NOFRAMES, WCAG, & UAAG (was Re: noframes & Content Guidelines)

aloha, charles!

the following is something that began as a response to paul booth, to be sent
to the WAI-IG list, but i postponed composition on it about a week or so ago,
as i was sick and woozy when i wrote it...  i'm still sick, but a bit less,
woozy, so here goes...

the problem with classifying NOFRAMES is that the functionality it is intended
to provide bridges two concepts -- access to content and navigation of
content...

therefore, since the NOFRAMES element is a container for alternative content,
and -- since such textual equivalents are a P1 WCAG requirement -- the use of
NOFRAMES is a P1 requirement as well, especially from an "until user agents..."
point of view, as the alternative means of providing non-graphical access to
the individual FRAMEs that comprise a FRAMESET -- LONGDESC -- has yet (at least
to my knowledge) to be implemented in the FRAMESET context...  

NOFRAMES are not only crucial to ensuring the accessibility of a framed site,
but its usability by frames-incapable browsers, as well as browsers (such as
Opera 3.6x) which allow users to disable the rendering of frames...  not
everyone has the hardware to support a graphical user interface, nor is
everyone fortunate enough to have unlimited dial-up access accounts, a free
local calling region, or a dedicated line...  thus, there are millions of
users, who, for a plethora of reasons -- be they physical, financial,
logistical, or philosophical -- choose to, or have no choice but to, access
web-based content using text-based browsers, or browsers, such as Opera 3.6x,
which allow users to turn off rendering of frames...  the easiest way for
individuals in such situations (and they number in the millions) to access
framed content is through the use of the NOFRAMES element...

as the HTML4 spec [reference 1] states,

quote
The NOFRAMES element specifies content that should be displayed only when
frames are not being displayed. User agents that support frames must only
display the contents of a NOFRAMES declaration when configured not to display
frames. User agents that do not support frames must display the contents of
NOFRAMES in any case. 
unquote

so, to answer one of paul's original questions, how does one properly use the
NOFRAMES element?

in order to avoid content replication, the links contained in the navigation
frame should be replicated in a structured manner in the NOFRAMES section
defined for the FRAMESET...  note that, by the phrase structured manner, i mean
that one would replicate the navigational links in either an ordered,
unordered, or (where appropriate) definition list...  one can also add other
content (such as a disclaimer, usage policy, or quick search form) which may
appear in other quote static unquote FRAMES that have been defined for the
FRAMESET in the NOFRAMES section, as well, as it is the FRAMESET page
containing the NOFRAMES element that will be bookmarked by the non-frames-using
user, and considered as the FRAMESET's home, or index, page...

as i stated earlier, i believe that the problem most people have in grappling
with the concept of NOFRAMES, in theory and in practice, is that the
functionality it is intended to provide bridges two concepts -- access to
content and the navigation of content -- which also happen to be 2 of the goals
of the User Agent Guidelines [reference 2] -- which is why i have carbon-copied
this to the UA list..  i would like to ask that -- where and if appropriate --
this thread continue on the GL list, to which charles posted the emessage to
which this emessage is a reply, unless the topic is being discussed solely from
a user agent standpoint...

despite the decision taken at the Redmond Face2Face meeting, i still argue that
the ability to render NOFRAMES rather than a FRAMESET is a Priority 1
requirement, under UAAG Guideline 3, for user agents that wish to comply with
the User Agent Accessibility Guidelines...  i also don't think that it is
sufficient that exposition of the NOFRAMES element is currently addressed only
in the UAAG Techniques document [reference 3], and not in UAAG Guideline 2 --
as special cases of UAAG Checkpoint 2.1 and UAAG Checkpoint 2.5 (offering the
user a choice of having the UA render either the contents of the NOFRAMES
element, or a list of LONGDESCs defined for the FRAMES -- especially as a list
of LONGDESCs is a sorry substitute for robust NOFRAMES content......  at the
very least, i would suggest that access to NOFRAMES be mentioned in the UAAG's
definitions of "alternative equivalents/alternative equivalents for content",
especially as it was decided at the Redmond Face2Face [reference 4] that the
checkpoint which originally called for user to be able to "turn off" the
rendering of FRAMES

quote
4.12 Allow the user to choose between a frameset or its alternative supplied by
the author. 
unquote

was dropped, in favor of adding a note on access to frames to what was (in the
22 October 1999 draft) then enumerated as Checkpoint 3.1

quote
3.1 Ensure that the user has access to all content, including alternative
representations of content. [Priority 1] 
unquote

but the mention of frames -- which was added to the introductory text for
Guideline 3 in the 22 October 99 draft [reference 5]

quote
Mechanisms for specifying alternative content vary according to markup
language. For instance, in HTML or SMIL, the "alt" attribute specifies
alternative text for many elements. In HTML, the content of the OBJECT element
is used to specify alternative content, the "summary" attribute applies to
tables, etc. In HTML, the NOFRAMES element specifies alternative content for
frames. The ability to access frame alternatives is important for users of some
screen readers and users with some cognitive impairments. 
unquote

does _not_ appear in the introduction to either Guideline 2 or Guideline 3 (or
anywhere else) in the current (15 January 2000) draft, although the note that
ian added to what was enumerated as Checkpoint 8.1 in the 22 October 99 draft
has been retained in the note for the 15 January draft's Checkpoint 7.1

gregory.

References
1. http://www.w3.org/TR/html4/present/frames.html#edef-NOFRAMES
2. http://www.w3.org/WAI/UA/WD-UAAG-20000115/
3. http://www.w3.org/WAI/UA/WD-UAAG-TECHS-20000115/
4. F2F Minutes:
http://www.w3.org/WAI/UA/1999/10/wai-ua-f2f-199910-minutes.html 
    Issues List: http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#91
5. http://www.w3.org/WAI/UA/WAI-USERAGENT-19991022/


At 01:05 PM 1/16/00 -0500, CMcN wrote to the GL list:
>Reposted to w3c-wai-gl@w3.org for the WCAG group to consider. Please followup
>there rather than IG.
>
>Paul,
>
>The main requirement for noframes is in checkpoint 6.5. This is Priority 2
>since in most user agents now it is opossible to access frames, at least to
>the extent of choosing between them. In other words, the working group felt
>that in general a meaningful title and name were more important for access to
>frames than a noframes element with useful content.
>
>One of the problems with a prioritised set of guidelines is that people are
>prepared to do what is essential, and then leave out the rest. In other
>words, it is accepted by some people as good enough if it is ont impossible
>to participate, but merely very difficult. I'm not sure that changing that
>could be done by the working group, and in the long riun i feel that there is
>value in prioritising to a certain extent, since meeting all the Priority 3
>checkpoints may not be as valuable as meeting 2 or three of the P1
>checkpoints. Of course I would hope that people endeavour to make triple-A
>conformance.
>
>Cheers
>
>Charles McCN
>
>On Thu, 13 Jan 2000, Paul Booth wrote:
>
>  Hi there,
>  
>  I've recently been running bobby over a few sites for people and explaining
>  the results to them.
>  
>  I noticed that when run over a site with frames bobby comes up with the
>  guideline: "Ensure that pages are readable and usable without frames."
>  
>  I cant find anywhere in the WAI Web Content Checklist that actually says
>  this however.
>  http://www.w3.org/TR/WAI-WEBCONTENT/full-checklist.html
>  
>  "Title each frame to facilitate frame identification and navigation." is in
>  there as is, "Describe the purpose of frames and how frames relate to each
>  other if it is not obvious by frame titles alone. "
>  
>  Although I did stagger on it somehow (through the mass of links and pages
>  you get from the checklist) in http://www.w3.org/TR/WAI-WEBCONTENT-TECHS/ as
>  point 4.10.3
>  
>  On further inspection, I guess its covered in part in checkpoint 1.1, but I
>  cant find anything that specifically mentions "noframes" in that part of the
>  techniques document.  I know space is a premium, but shouldn't some of these
>  points be given a bit more emphasis?
>  
>  I could see a potential problem that someone who has used frames could just
>  look under the "And if you use frames (Priority 1)" heading in the checklist
>  and think "well my frames all have titles so I'm ok!"
>  
>  Perhaps I'm missing something - I find navigating the techniques document
>  from the checklist quite difficult.
>  
>  Actually, this makes me think of another question:  Is it the case that
>  <noframes> isn't considered as important as it used to be? (hence why its
>  not specifically mentioned in the checklist?)  I appreciate that most
>  browsers cope with them by letting the user select the individual frames in
>  the frameset...  (In which case, is it bobby thats a bit out of sync?)
>  
>  ....Or perhaps I've just been working too late (again).
>  
>  - Paul
>  
>  --
>  Paul Booth, Project Officer, DISinHE Office.
>  The UK's national centre for Disability and Information Systems in Higher
>  Education,
>  Department of Applied Computing, University of Dundee, Scotland
>  w:  http://www.disinhe.ac.uk/   t: 01382 345050      f: 01382 345509
--------------------------------------------------------
He that lives on Hope, dies farting
     -- Benjamin Franklin, Poor Richard's Almanack, 1763
--------------------------------------------------------
Gregory J. Rosmaita <unagi69@concentric.net>
   WebMaster and Minister of Propaganda, VICUG NYC
        <http://www.hicom.net/~oedipus/vicug/index.html>
--------------------------------------------------------

Received on Wednesday, 19 January 2000 15:42:31 UTC