Takeaways from Accessing Higher Ground 2019

Two weeks ago, I attended Accessing Higher Ground 2019 on behalf of Tenon. We sponsored the event, and decided to have a personal presence as well as having brochures in the conference bags. This was my first time attending AHG (as it’s known), but Tenon (and Karl Groves) has a long history there.

In case you’re new to AHG, like me, here’s a brief overview. AHG is a conference dedicated to accessibility in higher education, organized by AHEAD, the Association on Higher Education And Disability. It takes place near Denver, Colorado, and attracts hundreds of people from colleges and universities across the US, as well as a couple dozen vendors and many other accessibility professionals. They also webcast some of the sessions. There are 2 days of workshops, followed by 3 days of 1-hour presentations (and a few workshops). Topics ranged across accessibility intros, practical tips, standards, organizational guidance and processes, tools, working with students with accessibility needs, Universal Design for Learning (UDL), and more. The organizers built in a long lunch break, and a little time to interact with the vendors each day, which decreased my FOMO.

There are some distinct characteristics of AHG as an educational –not just accessibility– conference that set it apart from similar industry conferences that focus on design and/or development. For example, there were many “how to use this tool” sessions, and multiple sessions on essentially the same topic: UDL (Universal Design for Learning), PDF and remediation, EPUB, and writing alt text all had multiple sessions by different presenters, where in industry conferences, you usually see a wider variety of topics with less repetition. I think this has to do with the needs of the higher-ed community, which has to be focused on reproducibility, different learning styles, and the ease of conveying information to a semi-technical audience, i.e. the hundreds or thousands of “content creators”  (faculty, staff, and students) in a university setting, who are only dimly aware of accessibility techniques and requirements.

Here are some highlights from my personal favorite AHG2019 presentations (in order of appearance).

Future of Accessibility Guidelines for Web and ICT

Presenter: Jeanne Spellman (Tetralogical, Co-chair of W3C Silver Community Group and Task Force), @jspellman

An overview of the future evolution of the Web Content Accessibility Guidelines (WCAG). Code-named “Silver” (a clever pun on the chemical symbol for silver being Ag, which is the acronym for Accessibility Guidelines, and also new emphasis on age-related disabilities), the next version of WCAG has some changes in store around 2021, and Jeanne was the perfect person to detail them.

My 3 takeaways:

• The current ratings (A, AA, and AAA) are unclear and not very i18n-friendly; the next iteration of WCAG will use the well-known olympics model of Bronze, Silver, and Gold, with scores for each checkpoint to tell you how you stack up.
• Cognitive disabilities and other oft-overlooked guidelines are being moved out of the AAA neverland into more general requirements.
• There’s new emphasis on atomic tests, measurability, disambiguity, and suitability for use and reuse in testing tools; this complements the W3C Accessibility Conformance Testing (ACT) project, and the new guidelines might even be available through an API for easier reuse. 

Writing Good Alt Text

Presenters: Gaeir Dietrich (Consultant), @Gaeir; Robert Beach (Kansas City Kansas Community College)

Despite being a well-known problem, this is a continual source of confusion and uncertainty for many people creating accessible content. The popularity of this topic is evidenced by the fact that there were 2 different sessions on alt text –as mentioned in my preface– and both were well-attended and useful (see my takeaway of the other one later in this post). 

(In fact, at CSUN next year, my old friend and colleague Mark Sadecki and I will revisit this topic with what we think is a new and productive take, in our session “Descriptions of Charts and Complex Images Made Easy”. Hope to see you there!)

My 3 takeaways:

• Publishers like to reuse the same graphic for different purposes, such as a map of the US that shows states, capitals, regions, and other labeled features; consider the context of the specific usage, and only describe pertinent features for the particular task at hand.
• The speakers praised the approach by Adam Alonzo in “A Picture is Worth 300 Words”, which offers 5 guidelines: be objective; be brief; be descriptive; be logical in sequence; be accurate. I noticed a similarity between these principles and the linguistic Gricean Maxims of conversational cooperation. 
• For “layered” graphics, such as a stepwise slideshow that animates each step over time, you should “flatten” the graphic, and describe it as a complete picture, probably described in the order of each step. 

Building a Culture of Accessibility through Innovation

Presenter: Evan Yamanishi (W. W. Norton & Company), @sh0ji

W. W. Norton & Company is a very progressive and nimble publisher, and they’re pushing hard to make all their content born digital and born accessible. Evan went over their evolution from traditional authoring and publication approach to a champion of accessibility.

My 3 takeaways:

• Evan summarized his own talk in 3 points (handy for me!):
  1. Accessibility should be used to push innovation forward, not hold it back
  2. Building a culture of accessibility is best accomplished through empathy and shared ownership
  3. Standards and best practices in web development can accelerate both accessibility and innovation
• He also demoed several projects:
  • Norton’s in-house Node.js accessible ebook platform
  • A component for automating the creation of accessible interactive transcripts, using HTML5’s <track> element (I’m pushing him to make this open-source!).
  • Their new design system, with accessible color palettes generated from raster image sampling, and an image description interface using universal design.
• Evan noted how the accessible color system wasn’t done by his accessibility division, but by the designers themselves; once he introduced them to the principles of accessibility, they embraced it and ran with it, which is an organizational triumph. 
  

Preparing Accessible Math Documents using MathJax

Presenter: Dr. Volker Sorge (Progressive Accessibility Solutions), @zorkow

Volker Sorge knows more about accessible math and graphics than most people on Earth. He’s a major contributor to MathJax, a JavaScript module for rendering math in the browser. This hands-on 2-hour session focused on how to use MathJax v3 to publish accessible math, which is crucial for STEM education. Volker was practiced in delivering this course (he’s offered it before, and will likely do so again); the course was pragmatic and useful, and I’d recommend it to anyone who has to handle math accessibility, if they get a chance.

My 3 takeaways:

•  Math is unsurprisingly difficult to render, and to navigate and understand for screen reader users; MathJax has put significant effort into multiple options for accessibility of math.
• Pandoc, which we used in the class to convert from LaTeX to HTML, is an amazingly versatile document conversion library. I’m going to add that to my toolbelt.
• Browsers should natively support mathematical rendering, but MathML (the ill-supported W3C standard for publishing math) might not be the answer; MathJax has already diverged from MathML in significant ways.

Learning Process Automation: Bringing Assistive AI to Accessibility

Presenter: Rajiv Narayana (ansrsource)

There’s a lot of hype about Artificial Intelligence and Machine Learning in general, and the accessibility field is no exception. Raj offered a sober look at current trends toward automation and the role of algorithms in our work future.

Karl Groves and I will puncture the hype balloon at CSUN with our talk, “AI and Web Accessibility: Opportunities and Pitfalls”. Don’t miss it!

My 3 takeaways:

•  AI, ML, algorithmic approaches, robots, and software automation have already changed the world of work, and they’ll continue to do so, but often in unpredictable ways.
• For the tasks that they’re doing, including automating image descriptions, the computer may be “99% accurate”, but it still makes critical mistakes, so a human needs to be in the loop. They see AI as a workflow enhancement tool to improve consistency, quality, and accuracy.
• Education and business, including the accessibility aspect of both, need to avoid leaning into the “As-Is” state of their operations, and let their process be guided by the “To-Be” state they anticipate for the future.

Writing Effective Alternative Text for Educational Content: Best Practices for Reducing Cognitive Load

Presenter: Valerie Morrison (Center for Inclusive Design and Innovation)

This was a lively session, packed with attendees, and very interactive. An excellent end of the conference for me. I attended this session, like the previous session, to pick up tips on delivering the message alt-text message myself at CSUN, and I wasn’t disappointed. Valerie knew the topic well, since it’s a major part of her job. I enjoyed Valerie’s approach, and while I didn’t learn much about the topic itself, I took away valuable reinforcement of my own thoughts on it, and notes on presenting it well.

My 3 takeaways:

• The key of an effective text description is to decrease the cognitive load on the reader, and give them just enough information to solve the task. This is my own experience, as well, and I’ve been saying this for a while, so it’s good to have it validated by someone who does this for a living.
• For complex images with symbols and graphical notation, Valerie suggests not describing the representation itself (e.g. “the red arrow line”), but to only focus on what the symbol represents (e.g. “the path of de Gama’s ship”). I don’t wholly agree, since I think there’s a place for more detailed description of the representation, but as the first-pass description, I think that’s a good rule-of-thumb.
• This picture (from Valerie’s slide deck) of a gargantuan cat lounging on the Istanbul skyline elicited some surprising descriptions from the audience, some of whom didn’t even notice the cat! (But were able to identify Istanbul by the Hagia Sofia…)

Conclusion

Personally, I found the event well-curated and productive. Even though I’m not the target demographic (I worked for a university only for a few months), many of the sessions were general enough that I had plenty of choices to learn and interact with other attendees. Most of the sessions during the 3 days I attended were high-quality, and though I’m pretty experienced, I got something out of each of them. And the “hallway track” and dinners were especially useful, making new connections and seeing old friends and colleagues from W3C, Benetech, CSUN, A11yTo, NC universities, or online. Though I was new to this conference, I knew so many people that I felt at home. (And of course, the accessibility community is especially welcoming and diverse.)

I’d recommend this conference to anyone in accessibility in the higher-ed world, or to anyone who has clients in higher education. It’s a good size, with lots of people but much less overwhelming than CSUN, and very focused on its goals.

I hope to see you there next year!