W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 1998

Re: comments on latest guidelines

From: Charles McCathieNevile <charlesn@sunrise.srl.rmit.edu.au>
Date: Fri, 24 Jul 1998 11:30:02 +1000 (EST)
To: Nir Dagan <nir.dagan@econ.upf.es>
cc: w3c-wai-gl@w3.org
Message-ID: <Pine.SUN.3.91.980724111723.2284B-100000@sunrise.srl.rmit.edu.au>
My comments interspersed:

On Fri, 24 Jul 1998, Nir Dagan wrote:

> * Concerning color blindness, a technique that is missing is:
> 
> Do not use images as text since users cannot override 
> the colors (or size) of images. Use stylesheets to color your text. 
> Color blind users may override all author's defined style-sheet 
> genereted colors. (and low vision users will use the appropriete font size)
> 
> * In B the technique: "...and bit-maptext with alt-text (for special text
> effects) may be used,.." contradicts what I said above. Images as text are 
> a big pain.  Look it up in WebTV's guide:
> http://developer.webtv.net/docs/xplat/Xplat.htm
> 
CMcCN::

This seems a question of general principle - the tables for layout debate
is the pointy end. 

The specific concern of the WebTV is that TV provides poor resolution. It 
should (as a UA) give access to the ALT stuff. In my view the use of 
bitmapped text with proper ALT is a hack, but I do not know if it 
produces any accessibility problems. (If WebTV doesn't give access to ALT 
text, then there is a problem. Mostly with WebTV, but we shoudl examine it)

ND::
> * concerning the technique: "Avoid 
deprecated elements >  and attributes... (TT ..."
> TT is not deprecated.

CMcCN::
I noticed this last night. We should just remove the reference to it.

ND::
> * the guideline "Enable keyboard operation of all page
> elements." 
> This is quite confusing for me as an author. I do not know 
> whether the user has a key board or not or how he navigates or 
> operates anything, and I realy shouldn't care. It would be better to 
> say "do not assume the user uses a GUI with a mouse. It is also 
> more consistent with the details.
> (also WebTV's guide is anti-mouse and hates image maps, 
> although TV is a graphical medium) 

CMcCN::
Somebody else thought this was not good enough. The basic principle is 
that any assistive technology can mimic keyboard events, so a page which 
is keyboard accessible is as good as it can get (from the point of view 
of guideline C - control device independence). I think it is better 
expressed in teh main Guideline. A large part of it is covered in section 
E, dealing with User Interfaces which are included in pages, and making 
applets (etc) accessible. It could still be cleaned up

ND::
> * How do I use D-link with FRAME?

CMcCN::
<BODY><A HREF="dframe1.htm" TITLE="description of this frame and its 
place in the universe">D</A> 

ND::
> * the rationale for using meaningful link phrases applies 
> to all users. Many users scan the page for links and have 
> to slow down if the descriptions are "click here", "click there".
> This is not a particular problem of blind users.
> (actually 95% of the guidelines increae usability for all users)
> 
> On the technique page:
> 
> In the ASCII art bit. the STYLE element should 
> not be inside a comment. Its content may.
> 
> Concerning:
>  "Ensure that pages are readable and usable 
> without style sheets for browsers that do not support 
> them or users who deactivate them. Since style
> sheets are a new phenomenon, older browsers will not 
> support them and it will take a while for new browsers to 
> support them in a standard way." 
> 
> The fact that stylesheets are a new phenomenon 
> has nothing to do with why HTML documents shouldn't 
> depent on them. 
> 
> Authors are not expected to write stylesheets 
> for all media (also new media can emerge after 
> a page is written),  and users can override the author's 
> stylesheet. Therefore pages should always work with 
> the user's default.
> 
> Concerning examples:
> 
> There should be one form, not a table version 
> and text version. I use the text version since I have to scroll
> horizontally to read the table one. The guidlines encourage
> to attempt to write one accessible version.   
> 

CMcCN::
I pretty much agree with all that. An important guideline we should be 
including is do not write !important stylesheets, since any 
CSS1-compliant browser may not allow the user to override them

ND::
> Nir Dagan                            
> Assistant Professor of Economics      
> Universitat Pompeu Fabra
> Barcelona (Spain)
> 
> email: dagan@upf.es
> Website: http://www.econ.upf.es/%7Edagan/
> 
> 
Received on Thursday, 23 July 1998 21:51:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:46:58 GMT