Re: Comments on RFD/A Primer

Bob,

Thanks for these detailed comments. We'll be sure to integrate them  
into the next revision of the primer, which I'm hoping to get to  
within a couple of weeks.

-Ben

On Nov 2, 2005, at 4:43 PM, DuCharme, Bob (LNG-CHO) wrote:

> Looks good. These comments focus more on the document's role as a  
> primer than on the technical ideas presented. While it does a good  
> job of telling RDF people how to incorporate RDF concepts into  
> XHTML, I understand a key goal of RDF/A to be making it easier for  
> XHTML people to adopt the concepts of RDF into their XHTML, so I  
> tried to focus on the potential perspective of an XHTML author  
> who's heard of RDF and is thinking of maybe (or maybe not!)  
> adopting RDF/A. Being a primer, it  needs to do a little more hand- 
> holding.
>
> Abstract:  "The reader is expected to be fairly familiar with  
> XHTML, and somewhat familiar with RDF <insert>and the n3 notation  
> of RDF</insert>."
>
> Purpose of RDF/A: A primer for something with such a cryptic name  
> should explain what the name means before starting in on syntax.  
> (If there's no simple explanation of what the "/A" means that would  
> make sense to mythical XHTML author mentioned above, maybe it needs  
> a new name...)
>
> 2.2  "This clickable link has an intended semantic meaning: it  
> <delete>is<delete><insert>references</insert> the document's license."
>
> (The link isn't the license!)
>
>  In RDF/A, the href attribute can contain a Compact URI (CURIE),  
> which makes the expression of <delete>numerous</delete> metadata  
> statements far more compact.
>
> (The antecedent of "which" is "a Compact URI", singular, and I  
> don't believe a single CURIE can contain numerous metadata  
> statements.)
>
> 3.2  "In certain cases, one wants to include multiple RDF elements  
> in a given document"
>
> Instead of the phrase "RDF elements" I would either say "multiple  
> RDF/XML elements" or change the word "elements" to something else.
>
> "Note how the meta and link apply to the parent node's about if  
> that about exists."
>
> This kind of language (e.g. the use of "meta" and "about" as nouns)  
> is OK for casual discussion, but inappropriate for a spec or for a  
> primer aimed at people approaching this material for the first  
> time. I'd suggest something like this:
>
> "Note how the meta and link elements apply to the subject  
> identified by the parent node's about attribute, if such an  
> attribute exists."
>
> 3.3 " it's quite easy to see how XHTML elements might represent RDF  
> entities on a regular basis"
>
> What's an RDF entity? Also, "on a regular basis" implies "regularly  
> over time" to me. Would it be more accurate to say "easy to see how  
> a wide variety of XHTML elements might store RDF metadata statements"?
>
> 4.2 Add one more sentence telling the mythical XHTML author why  
> reification is useful. e.g. "This can be useful to add metadata  
> about metadata to a document, such as the date that the licensing  
> metadata was added, as shown above, or the name of the person who  
> added some metadata".
>
> In general, I think RDF/A looks great, and it will provide a  
> superior alternative to practices like the commented-out RDF/XML  
> I've seen in MoveableType XHTML. To appeal to the people who make  
> such decisions, though, I think the Primer needs to balance all the  
> "we didn't forget this corner case" parts with more "look how easy  
> it is" parts.
>
> thanks,
>
> Bob DuCharme
>
>