Fw: Namespaces

[Mark Birbeck]

James, hope you don't mind me copying this to the group. I'd like to
know what the other James thinks.

James Anderson wrote:
> James Clark's notation is readily extended to describe the same 
> things as you
> desire. See my note in this thread. I suggest that it is to 
> be preferred to
> the A.3 notation as it is compacter - it can, in a sense, be 
> expressed 'in
> line'.
> For reasons which I do not fathom, (see his reply to same) he
> explicitly resists the extension and disregards the concepts 
> established in
> appendix a.

I don't follow you James. Where does he disregard appendix A?

On the extension, I think he is right to resist it - although to be fair
it is only a shorthand, not a new standard! If you think about what he
is representing with his current stuff, he has devised a simple way of
giving us clues as to how some post-parser software might be able to use
the data. Given:

	<n1:good a="1" />

and mapping it via James's syntax to:

	<{http://www.w3c.org}good a="1" />

the post-parser can say that 'good' is part of the http://www.w3c.org
namespace. Equally, given:

	<good n1:a="1" />

and mapping it to:

	<good {http://www.w3c.org}a="1" />

James has given us a further insight into the mind of the post-parser;
it now knows that 'a' is part of the http://www.w3c.org namespace too.
In other words, James's syntax is so far an accurate representation of
what the post-parser knows.

However, if we now 'extend' his syntax to map:

	<n1:good a="1" />

to:

	<{http://www.w3c.org}good {{http://www.w3c.org}good}a="1" />

as you seem to want (or at least Oren did, and I guess you are agreeing
with him although I can't find your original message), we've now
expressed something that is no longer consistent with what the
post-parser knows. We are either saying that 'a' is a member of the
namespace {http://www.w3c.org}good (if we treat the outer curly braces
as representing a namespace name) but that is not a valid uri. Or we are
saying that 'a' is a member of the namespace 'good' which is a member of
the namespace http://www/w3c.org, which is (sort of) closer to what is
going on in the intern representation of the XML namespace (according to
appendix A) but is meaningless at this level, because James is talking
about global namespaces, and so we need a uri. James did spell that out
when he introduced his shorthand. (Even if we allowed this, I would have
thought that the syntax {http://www.w3c.org}{good}a would be
preferable.)

In fact what the post-processor knows about the attribute 'a' is that it
has no uri prefix and is part of the element 'good'. That *can be*
expressed with James's syntax, provided that you keep 'a' with 'good'
(as I said before, it is like an 'exploded' XML document):

	<{http://www.w3c.org}good a="1" />

but James's syntax cannot be used to express anything about 'a' outside
of this, since you can't have:

	{{http://www.w3c.org}good}a

as we've already said, and:

	a

has lost all context. However, the extended attribute/element syntax I
used - from Appendix A - at least *can* express something about 'a',
without having to be in the original XML document:

	<ExpAName name='a' eltype="good" elns="http://www.w3.org" />

Personally, I prefer the latter method anyway, because it makes it clear
that we are looking at a different view on the data. The XML 1.0 or
parser 'view' of the data looks at elements and attributes, their values
and their relationships to each other. The 'post-parser' view is of the
same information but from the perspective of which namespace an element
or attribute belongs to. It is possible that an element or attribute
that has a direct relationship with an element in the first view has no
relationship to that same element in the second view. Unfortunately,
James's syntax mixes up these two 'views' of the data, by trying to
superimpose the post-parser's view (of namespaces) onto the parser's
view (of data).

Mark Birbeck
Managing Director
Intra Extra Digital Ltd.
39 Whitfield Street
London
W1P 5RE
w: http://www.iedigital.net/
t: 0171 681 4135
e: Mark.Birbeck at iedigital.net


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)