XLink: behavior must go!

[John E. Simpson]

At 08:35 AM 05/17/1999 +0100, Martin Bryan wrote:
>[I] wrote
>
>>Would it be fair to say that the behavior attribute of an XLinking element
>is intended to function as a sort of localized (element-context-specific)
>PI -- a trigger to XLink-smart processors that is ignorable by others? If
>so, in what way is such an attribute more problematic than actual PIs?
>
>The key point about the behaviour attribute is that it is instance specific,
>rather than element-context-specific. (You can deal with all
>element-context-specific problems in the stylesheet; it is instance specific
>changes that are the difficult ones to control without passing some form of
>control parameter.)

Instance-specific? That sounds totally wrong to me, even given your own
suggestion for the attribute's value (i.e., a URI). "Instance-specific"
says to me, "true of this document instance," not "true of this element
instance" -- the latter's what I meant when I said
element-context-specific. In the latter case, a style sheet could *not* be
used to establish behavior because there'd be no selector on the specific
occurrence.

It'd be like relegating href value assignment to a style sheet, no? The
value to be used for a specific occurrence of that sort of behavior is not
obtainable from the "meaning" of the tree structure at that point. Which
value to use would be known to the author (human or otherwise), as a
function of the "meaning" of that occurrence of that element, but unable to
be communicated to a downstream application using other XLink (or style
sheet selector) means.

At 09:04 AM 05/17/1999 -0400, Simon St.Laurent wrote:
>I'm starting to wonder if it isn't time to wait and see what the next
>(supposedly soon) draft says.

Yeah. I regret raising the question when I did... think I'll shut up now
and see what the WG's come up with for ver. 2.
==========================================================
John E. Simpson            | The secret of eternal youth
simpson at polaris.net        | is arrested development.
http://www.flixml.org      |  -- Alice Roosevelt Longworth

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)