Re: Some comments on GRDDL Spec

On Fri, 2006-11-10 at 17:00 +0000, McBride, Brian wrote:
> I have been reading
> 
> http://www.w3.org/2004/01/rdxh/spec revision 1.160
> 
> A few comments below.  I've type labelled the comments - typo, language,
> grammar, style etc are editorial comments.  They are intended as helpful
> suggestions to the editor and as such can be ignored if the editor
> doesn't find them useful.  But let me know if that is the case so I
> don't waste any time producing comments that aren't helpful.

It's always helpful for WG memebers to read carefully and suggest
changes. Numbering your comments and separating them into
typo, language, and style is probably overkill. I'm happy
to get one big pile of editorial suggestions in a message.

(For technical comments, i.e. changes that are observable
from a test, I prefer separate messages that include test
case sketches. You don't seem to make any technical comments
here; I'm just including this note for completeness.)

I made several of the edits you suggest in Revision: 1.163.

Details about which edits I made are below...

> 1. Intro
> 
> 1) Typo and necessity
> 
> [[
> Some offer more formally defined semantics and others more
> loosely-couple
> ]]
> 
> Should be "coupled"
>
> Is this sentence necessary?  It doesn't contribute usefully to the
> argument or description.

OK. struck.

> 2) language
> 
> [[
> Recently, two progressive encoding techniques have emerged
> ]]
> 
> I don't understand what makes these techniques progressive.  Is that
> "progressive" in the political sense?

I'm not familiar with the political sense of the term "progressive",
but several people have commented on it.

> To me "progressive encoding" is a technique like that used in
> representing images which allows them to be rendered with increasing
> fidelity as the data is received.

I'm striking the whole sentence about RDFa and microformats;
I think they're covered sufficiently elsewhere. They'd need
citations if they stayed.


> 3) language
> 
> [[
>   overlay additional semantics
> ]]
>
> Again the meaning of this is not clear to me.
> 
> Maybe something like:
> 
> [[
> Recently two new techniques have emerged for embedding structured
> information in XHTML documents.
> ]]
> 
> 4) grammar
> 
> [[
> Whilst this breadth of expression ...
> ]]
> 
> What breadth of expression?  I don't see any statement about "breadth of
> expression" to refer to.

" represent everything from poetry to prose, purchase orders
to invoices, spreadsheets to databases, schemas to scripts, and linked
lists to ontologies"

> 5) language and style
> [[
> to codify both common
> ]]
> 
> I have a personal aversion to "verbising" nouns.  However there is a
> verb "to codify" in the dictionary but its definition doesn't match the
> use here http://dictionary.reference.com/browse/codify
> 
> "to represent" would be better.

OK.

> 6) language
> 
> [[
> both common and customized meanings
> ]]
> 
> What is a "customized meaning"?  It would be better to avoid this sort
> of use of the word "meaning" as its meaning isn't clear.
> 
> Better would be "inspiring new dialects to represent information".

OK.

> 7) language
> 
> [[
> , it can prove to be a barrier
> ]]
> 
> "prove to" is redundant.

OK. struck.

> 8) language
> 
> [[
> , it can prove to be barrier to understand across different domains or
> fields.  How, for example, does software discover the author of a poem,
> ...
> ]]
> 
> I don't see how the example relates to the barrier to understanding.
> "Understanding" is not something that software does so this looks like a
> category error to me.  And I don't understand what the barrier to
> understanding is.  Understanding is something that human beings do and I
> don't understand how RDFa and microformats are a barrier to human
> understanding.  I don't' understand what RDFa and microformats have to
> do with spreadsheets and ontologies.
> 
> Then follows four examples that don't get referred to locally in the
> text and so I don't see what they are for.
> 
> So looking at that first section, perhaps the editor might prefer
> something along the following lines:
> 
> [[
> XML documents on the web use many dialects.  There are dialects of
> XHTML, XML and RDF that are used to represent many kinds of information
> including poerty, prose, invoices, spreadsheets, ontologies and many
> others.  Recently two new techniques, microformats and RDFa, for
> representing structured information in XHTML documents have emerged.
> These are simple, open data formats built upon existing and widely
> adopted standards.
> 
> With these new data formats a variety of structured information can
> easily be encoded in XHTML documents.  This however creates problems for
> software developers.  How are they to parse and represent this wide
> variety of structured information in their programs?  How are they to
> cope with the fact that the same information can be represented in
> different ways in different dialects? 
> 
> A solution to this problem is to use a common standard for representing
> information in a form that is convenient to process and to enable the
> designers of dialects of XML to publish transforms from their dialect to
> this common standard.
> ]]
> 
> Then again, he might not.

Hmm... no, I don't thik that text sets up the example as well.


> This commenting business takes longer than I like.  I've run out of time
> this afternoon, the editor may be pleased to learn.

;-)

> Brian

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Wednesday, 22 November 2006 15:00:45 UTC