Re: Tools, was Re: Manual validation

On Sun, 22 Sep 2002, Charles McCathieNevile wrote:

> For most people, learning accessibility testing is very difficult. Just using
> the list of checkpoints is almost impossible without already knowing what is
> in the techniques documents, and knowing some stuff that could be added to
> them.

I'd say also very tedious to do with any degree of thoroughness:
a recipe for oversights!


> The tools that have been produced for testing and repairing accessibility
> problems in Web pages have been enormously helpful by making it easier for
> people to do what they need (make acccessible pages). At times people have
> claimed that a tool can automatically test everything and there is no need
> for a person. I don't think that is true - I know that some tools recognise
> their limitations and explicitly ask the user to check some features, and
> others just test a certain group of features.

Any tool has to compromise somewhat between doing a thorough job and
overloading the user with mostly-irrelevant information.  Of course,
the only tool that can guarantee complete support for WCAG is one whose
report on any page is the WCAG itself.  But a useful tool is one that
adds to that, by drawing a users attention to specific things that are
(or might be) in violation of the WCAG.

In the case of an organisation managing a web team, another role for
tools is in automatic monitoring of updates in support of an
overall QA strategy.  When someone makes a "trivial" update and
commits some blunder through inattention, better an email reminder
from Valet the next morning than that it goes undetected!

> People have from time to time misunderstood or misused the information they
> get from tools - this is something we need to keep thinking about.

Yes.  The spellcheck is a useful analogy: the limitations on what it
can do are broadly comparable, and are familiar to anyone likely to
be using web-accessibility tools.

-- 
Nick Kew

Received on Sunday, 22 September 2002 19:42:49 UTC