W3C home > Mailing lists > Public > www-xml-linking-comments@w3.org > January to March 2000

Re: XLink 1999-12-20 WD questions

From: Eve L. Maler <elm@east.sun.com>
Date: Fri, 28 Jan 2000 15:45:18 -0500
Message-Id: <>
To: www-xml-linking-comments@w3.org
 >1. The WD includes various "sample" declarations of XLink 
elements/attributes. Is it safe to assume that the term "sample" means 
"non-normative"? In particular, I'm interested in the declaration of the 
role attribute. In several (all?) places, this is declared as being of type 
NMTOKEN; would it be possible/desirable to allow it to be of type NMTOKENS 
instead? This would permit a given resource to be associated with more than 
one arc at a time.

Yes; we should probably just say non-normative wherever we currently say 
sample.  The only constraints that are normative are the ones expressed in 
prose with "must" and, particularly, the ones expressed in the conformance 
constraint notes.

Because one of these constraints requires the role value to be a QName, 
NMTOKENS is not possible.

 > 2. What was the thinking behind the option of using XLink *elements* 
instead of XLink attributes? This seems rather perverse; at least, it goes 
against the grain of the general "XML shall be easy to process" design 
goal, since an XLink-aware application must now keep an eye cocked in both 
directions at once. It makes XLink harder to "grok" (providing *more* ways 
to perform a given task -- especially, as far as I know, without any marked 
advantage to one way over another -- *never* simplifies things). And it 
encourages the use of the (to my taste) horrid ANY content model.

You'll notice that the element option has disappeared...

 > 3. Minor editorial suggestions:

 > a. 1.3 [Terminology] In definition of "resource," change phrase "...that 
is participating in a link" to "...that may participate in a link." (As it 
stands, this "abstract sense" definition matches the definition of 
"participating resource": "A resource that takes part in a link.")

This is fixed in the January 19 version, in a rather different way.

 > b. 3.1.4 [Traversal rules for an extended link] Paragraph immediately 
preceding "Constraint: No Arc Duplication":
	...behavior of such a link with is the same....
 > Presumably "with" should be struck there.


 > c. 4.3 [Conformance Requirements for XLink Elements] In the numbered 
item 1, strike the word "either" in the first bullet. ("Either" already 
appears in the parent numbered item.)

This is fixed in the January 19 version.

Eve Maler            Sun Microsystems
elm @ east.sun.com    +1 781 442 3190
Received on Friday, 28 January 2000 15:44:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:21 UTC