W3C home > Mailing lists > Public > www-html@w3.org > January 2000

Re: Status of Iframe in XHTML 1.1

From: Murray Altheim <altheim@eng.sun.com>
Date: Fri, 14 Jan 2000 12:02:14 -0800
Message-ID: <387F80C6.58A23B10@eng.sun.com>
To: JOrendorff@ixl.com
CC: www-html@w3.org
JOrendorff@ixl.com wrote:
> 
> Murray,
> 
> you wrote:
> > JOrendorff@ixl.com wrote:
> > > > So, what is the fate of inline frame?
> > >
> > > Dropped from XHTML 1.1, unless a later draft resurrects it.
> >
> > Please read the XHTML 1.1 specification more carefully.
> > In the Abstract, Introduction, particularly Section 3. "The
> > XHTML 1.1 Document Type" and finally in the table in Appendix
> > A. "Changes from XHTML 1.0" describe XHTML 1.1 as explicitly
> > *not* providing features found in the HTML Transitional or
> > Frameset DTDs [1].
> 
> Respectfully, Murray, I think you overestimate the clarity and
> explicitness of W3C specs.  ;^)

Well, you don't need to argue with me about that. Everyone is so
incredibly pressed for time, IMO, that it's often difficult to 
write, edit and obtain suitable feedback. And as you may well
imagine, it's sometimes difficult to write anything by committee. 

> None of the places you cited
> actually come out and say that the features of the Frameset
> DTD are deprecated.  Section 3 comes the closest:  "It (XHTML 1.1)
> is not, however, as varied in functionality as the XHTML 1.0
> Transitional or Frameset document types."  (The other places
> do not mention "transitional" or "frameset" at all; the words
> "legacy" and "deprecated" are used instead.)
> 
> Nowhere, in any W3C document I've been able to find, are the
> frameset features formally deprecated.

Maybe this is where this whole thing is confused. Something doesn't
have to be deprecated to not make it into the next version of a spec,
especially since it makes little sense to go back into an earlier 
document and subsequently add in a deprecation.

XHTML 1.1 is simply a new document type based on XHTML 1.0 Strict,
itself based on HTML 4.0 Strict. We added in <applet>. The 'target'
attribute, frames, etc. were never part of Strict, so they didn't 
make it into 1.1. 

I note that in the XHTML 1.1 spec it says "Most significant is the
removal of features that were deprecated." This is an attempt to 
describe the transition from HTML 4.0 and XHTML 1.0 that were
specified via three DTDs, to XHTML 1.1 specified by one. I agree
that this is confusing the issue, and we'll look for an alternative.

> One has to check the table in the appendix, or read all the
> module descriptions, or fish through the DTDs, to be sure.

No, there are two places. If you look in the lists of elements and
attributes in HTML 4.0 [1][2] you can ascertain what is in Strict, 
then the changes table in XHTML 1.0 [3].

> Another thing has been bothering me.  If all this has been stable
> and well-known for so long, why isn't the change-- the most
> significant difference between XHTML 1.0 and 1.1 and arguably
> the biggest change ever to HTML-- mentioned in the "Future
> Directions" section of XHTML 1.0?

What change? Dropping of Transitional (named appropriately) and
Frameset features? We have tried to describe this in some detail,
and since HTML 4.0 it has been stated many times that these features
are either more appropriate to stylesheets, very problematic for
accessibility or i18n reasons, or have never had wide or consistent
implementation. If people want frames, they can continue to write
frame documents in HTML 4. We don't expect them to be supported 
in XML environments, and the vendors in the consortium have been
pretty consistent about that. The idea of Transitional has always
been that the features therein would eventually be dropped. We're
now dropping them in XHTML 1.1.

> Politics?

No, probably time. Many of us are involved in quite a number of
activities. I was talking to Masayasu Ishikawa when we were in
Munich, and I think he's on something like ten or twelve working
groups. I can't claim to that level of effort, but trying to hold
down the fort is more likely than politics. Sleep is a good thing too.

If there's any politics, it's that there's obviously a demand for
things like frames, but unless the world goes to one browser (and
it's actually moving away from two in the other direction, finally),
we can't rely on frames for the reasons mentioned above. People like
sugar but it's bad for their teeth, and it's impolitic to say "you
can't have sugar." We're trying to be responsible to the accessibility
community, and to those whose frame experience has been bad, in the
vendor, content provider, and user communities. But we're also not
trying to hide that fact.

And I'm sincerely apologetic about any failings in our specs. I'll
talk with Shane McCarron about trying to make this particular issue
more clear in the XHTML 1.1 draft.

Thanks,

Murray

PS. You might note that when we begin using XLink in XHTML 2.0, there is
a feature similar to target="_blank" that looks like: xlink:show="new".
There are really a lot of cool, exciting features coming with use of
XSL stylesheets in XHTML documents, truly professional publishing features,
and I'm working on adding interoperable navigation features [4], too.

[1] http://www.w3.org/TR/html401/index/elements.html
[2] http://www.w3.org/TR/html401/index/attributes.html
[3] http://www.w3.org/TR/xhtml11/changes.html#a_changes
[4] http://www.ornl.gov/sgml/sc34/document/0058.htm
...........................................................................
Murray Altheim                                   <mailto:altheim@sonic.net>
Member of Technical Staff, Tools Development & Support
Sun Microsystems, Inc. MS MPK17-102
1601 Willow Rd., Menlo Park, California 94025  <mailto:altheim@eng.sun.com>

   the honey bee is sad and cross and wicked as a weasel
   and when she perches on you boss she leaves a little measle -- archy
Received on Friday, 14 January 2000 15:08:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:40 GMT