WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Procedure of making web accessibility testing.

for

From: Wayne Dick
Date: Jun 9, 2009 12:10PM

Another approach is to prioritize into
three groups: critical pages no more
than 100 pages; important pages 300,
less used and important pages:300; and
rarely used or unimportant sites 300.
Don't split hairs. Get help from
others to develop your picture.

Make sure the critical pages are fixed
right away. Walk through and test 20
to 40 of them and evaluate using
manual techniques. That should give a
good picture of the whole site - all
thousand pages. Next make a
prioritized maintenance schedule and
follow it. Each time a page comes up
for maintenance fix it. If you set
your schedule for 3 years then fix the
critical pages and 300 / year.
Finally, never add a new bad page to
your site.

Don't be frightened by manual
evaluation. I checked 400 pages by
hand in two weeks.

This method is called the
refurbishment model taken from
construction. It worked well in the
80's-90's for architectural barriers.
For the web it is better because web
pages don't last for centuries.
Legal concerns: Institutions that are
following a realistic plan will not
big problems with regulatory agencies.

The nice thing about this method is
that after 3 years many unimportant
pages have died already.

Wayne Dick









On Tue, 09 Jun 2009 10:50:33 +0200
Christophe Strobbe
< <EMAIL REMOVED> >
wrote:
>
> At 06:18 9/06/2009, Rakesh Chowdary
>Paladugula wrote:
>>Dear all,
>> I have a doubt from the beginning.
>>I have a existing website with
>>1000 pages to be tested for web
>>accessibility should I test all the
>>1000 pages or some selected pages .
>>If so what are the pages I gave to
>>test.
>
> If you are looking for a sampling
>method, you could have a look at the
> one in the "Unified Web Evaluation
>Methodology":
> <http://www.wabcluster.org/uwem1_2/>;
> See chapter 4, "Scope and sampling
>of resources" in the "Core" document.
>
> Since you seem to know where all the
>pages of your website are, it
> should be possible to create a fully
>random sample, as described in
> section 4.3.2. (Section 4.3.1
>describes a sampling method that can
> be applied "manually".)
> I'd be very interested to hear how
>this works for you.
>
> Best regards,
>
> Christophe Strobbe
>
>>(...)
>> Thanks & regards
>>Rakesh Chowdary
>>Iridiuminteractive Limited
>
>
> --
> Christophe Strobbe
> K.U.Leuven - Dept. of Electrical
>Engineering - SCD
> Research Group on Document
>Architectures
> Kasteelpark Arenberg 10 bus 2442
> B-3001 Leuven-Heverlee
> BELGIUM
> tel: +32 16 32 85 51
> http://www.docarch.be/
> ---
> "Better products and services
>through end-user empowerment"
> http://www.usem-net.eu/
> ---
> Please don't invite me to LinkedIn,
>Facebook, Quechup or other
> "social networks". You may have
>agreed to their "privacy policy", but
> I haven't.
>
>