De-mystifying and Taming the Complexities of WCAG 2.1
Download as PPTX, PDF1 like167 views
The document discusses the challenges and strategies of implementing WCAG 2.1 accessibility standards at Optum, emphasizing the importance of tailored checklists to meet the specific needs of products and testing processes. It details the evolution of their accessibility program, the complexities of WCAG guidelines, and the benefits of custom groupings to enhance accessibility testing. The speaker, Bill Tyler, highlights his extensive experience and the growth of accessibility efforts within UnitedHealth Group since 2013.
![Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 14
Bespoke
Adjective [BRITISH]
made for a particular customer or user.
"a bespoke suit"](https://image.slidesharecdn.com/a11ytcmeetupdemystifyingwcag2020-06-22-200618021734/85/De-mystifying-and-Taming-the-Complexities-of-WCAG-2-1-14-320.jpg)
![Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 40
Other information for reports
β’ Additional options can be added for checkpoints and reports to communicate benefits
Example: βAffects users who areβ¦β [blind, low vision, deaf, cognitively impaired, etc.]
β’ Accessibility tools for testing
List of tools to test checkpoint β useful for developers and designers
References links to standards and resources
β’ Roles
List of roles that may own the issue or should be consulted about the checkpoint
β Example: color contrast issues (SC1.4.2) likely belong to visual designer](https://image.slidesharecdn.com/a11ytcmeetupdemystifyingwcag2020-06-22-200618021734/85/De-mystifying-and-Taming-the-Complexities-of-WCAG-2-1-40-320.jpg)
De-mystifying and Taming the Complexities of WCAG 2.1
- 1. De-mystifying & Taming the Complexities of WCAG 2.1 Bill Tyler June 22, 2020 Accessibility Twin Cities MeetUp #bespokewcagchecklists SlideShare: tbd
- 2. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 2 Meet the speaker My Super Power β’ Experienced βMature Methaneβ β UX developer & designer experience back to 1984 Education & Experience β’ BA in Computer Studies β’ Full-time accessibility professional since 2013 β First exposure & experiences go back to 2002 β’ W3C Invited Expert on Role-based Accessibility When did my accessibility journey truly started β’ Learned a lot about the challenges of the blind through a friend born without eyes over 20 years ago (Don Wardlow) Bill Tyler Principal Digital Accessibility Engineer btyler@optum.com @billtyler
- 3. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 3 Approach & techniques used at UnitedHealth Group for WCAG testing β’ Optum Technology Accessibility Center of Excellence (A11y CoE) β’ 6+ yrs. of ongoing accessibility testing research & analysis β’ Applied to 200+ UnitedHealth Group (UHG) products
- 4. Accessibility Center of Excellence A brief history 4
- 5. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 5 0 15 30 45 Staff size before, during & after take-off 11 31
- 6. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 6 Growth and change over past 6+ years β’ Primary practice in UnitedHealth Group (UHG), Optum and UnitedHealthcare (UHC) β’ Started December 2013 with 2 people (including me!) β’ Through internal mergers and restructuring grew to 10 people (including managers) β’ Summer 2016 (Take-off): Complete assessments on 200+ products β Increased 300%: 11 to 31 over 6 months β’ Starting 2018 transitioned to consulting, governance, separate UHC digital accessibility program (DAP) β’ Today: Maturing of Optum A11y CoE (40) and UHC DAP (5) But still growing!
- 7. WCAG 2.1 A brief summary 7
- 8. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 8 WCAG 2.1 summary WCAG: Web Content Accessibility Guidelines from the W3C 4 (POUR) Principles β’ Perceivable β 4 guidelines β’ Operable β 5 guidelines β’ Understandable β 3 guidelines β’ Robust β 1 guideline 78 success criteria across 3 levels β’ A β 30 criteria β’ AA β 20 criteria β’ AAA β 28 criteria
- 9. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 9 WCAG 2.1 conformance Target: WCAG AA β’ WCAG 2.0 (2008) β 38 (A & AA) criteria β Current US Section 508 (and many others, including most existing legal cases) β’ WCAG 2.1 (2018) β 50 (A & AA) criteria β Started in EU (EN 301 549) some organizations such as University of Minnesota β UHG / ACoE Standard as of January 1, 2019
- 10. The trouble with WCAG 0
- 11. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 11 WCAG is great for learning and organizingβ¦ β¦but βone size does not fit all.β
- 12. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 12 WCAG off the rack is not really useful for testing Principles β single-word abstractions β’ Great for organizing, but not for testing Guidelines β goals β’ Elaborate on principles, but do not add much detail Success Criteria β testable requirements! β’ Targets to be met β’ Basis for conformance β’ But the text is still not checklist or checkpoints β Understanding documents includes situations, techniques, failures, and more
- 13. What did we do? Bespoke checklists Custom tailor WCAG to our needs
- 14. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 14 Bespoke Adjective [BRITISH] made for a particular customer or user. "a bespoke suit"
- 15. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 15 Why bespoke (custom tailored) checklists? β Before take-off No budget, but lots of time β’ First couple years we had the luxury of time to prepare β We knew what was coming β’ With no budget for off-the-shelf tools we had to build our own checklist β’ Excel was the tool of choice β Standard issue β Basic database operations β Sorting β Filtering β Import/Export
- 16. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 16 Why bespoke (or custom) checklists? β After take-off Little budget, no time, and nothing fit properly β’ Had to build and adapt on the fly using the tools at hand β’ Rapid ramp up (from ~3 to 200+ products) left little time for new tools β Time needed to hire people, β Not evaluate, buy, deploy and start using new tools β’ Still limited budget, especially for size of organization β’ Off-the-shelf tools were not viable β Incomplete automation and coverage β Concerns about fitting and integration into the future β UHGβs size and number of products were at a scale beyond most tools
- 17. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 17 Bespoke checklist benefits β’ Tailored β explicitly to enterprise needs β’ Integrated β into existing corporate tools and processes β Import from automated testing tool β Export to Governance Risk Compliance (GRC) application β’ Optimized β for changing accessibility program requirements β’ Control β multibillion dollar companies donβt like risk
- 18. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 18 Downsides? Many β’ Time to build β’ Cost and resources β’ Support and maintenance Not for everyone, but the lessons and techniques can still be useful
- 19. Bespoke checkpoints Custom tailored testing of success criteria #bespokewcagchecklists
- 20. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 20 Checklist v1.0: One checkpoint per success criterion Too small β’ Obvious, logical mapping for 38 checkpoints (WCAG 2.0) β’ 1:1 relationship ensured full coverage for WCAG conformance (and VPATs) β’ Didnβt have many projects so it seemed acceptable initially Problem: WCAGβs variable success criteria complexity β’ Many criteria are much more detailed than others β’ SC1.1.1 Non-Text Content β SC3.1.1 Language of Page β’ Often too broad, covering multiple different cases β’ SC1.3.1 Info and Relationships: covers forms, labels, tables, groupings, lists, etc. β’ Often difficult to describe effectively to other roles (developers, designers, authors)
- 21. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 21 Checklist v2.0 β One checkpoint for each sub-case (in success criteria) Too large and cumbersome β’ Understanding documents for each criteria provide the real details useful for testing β’ One for each situation β standard WCAG building blocks β’ Add other sub-cases such from success criterion text as needed (requirements, options) β’ Added our best practices and priorities β’ Checkpoint potential count of 113, but started with 84 at take-off Problems: Duplication, overlap, confusion, extra work β’ Examples β SC1.1.1 Non-Text Content went from 1 to 11 checkpoints β Multiple references to required attributes β’ Example: SC1.3.1 Info and Relationships, SC3.3.2 Labels or Instructions
- 22. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 22 Checklist v3.0: Reviewed, revised, rewritten, reduced Custom tailored to our needs β’ Reviewed all 84 checkpoints β’ Revised and rewritten for clarity based on experience of growing team β’ Merged checkpoints to remove duplication and overlap β’ Removed checkpoints mapped into new versions for continuity β Still at least one per success criterion to ensure WCAG AA conformance testing β’ Reduced count by 22 from 84 to 62 (26%) β WCAG 2.1 (2019) added 15 making current count 76
- 23. Bespoke groupings Custom tailored principles and guidelines #bespokewcagchecklists
- 24. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 24 WCAG principles and guidelines are for accessibility professionals They do not work well for product testing β’ Accessibility topics are not familiar to business owners, designers, authors, developers β Some terminology is β’ Particularly obscure (time-based media) β’ Vague or not actionable (adaptable, robust, compatible) β’ Difficult to find β Do not align with roles or features (forms, layout, presentation, semantics) β Some topics span success criteria, guidelines and principles
- 25. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 25 Custom groupings Group checkpoints for your audience and their needs β’ Size to audience needs β Fewer, simpler groupings for stakeholders (8 categories) β More detailed groupings for testers (10-11 test groups) β’ Organize by subjects β Functional: keyboard operation, forms, tables β Role: presentation, content and semantics, programming β’ Structure to meet needs of product types β Emphasize key functions (error prevention and handling)
- 26. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 26 Custom groupings Benefits β’ Use groups appropriate for testing and reporting needs β Test using groups optimized for assessments β Create stakeholder reports using more meaningful categories β’ Filter by topic (display all the table-related or image checkpoints) β Easier search β no need to memorize guidelines or criteria β’ Filtering allows easier βbatch settingβ of values for videos, forms, tables
- 27. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 27 Custom groupings Test groups for accessibility engineers (WCAG 2.1) 1. Keyboard / Input 2. Mobile 3. Navigation 4. Forms 5. Tables 6. Images 7. Presentation 8. Semantics 9. Content 10. Code 11. Time-Based Media (TBM)
- 28. Bespoke sort orderings Custom tailored numbering and sequencing #bespokewcagchecklists
- 29. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 29 WCAG order is a jumble Principles, guidelines, criteria order are good for learning, butβ¦ β’ WCAG order is inefficient for testing β SC1.1.1 Non-text Content ο GL1.2 βVideoβ ο SC1.3.1 Info and Relationships ο SC1.3.2 Meaningful Sequence ο SC1.3.3 Sensory Characteristics β’ Must switch testing modes and techniques or constantly jump back and forth β SC1.3.1 Info and Relationships ο GL3.3 Input Assistance β’ Critical issues come later β SC2.1.1 Keyboard Operation is the 21st criterion of WCAG 2.1, A & AA only β WCAG 2.0 order wasnβt great, WCAG 2.1 made it worse β New criteria added after originals, starting over with level A
- 30. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 30 Custom sort orders Create a logical sequence for testing needs and efficiency in addition to WCAG β’ Prioritize group order β Common, critical groups first (such as keyboard, navigation, forms) β Less critical later (time-based media) β’ Start with showstoppers to confirm testing can proceed β Example: focus ο keyboard operation ο keyboard trap β’ Sort checkpoints within groups to follow more efficient testing protocol β Example: content images ο images of text ο complex images ο decorative images
- 31. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 31 Checkpoint numbering tips Encoding IDs β’ Mandatory IDs β row numbers become meaningless after sorting β’ Unique IDs for each checkpoint β duplicates will get out of order after multiple sorting β’ Extra digits β for consistent sorting of multiple checkpoints to a criterion or sub-case β’ Base on WCAG β provides quick reliable default WCAG order (after using test sort order) β’ Leading zeroes β for GL1.4 β 1.4.01 or 1.401 β If not, they never sort into to the expected WCAG order β’ 1.4.1 β’ 1.4.10 β¦ β’ 1.4.13 β’ 1.4.2
- 32. Bespoke severity levels Custom tailored level indicators #bespokewcagchecklists
- 33. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 33 A, AA and AAA donβt work well for testing WCAG AA isnβt very meaningful for prioritizing checkpoints Itβs really all one group (all A & AA) β’ AAA (usually) not included β’ Does anyone target Level A conformance? A and AA are too broad for processes like Agile β’ Severity scales typically have four levels β Example: MoSCoW method β Must / Should / Could / Wonβt β’ Which are showstoppers? blockers? major or more minor barriers to use? β SC2.1.2 No Keyboard Trap = SC4.1.1 Parsing? Both are Level A
- 34. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 34 Custom severity levels Sample levels β’ Critical Showstoppers or blockers to site access, operation and testing β’ High Could be very challenging to overcome, impossible for most β’ Medium May be very frustrating or time consuming for all β’ Low Might have significant impact for many Benefits β’ Not forced to WCAG levels (Example: raise SC2.4.7 Visible Focus from AA to Critical) β’ Align to product and project needs (like minimum viable or accessible product) β’ Write so they cannot be optional (must include only) β’ Align and reference by WCAG levels as needed
- 35. Other bespoke details Additional custom tailoring #bespokewcagchecklists
- 36. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 36 WCAG options WCAG version β’ v2.0 β only β’ v2.1 β complete (2.0 and 2.1) β’ v2.1 β supplement (new 2.1 checkpoints only) WCAG level indicators β’ A β only β’ AA β only β’ AA β conformance (A & AA) β’ AAA β if applicable (A & AA & AAA)
- 37. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 37 WCAG options (cont.) The Four POUR Principles β’ P β Perceivable β’ O β Operable β’ U β Understandable β’ R β Robust Could include Guidelines β’ 1.1 Text Alternatives β’ 1.2 Time-based Media β’ 1.3 Adaptable β’ β¦
- 38. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 38 Custom checkpoints can be distributed by skill Optimizing work to screen reader and visual testers Custom checkpoints can be tailored and grouped based upon skills β’ Screen Reader Tester / SRT AT expert using screen reader β’ Visual Assistant / VA User working relying on vision, not AT Filtering allows clear, consistent work assignments β’ SRT Screen Reader Tester preferred β’ Either Screen Reader Tester or Visual Assistant can perform equally well β’ Dual Both should test (AT experience very different from visual) β’ VA Visual Assistant preferred (usually for efficiency)
- 39. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 39 Map checkpoints to non-WCAG standards Section 508 (1998 or βpre-refreshβ) β’ Β§ 1194.21 Software Applications and Operating Systems β 12 criteria (a) β (l) β’ Β§ 1194.22 Web-based Intranet and Internet Information and Applications β 16 criteria (a) - (p) β’ Β§ 1194.24 Video or Multimedia Products β 5 criteria (a) β (e) And possibly transition to future standards like AG (WCAG 3.0)
- 40. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 40 Other information for reports β’ Additional options can be added for checkpoints and reports to communicate benefits Example: βAffects users who areβ¦β [blind, low vision, deaf, cognitively impaired, etc.] β’ Accessibility tools for testing List of tools to test checkpoint β useful for developers and designers References links to standards and resources β’ Roles List of roles that may own the issue or should be consulted about the checkpoint β Example: color contrast issues (SC1.4.2) likely belong to visual designer
- 41. Final fitting Putting it all together #bespokewcagchecklists 1
- 42. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 42 You can and must support all WCAG β’ Success criteria β’ Guidelines β’ Principles β’ Conformance β’ Reporting (VPAT) How you get there is up to you.
- 43. Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 43 Custom tailor WCAG to your needs with bespoke checklists β’ Success Criteria ο Checkpoints β Create as many as appropriate with minimum of 1 for each criterion β Structure using Understand Documents (leveraging situations, options, best practices, etc.) β’ Principles and Guidelines ο Groupings β Tailor and size them to needs (and vocabulary) of target audiences β’ Numbering ο Sort order β Encode IDs so they provide consistent, helpful, efficient sequences for testing β’ Level Indicators ο Severity levels β Prioritize to needs of testing and delivered product β’ Other details as needed
- 44. Thank you. Bill Tyler Principal Digital Accessibility Engineer btyler@optum.com @billtyler
- 45. 45Β© 2020 Optum, Inc. All rights reserved. Confidential property of Optum. Do not distribute or reproduce without express permission from Optum. 4