W3C home > Mailing lists > Public > www-tag@w3.org > July 2002

Re: xlink:href

From: Christopher B Ferris <chrisfer@us.ibm.com>
Date: Fri, 12 Jul 2002 18:00:51 -0400
To: Tim Bray <tbray@textuality.com>
Cc: Paul Prescod <paul@prescod.net>, Steven Pemberton <steven.pemberton@cwi.nl>, www-tag@w3.org, www-tag-request@w3.org
Message-ID: <OF1F6DDA5B.7152B977-ON85256BF4.0078E023@rchland.ibm.com>


Christopher Ferris
Architect, Emerging e-business Industry Architecture
email: chrisfer@us.ibm.com
phone: +1 508 234 3624

                      Tim Bray                                                                                                    
                      <tbray@textuality        To:       Steven Pemberton <steven.pemberton@cwi.nl>                               
                      .com>                    cc:       www-tag@w3.org, Paul Prescod <paul@prescod.net>                          
                      Sent by:                 Subject:  Re: xlink:href                                                           
                      07/12/2002 05:39                                                                                            

Steven Pemberton wrote:

> I think it is wrong to characterise the HTML and SYMM groups as some sort
> villains in this piece. It just demonstrates the importance of reaching
> consensus in the specification-making process. Both WGs reported many
> that the linking group were not addressing their issues, and the linking
> group seriously suggested that HTML should not use XLink since HTML
> had perfectly useful and successful linking constructs (I'll dig out the
> if you are interested). The HTML WG did not agree with this approach, but
> there you are. If a WG doesn't listen to its clients, it will produce a
> product that doesn't match the requirements of it.

Just to clarify the history, there was a substantial and never-resolved
disagreement between the XLink and HTML WGs.

The HTML WG read the XLink charter to mean that XLink should be designed
so that HTML links as they currently exist could automatically be
treated as XLinks based on external declarations, for example in the DTD.

The XLink WG persisted in believing that they were chartered primarily
to build a next generation of hyperlinking markup that could be
recognized in the instance and used to express things that HTML's A,
IMG, and LINK elements just couldn't do, and failed to find a good way
to do this that also grandfathered the existing HTML syntax.

An aggravating factor was the incredibly slow pace of operations of the
XLink working group (I speak as its one-time chair); a combination of
factors including personalities caused it to waste literally years at a

HTML today has pretty well the same repertoire of hypertext markup that
it's had since day 1, and XLink is substantially ignored.  Really quite

I think that this issue is worth investing some TAG time in.  My
reasoning may be naive, but here it is:  Today's HTML hyperlinks changed
the world, even though they are metadata-light, single-ended, and
without builtin indirection.  If you could add some metadata, linkbases,
and multi-endedness without compromising the web architecture, the world
might get changed again.  XLink is the best attempt yet in this
direction.  If the idea is basically wrong, we should say so.  If it's
OK but the design needs fixing, we should say so.  If it's right, we
should apply TAG pressure to enrich the Web's repertoire of linking

I can't imagine an issue which is more central to the Web architecture.
Received on Friday, 12 July 2002 18:07:27 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:55:52 UTC