techniques draft

The status should point out that this is a working draft, updating the
NOTE.

It would be helpful to have in the techniques document where the checkpoints
are listed, things that are applicable to that checkpoint. For example
checkpoint 3.1 "Use appropriate markup languages" should provide  list of
languages that are appropriate for basic tasks - Math, Chemistry "general
documents", graphics, music, etc. The information on structure and
presentation is not directly relevant, but further background. Mkaing someone
go to the same trouble to find the direct, simple answers and the explanation
of why to use them seems to be a mistake.

3.4 Use relative sizes...
It is not clear what the relevance of directly accessible applets is.

1.5 and 4.1 both have a link to "text equivalents", but they go to different
places. This does not allow the user to have an idea of where the link leads
without context...

This is also the case with two links under checkpoint 9.3 - keyboard access

HTML TECHS 6.3 - Keyboard access.

An important way of enabling keyboard access is to replicate an onMouseover
with an onfocus that does the same thing (since it will of course be
accessible *grin*) and likewise redundant actions for onblur and onmouseout.
Most of the stuff in 5.4 "Keyboard Access" in the main techniques document
(image maps, etc) seems to belong in the relevant language-specific
techniques.

checkpoint 13.9 - provide information about document collections. It is not
clear whether this is satisfied by providing this information in the document
content or whether it has to be metadata or machine-comprehensible.

5.1 Structure and presentation
It is not true the that XML inherently provides separation of structure and
presentation. WML (the language created for WAP) is a clear counter-example.

5.2 Text Equivalents
I don't think text is accessible to almost all users. Substantial amounts of
text are inaccessible to perhaps most users (i.e. more than half), and we
should not continually fuel the "text-only is fine" approach. Text merely
provides another form of access to many types of content, for users who
cannot (for example) see graphics.

5.2.1 Overview of technologies
At least PNG, JPEG, GIF, WebCGM and SVG allow the inclusion of text. zThey
also allow the inclusion of markup, which can be used to add metadata to an
image, providing a method for clearly associating the alternative content in
a machine-comprehensible way. A demonstration tool has been developed to do
this to jpegs (reading out and adding in RDF data, such as Dublin Core
information), and the Jigsaw web server now has the abiliity to extract the
metadata that is included in a jpeg ad serve it seperately.

5.3 Alternative pages
The text makes it clear that what is referred to should be an accessible
page, but in the example it is called a text-only page. These do not mean the
same thing, and the example should be corrected.

5.6.1 Comprehension
Jakob Nielsen has a short, easily understood essay on writing for the web
which is a good reference. http://www.useit.com/alertbox/9606.html which is
linked from http://www.useit.com/alertbox/9710a.html - itself a good
reference on how people read the web.

5.6.2 Multimedia
Nte also that people who use American Sign Language, British Sign Language,
and Auslan, are likely tohave more reading ability in English than another
language, but their primary sign languages are different. Like English and
German are different.

5.8 autorefresh
I am not sure that having this as the major part of the metadata section in
the HTML techniques is a good idea. It isn't even a metadata technique
(although it happens to use that syntax).

5.11.3 User Scenarios
This should have a link to the document "How people with disabilities use the
web". Even if that is only a draft.

definitions: braille
Some braille uses 8 dots. And there isn't the word accessible presented in
braille, there is an image of it.

natural language:
Braille is not a natural language. It is a format for representing one, as
the gothic script alphabet is.

references: Bobby
Bobby is not an automated tool. It is a semi-automated tool

triple-A WCAG compliant: 11.1 (XHTML)? 13.1 (link targets)? 13.2 (Metadata)?

Charles McCN

--
Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative                      http://www.w3.org/WAI
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053
Postal: GPO Box 2476V, Melbourne 3001,  Australia

Received on Tuesday, 15 February 2000 23:54:09 UTC