W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2005

Re: [techs] test case review assignments

From: Ben Caldwell <caldwell@trace.wisc.edu>
Date: Tue, 26 Jul 2005 14:19:41 -0500
Message-ID: <42E68CCD.10405@trace.wisc.edu>
To: "WAI GL (E-mail)" <w3c-wai-gl@w3.org>

Here's the results of my action item for test case reviews. Specific 
comments about the tests I was assigned below with some more general 
comments and proposals related to the test suites at the end.

-Ben

> * Document must contain a TITLE element. 
> <http://www.w3.org/WAI/GL/WCAG20/tests/test50.html>

This test has already been listed as accepted, only a few minor
adjustments are needed:

1. Propose changing the test title from, "Content contains a title
element." to "Delivery units contain a title element." This is more in
sync with the success criterion and removes ambiguity about what is
meant by "content." (suggest replacing "content" with "delivery unit"
throughout -- applies to technique 1.2 as well)

2.) The related SC reference needs numbering fix (was SC4 is SC3 in
latest draft)

3.) the example in test file 50-2 has a somewhat cryptic title - suggest
revising this from, "<title>OAC Testfile - Check #50 - Negative</title>"
to "<title>Test 50 - HTML Test Suite For WCAG 2.0</title>." (note: this
test file also appears in tests 51, 52 and 53)

4.) we may want to consider associating this test (once resolutions on 
guideline 4.1 are made) with validity as the presence of the title 
element would be required at a higher level. (current mapping is to a 
level 3 criterion)

> * TITLE must contain at least 1 character. 
> <http://www.w3.org/WAI/GL/WCAG20/tests/test51.html>

Another one that has already been accepted. A couple proposed edits:

1.) The procedure section includes a note that says, "For this test the 
term "contains text" excludes whitespace. The title element must contain 
at least one non-whitespace character." I may have missed it, but didn't 
see anyplace where we were using the phrase "contains text" to include 
whitespace, so I propose removing this note and replacing it with 
something like, "the title element contains at least one printable 
character."

2.) Expected results includes the phrase,  "title contains text." 
Propose that we revise it to read, "The title element contains text." in 
order to avoid confusion with the title attribute (note: this editorial 
suggestion applies to multiple tests in the current test suite)

3.) With the exception of test 50 (content contains a title element), 
each of these tests should be moved to general techniques since the 
concept of title is not unique to HTML.

> * TITLE should be less than 150 characters. 
> <http://www.w3.org/WAI/GL/WCAG20/tests/test52.html>

Propose that we delete this test since it seems to fall more on the side 
of usability than accessibility. If we were to keep it, we'd need a way 
to identify it as good advice or optional in some way.

1.) test 51 seems to be a prerequisite test for this.

2.) test file 52-1 has a typo in it. "Yes, let's all shorted the..." 
should be, "Yes, let's shorten the..."

> * TITLE should not contain placeholder text. 
> <http://www.w3.org/WAI/GL/WCAG20/tests/test53.html>

Another one that is already listed as accepted.

1.) Suggest adding additional language specific values and test files 
for "Enter the title of your HTML document here," (XMLSpy default) 
"Untitled Document," (DreamWeaver Default) and, "No Title" (Amaya 
default), and "Untitled Page" (GoLive default). I'm sure there are others...

> * TITLE must describe the document. 
> <http://www.w3.org/WAI/GL/WCAG20/tests/test54.html>

This test is very similar to the SC itself. Do we want to include tests 
like this or could this information be covered by the SC and supporting 
materials in the guide doc?

-------------------

Items that relate to the entire test suite:

1.) links to "priority levels" in the intro should point to
http://www.w3.org/TR/WCAG20/#conformance (this applies to all files in
the test suite)

2.) Pass instructions for many of the tests include the phrase "There is
no requirement that these tests be performed." The text appears 
preceding references to related tests and seems to imply that these 
related tests are optional, but I don't think this is what was intended. 
For now, I suggest removing this phrase.

3.) As written, the priority level of tests in the test suite currently 
inherit the priority level of the success criterion to which they are 
(indirectly) related. However, tests are designed to test whether a 
technique has been implemented correctly. Since techniques may or may 
not be applicable to a given set of Web content, I'm not sure this 
indirect priority inheritance is a good idea and am concerned that it 
will cause confusion. One of the main issues it creates is that it 
implies that performing all of the "level 1" tests would ensure 
conformance to WCAG 2.0 level 1.

I propose removing priority level information from tests altogether and 
creating a mechanism that allows us to identify tests as optional 
instead. That way, the model for determining conformance would be for 
authors to first determine which techniques they are going to make use 
of to meet the guidelines (note: the details to support this are not yet 
included in current techniques drafts) and, second, determine which 
tests must be performed in order to determine whether those technique 
have been correctly applied.
Received on Tuesday, 26 July 2005 19:21:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:39 GMT