Transcript MS PowerPoint format

Forcing Standardization or
Accommodating Diversity?
A Framework for Applying the
WCAG in the Real World
Brian Kelly
UKOLN
University of Bath
Bath, UK
Email
[email protected]
Co-authors:
David Sloan, University of
Dundee
Lawrie Phipps, TechDis
Helen Petrie & Fraser
Hamilton, City University
UKOLN is supported by:
A centre of expertise in digital information management
www.ukoln.ac.uk
Background To The Paper
This paper:
• Based on experiences of promoting & supporting
best practices for Web accessibility within the UK
higher & further education (HE/FE) & cultural
heritage communities
• Five authors from four HE organisations in the UK
(including 2 national advisory services)
Web developer
Consultant and
Senior adviser
since 1993 &
researcher in
at TechDis
adviser to UK
inclusive design
– a national
HE/FE & cultural
and accessibility,
advisory
heritage
University of
service to
in digital information management
communitiesA centre of expertise
Dundee
HE/FE
2
Senior
Professor at
City University researcher
at City
with interests
University
in usability &
accessibility www.ukoln.ac.uk
The WAI Model
WAI has been tremendously successful in raising
awareness of Web accessibility and providing
guidelines to achieve this.
WAI guidelines are
based on:
• WCAG (Web Content …)
• ATAG (Authoring Tools ..)
• UAAG (User Agents …)
The model is simple to grasp. But is this model
appropriate for the future? Does the model:
• Reflect the diversity of users & user environments
• Reflect the diversity of Web usage (not just informational sites)
• Reflect real-world technical environment and developments
Reflect
real-world
political management
and cultural environments
A•centre
of expertise
in digital information
www.ukoln.ac.uk
3
Limitations Of The Model
This model:
• Requires all three components to be implemented
in order for the WAI vision to be achieved
• Is of limited use to end users who have no control
over browser or authoring tools developments
• Is confusing – as many think WCAG is WAI
How does this model address:
• Delays in full conformance? (We're still waiting for
"until user agents …" clause to be resolved)
• Real-world reluctance to deploy new software
(issues of inertia, testing, costs, …)
• Real world complexities
4
Is there a plan B in case this model fails to ever take off?
Is it desirable to base legal requirements on an unproven
A
centre of expertise
in digital information management
www.ukoln.ac.uk
theoretical
framework?
WCAG Conformance
Page authors can only follow WCAG guidelines.
Several surveys carried out using automated tools
(which gives upper limit on accessibility)
• DRC report: 19% A, 0.6% AA conformance based
on 1,000 Web sites
• UK Museums report: 42% A, 3% AA conformance
based on 124 Web sites
• UK Universities surveys (2002, 04):
43%/58% A, 2%/6% AA based on 160+ Web sites
(NB figures give numbers which comply with the
metrics the tool measure)
Implications
DRC
Disability
Rights Commission,
independent body
These– low
conformance
levels can indicate:
legislated
to stop discrimination
and promote equality of
• Organisations
don't care
opportunity
of disabled
people.
• Guidelines
are difficult
to implement
• Guidelines are inappropriate, misleading, wrong, …
• Technical
OSs, …) overcomewww.ukoln.ac.uk
barriers
A centre
of expertisedevelopments
in digital information (ATs,
management
5
WCAG Difficulties
Certain Priority 2 and 3 guidelines cause concerns:
11.1 Use W3C technologies when they are available
and appropriate for a task ...
• Promotes own technologies
• Appears to ignore major improvements in
accessibility of non-W3C formats
11.1 … and use the latest versions when supported
• Goes against project management guidelines
• Logical absurdity: when XHTML 1 came out WAI
AA HTML 4 compliant sites downgraded to A!
3.2 Create documents that validate to published formal
grammars
• Dodgy HTML (<br />) can be rendered by
browsers – this is an interoperability issue
A centre of expertise in digital information management
6
www.ukoln.ac.uk
Usability Issues (1)
"WCAG provides the highway code for accessibility on
the information superhighway"
"Fine – but what if the accelerator and brake pedals
differ on every car. I'll still crash!“ (and I want to fly)
DRC survey also carried out usability testing:
• Exemplar accessible Web sites did not comply with
WCAG guidelines (WCAG A)
• WCAG compliant sites (according to tools) were not
accessible or usable
DDA requires users to be able to access & use services
The subjectivity of
"I don't claim people should
DDAguidelines
– UK's Disability Discrimination
ActI say"
usability
do 100% of what
seems to be recognised
Jakob
Neilson
A centre of expertise in digital information management
www.ukoln.ac.uk
7
Usability Issues (2)
What is the relationship between usability &
accessibility?
Accessibility
Usability
Usability
Accessibility
Usability
Accessibility
Accessibility
Usability
A centre of expertise in digital information management
8
www.ukoln.ac.uk
Confusion
SiteMorse’s automated accessibility survey of UK
disability organisations’ Web sites generated heated
debate
• SiteMorse: Low WCAG conformance found:
• Response: doesn’t matter, manual testing gives
OK results
What do such comments say about disability
organisations’ views of WCAG ?
Note that the RNIB actively promote WCAG
guidelines – and also promote use of accessible
Flash, without flagging any inconsistencies.
9
Organisations may publicly support WCAG whilst rejecting
(parts of) it (for understandable political reasons). But we
A centre of expertise in digital information management
www.ukoln.ac.uk
want to develop an alternative approach.
Nitpicking?
“This is just nit-picking! WCAG is valuable – don’t
knock it!”
WCAG is valuable, but we need to:
• Build a robust framework for the future
• Ensure clarity and avoid ambiguities to avoid
different interpretations
• Reflect on experiences gained since 1999
• Avoid dangers of inappropriate case law being set
Nightmare Scenario
Case taken to court in UK.
Defence lawyers point out ambiguities & inconsistencies.
Case lost, resulting in WCAG’s relevance being
diminished.
A centre of expertise in digital information management
www.ukoln.ac.uk
10
Holistic Approach
Kelly, Phipps & Swift1 have argued
for a holistic framework for
e-learning accessibility
This framework:
• Focusses on the needs
of the learner
• Requires accessible
learning outcomes,
not necessarily e-learning
resources
This approach reflects current UK emphasis on
blended learning (rather than e-learning)
1 Developing A Holistic Approach For E-Learning Accessibility,
A centre
of expertise
in digital
informationand
management
Canadian
Journal
of Learning
Technology, 2004, Vol. www.ukoln.ac.uk
30, Issue 3
11
Legal Issues
UK legislation requires organisations to take
"reasonable steps" to ensure disabled people do not
face unjustified discrimination.
This approach:
 Is technology-neutral
 Is both forwards-looking and backwardscompatible
 Acknowledges differences across providers of
services
 Doesn’t differentiate between real-world and
online accessibility (or between Web and other
IT accessibility)
 Avoids change-control difficulties
The approach outlined in this paper appears to fit in well
with
UK legislation
A centre of expertise in digital information management
www.ukoln.ac.uk
12
Need For Engineers
The Web and the notion of universal accessibility was
initially build on a great vision
But as:
• The Web gets more complex
• Real world complexities become apparent
• Alternative approaches emerge
• We learn from user experiences
we argue the need for an engineering (evidencebased) approach rather than a visionary/dogmatic one:
• Robust solutions
• Fail-safe
• Reflect user experiences rather than ideologies
A centre of expertise in digital information management
13
www.ukoln.ac.uk
WAI WCAG 2.0
What about WCAG 2.0?
 Modularity
 Uncertainty of the broader context
 It is tolerant of non-W3C formats
 Long gestation period
 Why address writing style? (this is for real world)
 Complexity and relationship with WCAG 1.0
 Horrendous migrations implications
"Though
this address
be madness,
there
is method in it"
 Doesn’t
limitationsyet
of WAI
model
Addressing
writing style has attractions, but:
 Still closed:
• Applies
a single
dimension (ease of understanding)
 Lack
of extensibility
 Failsricher
to address
of other
developments
e.g.
to a much
areaimplications
(learning,
culture,
…)
models of accessibility; technical developments such as
• Liable topersonalisation
undermine core accessibility issues
• Hostage
 …to fortune – "They're banning Shakespeare"
• Makes divisions between real world and Web
A centre of expertise in digital information management
www.ukoln.ac.uk
Similar
concerns have been raised by Joe Clark, AListApart
14
Accessibility In Context
External factors: Institutional issues (funds, expertise,
policies, security…)
Digital Library Programme
Context
Purpose Sector Funding Resources Research
…
Policies
Standards Accessibility/Usability Privacy Finance …
Compliance
External Self-assessment Penalties Learning Broken
A framework is
being developed
which places
accessibility &
usability within a
wider context:
• The context
• A range of
policies
• A compliance
regime
Accessibility
guidelines should
This approach embraces relativism and context be usable in wider
rather than
the
current
absolute
approach
context
A centre
of expertise
in digital
information management
www.ukoln.ac.uk
External factors: Legal issues; cultural factors; …
15
Conclusions
16
To conclude:
• Importance of Web accessibility widely accepted
But:
• WCAG P2 & P3 guidelines aren’t working (cf.
disability organisations, W3C members, W4A, …)
• Some guidelines are flawed
• Usability issues are being lost
• User focus on WCAG conformance rather than
providing usable & accessible resources
• World has moved on since 1999
Need for:
• Modularity
A centre
expertise in digital information
management
www.ukoln.ac.uk
• ofEngineering,
not ideology
Questions
Questions are welcome
A centre of expertise in digital information management
17
www.ukoln.ac.uk