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

Fwd: Re: Call for Review: Website Accessibility Conformance Evaluation Methodology (WCAG-EM)

From: Shadi Abou-Zahra <shadi@w3.org>
Date: Fri, 28 Sep 2012 08:32:00 +0200
Message-ID: <50654460.1080100@w3.org>
To: Eval TF <public-wai-evaltf@w3.org>
Forwarding some more comments that don't seem to have gotten through.


-------- Original Message --------
Subject: Re: Call for Review: Website Accessibility Conformance 
Evaluation   Methodology  (WCAG-EM)
Date: Thu, 27 Sep 2012 21:48:00 +0000
From: Ramón Corominas <rcorominas@technosite.es>
Organization: Technosite
CC: public-wai-evaltf@w3.org <public-wai-evaltf@w3.org>

Dear Eval TF,

I'm finally having some time to review and comment on this document.
Congratulations for all the good work, it is already in Working Draft
phase! I hope I'm not too late to contribute :)

I'm in the process of reading the full methodology as a "fresh review".
This means that, although I've read many of the comments here, it is the
first time I read the complete document, so please apologies if I say
something that you have already discussed.

In this first e-mail I will just comment about the introduction, but
since some comments are "philosophycal issues" they will probably
affects to other sections. I expect to send more comments in the
following days.

---
Section 1. Introduction

1.1 Scope of this Document

"Representative pages" in single-page apps

Although "website" usually refers to more than one page, in single-page
applications the whole website is just ONE page. Thus, the sampling of
"representative pages" would immediately lead to considering that page,
which seems to mean "consider and test ALL the functionalities of that
page, that is, test "everything in the app". I would suggest to avoid
this saying "representative pages and/or tasks" instead of
"representative pages" (I use the word "task" as a tentative term, to
avoid "process", since WCAG 2.0 uses "process" in the context of
multiple pages).

For example, 1st paragraph of Section 1.1 says: "sampling representative
web pages where it is not feasible to evaluate all web pages of a
website". Since there is only one page in a single-app, we must evaluate
"all web pages", so the wording seems to mean that sampling is not
required in this case.

Practical example: how to use WCAG-Em to evaluate Google Docs?


"Applying Success Criteria"

In the same paragraph, it says: "applying the WCAG 2.0 success
criteria...". Shouldn't it be "verifying the application of..." instead
of "applying"? We usually understand that "applying" the SC means
"making the necessary changes" to the website, while checking their
application does not imply changes. Maybe this is just a matter of my
dual evaluator/developer profile ;)


Applicability and types of "websites"

Then it says that "it is applicable to all websites, including web
applications, websites intented to be used with mobile devices...".

In principle, it sounds ok. However, I'm wondering how WCAG-EM could be
applied to websites that are embedded into native mobile apps (or other
similar contexts). For example, if a native app uses one main -native-
interface and the "website" is rendered in an embedded window within the
app. In this situation, there might be some components of the whole
experience that are in the "native" part, and therefore cannot be
evaluated according to this "web" methodology; however, since they
contain essential parts of the navigation or structure, the embedded
website would fail conformance (out of context). Maybe this situation of
"websites that are part of a higher level product" can be included in
"Section 4. Considerations for Particular Situations". This would also
allow WCAG-EM to be applicable to partial content or mock-ups.


"Only after development"

I assume there is a strong reason for this (probably due to difficulties
to apply it to partial content), but it sounds really discouraging if we
pretend to introduce an evaluation methodology as an essential part of
the QA process. Moreover, it resembles to the old philosophy of
considering accessibility only at the end of the road. Developers need a
consistent way of evaluating the accessibility of their developments in
different phases before the website is shipped, and I think this
methodology should cover this to gain wider acceptance. I know that
other resources are provided, but this paragraph really limits
applicability for many real-world scenarios where clients want a single
procedure.

In addition, in Section "1.2 Target Audience", it is mentioned that the
methodology is relevant to "Website developers who want an evaluation
during the development process". How could they use WCAG-EM if it
explicitly says that it's not applicable in this situations?


Section 1.4 Terms and Definitions

I don't understand very well what "Common functionality" tries to
define. Could it be something like "essential functionalities", in a
similar way to the definition of "essential" in WCAG 2.0? I mean,
"essential" is defined as something that, if removed, produces a
fundamental change in information or functionality. Is this the case?

It sounds a bit strange to me that for "evaluator", "evaluation
commisioner", and "web owner" you use "the person, team of people,
organisation..." but for "web developer" you use "anyone involved...".
Maybe this is because "Web developer" includes different profiles, but
it seems to reflect a difference in the "seriousness" of those guys
called "web developers".


Regards,
Ramón.

> WCAG-EM Working Draft 20 Sept 2012
Received on Friday, 28 September 2012 06:32:44 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:52:15 GMT