Re: new TAG issue TagSoupIntegration-54

/ Mark Baker <distobj@acm.org> was heard to say:
| I maintain though, that the situation Ian describes is desirable; if a
| publisher intends XHTML/XML semantics, then they should use
| application/xhtml+xml.  So if Alice had sent the document to Bob as
| application/xhtml+xml, then Bob would have said "Sorry Alice, I can't
| handle that".  Even if Bob had ignored the media type, done his edits,
| then sent it back to Alice as text/html, her UA should process it as
| HTML, not XHTML (unless she overrides that, in which case it's her
| fault).

I'm sympathetic to that point of view. I want tools that work that
way. But I'm...not like most people in this regard. I see the benefits
of XML because I've been taking advantage of them for more than a
decade and because I get a sense of personal satisfaction from
knowing that my data is marked up clearly and precisely.

To your average user, for whom all this angle bracket stuff is just a
means to an end, the end being his or her actual day job, the benefits
of namespace well-formed XHTML over HTML tag soup that works in the
browser just aren't apparent.

I think I imagined that in a post-XHTML world, vendors of editors,
browsers, and other UAs would be so attracted to a world in which they
could discard the enormous complexity of error recovery that they'd be
fighting with each other to find the quickest and easiest way to lure
users to migrate away from tag soup.

/me pauses until the howls of laughter subside to a dull roar

The economic realities didn't actually play out like that.

I think the TAG has a responsibility to consider the reality of the
situation in some detail. Hopefully, we can have this discussion in a
calm and rational manner without politics or histrionics.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh
XML Standards Architect
Sun Microsystems, Inc.