't normative on charmod.  or c14n?


General
   - some spelling needs to be in US english.  "labeled" maybe
   and w3c doc style, refs, but you know that.

Dave

Received on Thursday, 25 July 2002 10:35:05 UTC