W3C home > Mailing lists > Public > public-wai-evaltf@w3.org > April 2012

AW: Templates and use cases

From: Kerstin Probiesch <k.probiesch@googlemail.com>
Date: Fri, 20 Apr 2012 08:02:01 +0200
To: "'Elle'" <nethermind@gmail.com>, "'RichardWarren'" <richard.warren@userite.com>
Cc: "'Alistair Garrison'" <alistair.j.garrison@gmail.com>, "'Eval TF'" <public-wai-evaltf@w3.org>
Message-ID: <4f90fbd5.c507b40a.71ca.73ea@mx.google.com>
Hi all,

I believe it will be helpful to use different terms for three different
situations (and use cases):

a. testing templates
b. testing template-based pages
c. testing template-based elements (forms, tables, ...)

It should be possible that an evaluator could evaluate templates (scope
definition) and offer in his report guidance for implementing (IDs, heading
levels, skip links - problems described by Richard and others). 



Von: Elle [mailto:nethermind@gmail.com] 
Gesendet: Freitag, 20. April 2012 00:23
An: RichardWarren
Cc: Alistair Garrison; Eval TF
Betreff: Re: Templates and use cases

Richard and Alistair:

I think you both (and I) agree on most of the important points of this
• Testing templates is efficient and it saves time
• It doesn't replace full page testing for conformance
• Whether we can get the templates directly from the website owner or have
to sleuth out the commonalities is probably a case-by-case situation

My goal in participating in this group (besides learning from all of you) is
to ensure that any evaluation methodology that the W3C creates is pragmatic
enough to be used in a real world scenario at a large organization. With
that said, content management system templates, business and IT use cases,
URL lists, and other work that has been done before I arrive to test are a
requirement if I am to be successful in measuring conformance on all the
sites our company owns.


On Thu, Apr 19, 2012 at 6:10 PM, RichardWarren <richard.warren@userite.com>
Hi Alistair,

We do template testing as a separate service to be used during the
development of a site. However, even fully compliant templates do not always
produce compliant web-pages. Examples we have found are duplicated IDs.
overlapping areas when zoomed and "skip to content" links that don't work
properly. I do agree that testing the actual templates is a good way of
saving time and effort (that's why we do it) but it is not necessary for the
methodology we are working on.

What we are interested in is the the completed pages. We are only
recommending identifying templated or common areas as a method of reducing
effort on the part of the evaluator and for ease of reporting findings.  By
identifying and reporting on these areas separately we can provide the
developer with an efficient list of "quick fix" improvements within our
overall report.


-----Original Message----- From: Alistair Garrison
Sent: Thursday, April 19, 2012 9:11 PM
To: RichardWarren ; Eval TF

Subject: Re: Templates and use cases

Actually Richard, it is 100% the actual templates I would be interested in -
for exactly the reasons I indicated.

Having worked extensively with content management systems I don't know of
any which leave helpful ids relating to templates - which is another reason
why it would be good to have access to the actual templates.

And, I would also point out that it might not be so easy to work out the
templates used on a 100 page plus site, let alone a 100,000 page plus
website... In any case, why would you want to waste time trying when you can
ask the evaluation commissioner for them directly...

All the best


p.s. Dreamweaver is an HTML authoring tool - not a Content Management System

On 19 Apr 2012, at 20:38, RichardWarren wrote:
Hi All,
Sorry I missed the teleconference. I feel that the problem problem probably
starts with the wording of section 2

Requirement 2a “ During this step the common web pages of the website and
templates available to the evaluator are identified and documented. "

It is not the actual templates that we are interested in but how they are
applied. I think they would be better described as Elle calls them  -
"templated areas". Some CMS such as Dreamweaver helpfully leave comments in
the code naming the template and the area. Where this is not available it is
often easy to work out from the <div> IDs. Using this makes it helpful for
the developer as we can say -'in template XYZ there are the following
errors..' and then concentrate on the main content area of relevant pages.


From: Alistair Garrison
Sent: Thursday, April 19, 2012 6:18 PM
To: Elle ; Detlev Fischer ; Eval TF
Subject: Re: Templates and use cases

Hi Detlev,

To my mind there is true value in having evaluators having access to the

Lets say that your first move when evaluating is to quickly evaluate and
repair the templates - correcting alt missing type issues, etc... and then
having them put back into the live site.

By doing this simple move, when you come to run your automated tools over
the site - instead of picking up ever page due to the recurring errors in
templates you only pick up those pages which have other errors...

Hopefully making the whole process smarter and more efficient...

Otherwise, I'd mirror what Elle said.  I think we need to work 'smart' - by
utilising information that already exists and by drawing on the knowledge of
the evaluation commissioners.

It is important to recognise that we do not need to turn over every stone
during an evaluation... Really, just enough to conclusively show if success
criteria have been failed.

I'd be super happy if the evaluation commissioner provided a full site map
(in the sitemap.xml protocol), use cases for all major functions and empty
templates for all page types - just by doing this they would save me so much
time and effort...

Hope this helps


On 19 Apr 2012, at 17:37, Elle wrote:


I think, as you mentioned later in your email with use cases, that templates
provide a good starting point. They are only part of the picture for an
evaluator, but they represent an efficient way to start the evaluation
process. Using templates and reporting on them as such allows an evaluator
to identify common elements needing correction in a way that communicates
well to the website owner.  It's how he or she already views the site in
question. As you say, it's not at all a replacement for evaluating live
content and its interaction within the template, but it reduces the
redundancy of page-level testing.

When we remediate our websites, we look first for common elements, often
templated elements. Then, we look on a page level basis. If, on the page
level analysis, we see that an interaction needs to change because of the
conflict with template and page components, we can decide whether to change
the template or the page content itself.  That's a decision that's harder to
make if we don't view these as separate items, I believe.


On Thu, Apr 19, 2012 at 11:29 AM, Detlev Fischer
<detlev.fischer@testkreis.de> wrote:
Hi list,

just two points that cam to mind after today's teleconference:

(1) Templates

The text now (in Step 2a)  talks about "The common web pages of the website
and templates available to the evaluator"

For templates that just cover a part of the web page (e.g. portlets), some
aspects of conformance depend on their place in the context of the overall
page - think of 1.3.1 and 2.4.1 (heading levels), 1.3.2 Meaningful Sequence,
etc. I wonder what the evaluator will actually *do* with the templates
provided by the commissioner - if they are empty or out of page context, a
complete evaluation seems impossible or of doubtful value. But this may not
be intended by the description - not sure. Many observations (potential
fails) also result from interactions between template and content on a real

I am not against taking stock of templates per se, just not sure what added
value that would bring and how they would be assessed outside a normal page
context. Thoughts?

(2) Use cases

In cases where evaluations of web applications are commissioned by clients,
checking the use cases provided is certainly a good starting point. It's
just important to also go beyond documented use cases, especially regarding
error handling and unexpected user actions, to capture violations.

Detlev Fischer
testkreis - das Accessibility-Team von feld.wald.wiese
c/o feld.wald.wiese
Borselstraße 3-7 (im Hof)
22765 Hamburg

Tel   +49 (0)40 439 10 68-3
Mobil +49 (0)1577 170 73 84
Fax   +49 (0)40 439 10 68-5

Beratung, Tests und Schulungen für barrierefreie Websites


If you want to build a ship, don't drum up the people to gather wood, divide
the work, and give orders. Instead, teach them to yearn for the vast and
endless sea.
- Antoine De Saint-Exupéry, The Little Prince

If you want to build a ship, don't drum up the people to gather wood, divide
the work, and give orders. Instead, teach them to yearn for the vast and
endless sea.    
- Antoine De Saint-Exupéry, The Little Prince
Received on Friday, 20 April 2012 06:02:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:40:20 UTC