[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Re: Can XLink be fixed?

• To: "'Simon St.Laurent'" <simonstl@s...>, <xml-dev@l...>
• Subject: RE: Re: Can XLink be fixed?
• From: "Didier PH Martin" <martind@n...>
• Date: Sat, 17 Aug 2002 16:34:46 -0400
• Importance: Normal
• In-reply-to: <r01050300-1015-AC536F69B21211D683CE0003937A08C2@[192.168.124.21]>

Hi Simon,

Simon said:
"XLink is a W3C Recommendation and therefore you must abide by it"
doesn't seem like a particularly reasonable thing to say to people whose
comments on XLink appear to have been rejected out of hand with little
explanation.

Given my prior experiences with this WG, I'm prone to take all of its
comments with a ton of salt, but a coherent explanation of the why
behind the spec might help wash some of that salt away.


Didier replies:
Like you I am not always please with what's coming out of W3C but most
of all I am not pleased at all having face a schizophrenic behavior and
no consistency. If W3C came out with a link specs, good or bad, its
better that other W3 specs use it for internal consistency or, if a
major breakdown is discovered, that the xlink spec is updated. I won't
defend each groups but I want to know, in a structured document, the
reasons why Xlink is a showstopper for XHTML. It would help to
understand why xlink is broken. In science, (again, sorry for our
philosophy allergic fellows) it seems to be better to take the
epistemological(1) approach recommended by Karl Popper(2). Thus, to try
as much as possible to prove that a theory is wrong instead of trying to
prove it is right ( a major exception to this are mathematics).
following this approach, it is better to know what is the experiment or
experience that leads to invalidate the xlink theory/specifications than
to know why xlink is validate (simple economy of thoughts and time).
Let's pretend, at first, that xlink is useful and OK and then let's see
why its not useful and OK by use cases, experiments, experiences. 

The SVG WG used xlink, and, a in day to day practice, it doesn't look
like a showstopper. So, SVG, up to date, didn't invalidate the xlink
theory/specifications.

Off course, we may not like the namespaces or the way namespaces are
handled but it remains that from W3C, as an institution, we expect
consistency and coherency. The actual problem is that a particular W3C
WG says that there are some showstoppers with Xlink, a respected member
of this WG, Ann assured us that we'll have a document stating the
reasons why such breakdown is happening (and thus invalidating the xlink
theory/specifications). With this document that I guess will be better
structured than a puzzle to be assembled from scattered emails, we'll be
able to make our mind and we will be in a better position to understand
why xlink is invalidated in the case of XHTML.

Off course if the only leitmotiv is political and Byzantines fights,
there are little hopes that a sound and rational process is followed.
However, I hope this is not the case and will place my trust in Ann's
words; we'll get a document giving us the answers. 

(1) http://www.dictionary.com/search?q=epistemological
(2) http://www.eeng.dcu.ie/~tkpw/

Cheers
Didier PH Martin

• Follow-Ups:
  • Re: Re: Can XLink be fixed?
    • From: Uche Ogbuji <uche.ogbuji@f...>
  • Re: Re: Can XLink be fixed?
    • From: Arjun Ray <aray@n...>

• References:
  • RE: Re: Can XLink be fixed?
    • From: "Simon St.Laurent" <simonstl@s...>

• Prev by Date: Re: linking, 80/20
• Next by Date: RE: Tearing the Hyperlink Problem Apart (was Re: Re: Can XLink be fixed?)
• Previous by thread: RE: Re: Can XLink be fixed?
• Next by thread: Re: Re: Can XLink be fixed?
• Index(es):
  • Date
  • Thread