Priorities - a proposal

We have what we claim is a simple rule for distinguishing between P1 and P2
items - whether they make things impossible. The difference between P2 and P3
seems much less well defined.

I don't think there will ever be a hard and fast rule, but a bit better
guidance would be good. In addition to the definitions that we use at the
moment, (essential, important, helpful, or some variation) I suggest we look
at efficiency as a rough guide. For example, something that takes 10 times as
long or more, and is a task that "normally" takes several minutes suddenyl
becomes a task that takes an hour or more. In this case I would suggest that
it represents a barrier of P1 level - in particular in a work situation this
makes it practically impossible.

For the difference between P2 and P3 I think that things which do not cause
this level of efficiency blow-out, but mean that a task takes twice as long
or more, a P2 is justified.

A P3 should be solving a problem that is less significant than that.

I realise that these are rough figures, and working out how to apply them
will still involve a measure of subjectivity, but maybe we can lessen it
somewhat. I think that would help us.

A second part of the proposal is to look at different tyypes of requirements.
It seems reasonable to assume that a person with disabilities makes use of
their software through the standard interfaces, so requiring special skills
such as interpreting HTML source should not be considered as an access method
in determining the impact of something. On the other hand, there are
peopole who can make use of such functions, which are after all common.
Although this represents perhaps 10% of the user community, makiing such
functions available is an important repair strategy. SInce it seems that in
the near future no single strategy is going to solve everyone's problems, we
should be prepared to include these things (view source is one example) as
requirements, with their priority based on the difference they can make. As
an example, if a user can look at the source of a document, and in particular
of scripts, they can determine how to get access to a site that is otherwise
completely inaccessible. (As sites I have used this technique for, there is
the Boston "T" system - http://www.mbta.com, the first version of the Sydney
Olympics Website, and others that I forget now). In other words, with this
technique, access that was otherwise impossible becomes possible. So the
requirement is at P1 level - removing an (effectively) total barrier.

What do people think?

Charles McCN

-- 
Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
W3C Web Accessibility Initiative     http://www.w3.org/WAI    fax: +1 617 258 5999
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia
(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)

Received on Sunday, 11 March 2001 11:33:45 UTC