Re: [editorial] white space reference to XML specification

At 10:56 PM 2001-09-24 , Susan Lesch wrote:
>>  The section cited above deals only with 'white space' appearing outside
the
>>  markup.  For the definition of what is to be considered as white space it
>>  depends on the syntax presented earlier in
>>
>>   Common Syntactic Constructs
>>  
<http://www.w3.org/TR/REC-xml#sec-common-syn>http://www.w3.org/TR/REC-xml#se
c-common-syn
>>
>>  which is a better location to point to for the definition as used in XML
>>  1.0.
>
>Good point. I could use that in the future. The point of pointing to
>the XML Recommendation is only to show that it spells "white space" as
>two words. (The link I gave shows that in a heading.)
>

Susan, I don't want to beat on _you_ because you done good.  You didn't
need to
use the precise reference to the 'correct' point of definition.  But in the
Grammar spec, we still do need to.

The point of pointing to the XML specification from your comment is to show
the
difference in spelling between the two documents.

The point of pointing to the XML specification from the Grammar specification
(which I recommended) is a bit different.  The Grammar specification needs to
inherit the technical definition of white space [sic] from the XML
Specification or the Grammar specification is -- a) undefined or b) defined in
conflict with the senior specification -- in the normative things it says
about
white space.  The reference is here a piece of specification code which is
required to construct the specification correctly.  You can cite the whole
document in which case the user knows to search the document for the defining
occurrence.  Or you cite the defining occurrence.  But those are your
choices. 
To document the relationship where there is a dependency you don't cite a
random other occurrence somewhere else inside the document.  That would be
misdirection, because people would expect you are pointing them to the
definition that you need.  Pointing a dependency at anything more local than
the whole configuration item usually connotes a reference to where the thing
you depend on is defined in the defining document.

So, for the purposes of the editors of the Grammar spec, it makes a
difference.  

To point out the problem, it doesn't matter.  In this case, to fix the
problem,
it does matter; otherwise we introduce new problems.

Al

Received on Tuesday, 25 September 2001 00:11:14 UTC