Re: xlxp-dev: XML XLink Requirements Version 1.0

Simon St.Laurent writes:
 > After several readings of the XML XLink Requirements document, I'm both
 > delighted and disturbed.  While I'm pleased to see that several key issues
 > (4.B.7, 4.B.8, and 4.B.10) have finally been raised within the WG, I'm

Simon,
  As someone who's trying to learn more about XLink & XPointer and
apply them within a very specific context, I have more concerns about
the draft than the requirements.  I read the reqs when they were
released, and thought them to be pretty good; nothing jumped out at me 
as being a particular problem.  (I'm neither an academic nor any other 
hypertext expert.)
  Now for my concerns (not terribly related to your post, other than
my being inspired to post them at this point).
  The context for my concern lies in my application.  The XML Bookmark 
Exchange Language (XBEL, http://www.python.org/topics/xml/xbel/)
doesn't include support for links to other bookmark resources.  I
expect that XLink would be the best way to add a reference from an
XBEL instance to "more bookmarks, from over there."  This is clearly
an application of simple links (clear to me, at any rate; I'm sure an
extended link could be wrangled out of it).  What I'd like to have in
an XBEL instance is something like this:

	<link href="...URI for another bookmarks file...#pointer"
	      added="1999-05-22">
	  <title>Title of remote bookmarks resource</title>
	  </link>

  This seems simple enough, but there are several aspects of linking
which are not clearly specified.  Aside from the issue of placing the
title of the remote resource in an element rather than an attribute,
the "show" and "actuate" attributes aren't described with sufficient
clarity.  Specifically, is actuate="auto" intended to mean that the
resource is always loaded, or that the application should decide when
if & it's loaded?  I can live with show="embed" I think, and that can
be #FIXED in the DTD; maybe that's not so bad.  I am concerned that
fixing the value in the DTD may deter applications from loading the
resource as if show="new" or show="replace", and either could be a
reasonable behavior.  I'm not convinced that it would be reasonable to 
allow the behavior to be specified in the instance.
  I'd like the XLink specification to discuss how these attributes
should be handled in this context; is there an attribute that says
"application decides," or does ommission of the attribute from the DTD 
imply that?  (The ommission of various attributes doesn't appear to be 
dealt with; that needs to be added to ensure clarity.)  [If I'm
missing something in the draft, I'd love to be corrected, especially
with a specific pointer into the draft!]
  For now, I can place changes to XBEL on the back burner, but would
like to be able to address the needs of the application appropriately
before too much longer.  (Yes, I understand the timetable for XLink;
I'd just like to avoid creating something that will need an enormous
amount of work to become compatible with the final specification.)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake@acm.org>
Corporation for National Research Initiatives

Received on Monday, 22 March 1999 13:47:10 UTC