Re: XLink 1.1: Xlink vs "legacy" linking

/ Bjoern Hoehrmann <derhoermi@gmx.net> was heard to say:
| Describing a new format's integration into existing infrastructure is
| about the most basic and obvious requirement for W3C Technical Reports.
| I want to make a user agent that is both a conforming XHTML user agent
| and a conforming XLink application. That's either a stupid idea in which
| case this needs to be pointed out in XLink or it's reasonable in which
| case conformance to the specifications must not be mutually exclusive.

I see your point, but I agree with Daniel. The XLink Recommendation
can describe how an XLink processor must behave if processing an XLink
element. Another specification can describe how to behave if
processing an element in its vocabulary.

In the case where an element occurs in both vocabularies and these two
actions are not the same, it's likely that the application will have
to intervene. Perhaps give the user a choice, or a warning?

The XLink specification cannot proscribe HTML behavior on an XHTML "a"
element and reasonably expect to get public support. I suppose it
could defer to another vocabulary in which it appears, but in that
case an XLink processor unaware of the other vocabulary cannot
function.

A user that produces

  <a href="x" xlink:href="y" ...>

has probably made an error, but only he knows for sure.

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM / XML Standards Architect / Sun Microsystems, Inc.
NOTICE: This email message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.