ANN: Draft Sildeset for Links and Relations in XML

[John Cowan]

Ingo Macherius asks:

> 4 : Is the notation urn:http://foo.com/ correct ?

It is formally correct, but I have not seen any drafts documenting
such a thing as URL-based URNs.

> 6 : What really is the difference between document and group ?

An extended link group is just a kind of extended link, and an
extended link document is just a kind of locator.

The point about extended link groups is that (unlike other extended
links) they necessarily refer to whole documents, and their
semantics is that the documents are associated in some way.

> Do you consider the visualization an oversimplification ?

Yes.  "Locator" should not have a separate picture, and
"document,group" should just be "group" and should show a
part of a document that refers toa bunch of *whole documents*
(shown in reduced size).

> 7 : Any idea how to generally visiualize the concept of a 
> link without discriminating inline/off-line and simple/extended?

No, nor do I know how to show a picture of apples which does
not discriminate between "two apples" and "many apples".  Sometimes
a word is worth a thousand pictures.

> May a locator really be a link of type locator?

A locator is just part of the machinery of an extended link; it
has no independent use.  So locators are not links.


> 9 : Have I missed a feature ?

Relative locations should be at least mentioned, even though they
are explained on the next slide.

> 10: Formalisms tend to be erroneous, please check.

The "preceding" list should be 1, 2, 5; the "following" list should be
7, 2, 3, 4, 8, 12, 9, 1.

You can see this by writing out the XML format:

<X1>
  <X2>
    <X5></X5>
    <X6>
      <X10></X10>
      <X11>
        <X13></X13>
      </X11>
    </X6>
    <X7></X7>
  </X2>
  <X3></X3>
  <X4>
    <X8>
      <X12></X12>
    </X8>
    <X9></X9>
  </X4>
</X1>

and applying the rules given at 3.3.7 and 3.3.8.

Ancestors always both precede and follow; descendants never
precede or follow.

> 11: " " " " " " " .

2nd example is in error.  "Root()" gets the "speech" element;
the second child counting from the right is the content
"Fare you well, my lord", and then you want the 3rd sibling
previous to that --- but there are only 2 siblings previous.
Change to "root().child(-2,#all).psibling(2,#all)".

3rd example is also in error.  The term "descendant(-1)" means
"descendant(-1,#element)" and cannot select a string.  You
will get the whole treewise rightmost element, namely
"<direction>To Ros.</direction>".

4th example is also in error.  Since neither "root()" nor
"origin()" is present, "root()" is assumed, but nothing
follows the root, since descendants (see last slide) are
neither following nor preceding.

-- 
John Cowan	http://www.ccil.org/~cowan		cowan at ccil.org
	You tollerday donsk?  N.  You tolkatiff scowegian?  Nn.
	You spigotty anglease?  Nnn.  You phonio saxo?  Nnnn.
		Clear all so!  'Tis a Jute.... (Finnegans Wake 16.5)

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/
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)