How did Thomas Logan get started in accessibility? It all started in 2002 when he was a student at the University of North Carolina. There, he met Jason Morris, a graduate student at UNC. Jason needed access to the ancient maps as part of his work toward a degree in the Classics Department. Jason couldn’t see the maps, so Thomas got involved in creating accessible ancient world maps.
This was Thomas’s first introduction to accessibility and caring about it knowing someone that needed it and could benefit from the work. This gave Thomas’s work meaning and continues to drive him today in his accessibility work.
Dr. Gary Bishop, Thomas’s professor, met Jason and talked to him about what he needed. He asked about the access challenges he had. Jason explained that he couldn’t access the maps to do his research. Dr. Bishop turned it into a project called Blind Audio Tactile Mapping System (BATS).
Dr. Bishop posed this problem in computer science class. Thomas credits his professor for his interest in accessibility. The media reported on the project as Wired and Python Success Stories wrote about it.
However, having the story of helping a person with a disability do something with technology doesn’t mean that it solved the problem. The same problem existed 15 years later and Thomas worked on it again as an accessibility expert witness in a legal case that involved all of the large corporate mapping providers
Thomas’s first project only addressed one map, and working on his later legal project made him realize that there were still many maps that remained inaccessible. The question became how do you make every map accessible? Accessibility supporters need to influence the big companies that are responsible for building these technologies to get them to make the change. Change can’t come from academic technical demos. It has to come from the commercial world.
After completing the BATS project for the ancient world, Thomas met Barb Riverdahl, a blind person who was at the University of North Carolina at Chapel Hill. She shared the barriers she ran into around campus. She explained that she creates a cognitive map that she mapped in her mind.
Nonetheless, the campus continuously added new barriers or something broke down, such as tearing up a sidewalk that she didn’t know about. The space between the sidewalk and the street was not detectable as there was no change in the pattern. She also walked into many side mirrors on trucks and vehicles that were illegally parked.
Thomas said these problems still exist. There are opportunities to make a change. He believed that someone working in technology or computer science only thinks about what’s in front of them on the screen. The key is to meet people and learn about the barriers they run into every day.
Microsoft and Test Automation
When Thomas went to Microsoft, there wasn’t much work in accessibility or training. Microsoft did a PR piece with the title “Student finds life’s work.” Thomas didn’t know how he felt about the work then. Twenty years later, he admitted Microsoft was right that he couldn’t get away from accessibility. A lot of the people he worked with in accessibility still do after more than two decades.
When people get passionate about accessibility, it’s often hard to go into another area because you get invested in it. Once this becomes a passion, you can’t stop being passionate about it.
A lot of the ideas that come up in technology are celebrated as new ideas. But there’s usually prior implementation. One of the first things working in his job at Microsoft was the idea of these control patterns called UI Automation.
Here’s the explanation of UI Automation from Microsoft: “UI Automation provides programmatic access to most user interface (UI) elements on the desktop, enabling assistive technology products such as screen readers to provide information about the UI to end users and to manipulate the UI by means other than standard input. UI Automation also allows automated test scripts to interact with the UI.”
Thomas believed the original implementation was better than what exists today. There were a lot of arguments over whose implementation was better. That didn’t matter. Ultimately, what matters is to implement something that works and fixes the problem.
The control patterns were this idea of speech control of software interfaces and having speech synthesizers for people who are blind. The gist was the interface could read out and you can control the interface with your voice. Unfortunately, controlling the interface has a lot of security concerns.
This was another problem that hasn’t changed in 20 years. If you let someone control a computer, that could be a hacker or it could be someone helping a person with a disability. How do you balance security and accessibility? Giving permissions to allow things to be controlled is tied to security and accessibility. Even so, the same problem exists today in that you can’t enable control without introducing security problems.
Japanese in Windows Vista
While working on Windows Vista at Microsoft, Thomas collaborated with Japanese project managers on enabling access to type and speak Japanese on Windows Vista. After living in Japan for four years, the problem made more sense to Thomas than when he worked on it.
There are three ways to type “cat” in Japanese. There’s the kanji, hiragana, and katakana. This is what he needed to know and understand to design the feature to work well for people. Being assigned to this at Microsoft, he didn’t know anything about Japanese, much less there were three ways to write letters. He was told to make it accessible.
He thought he did a good job. But the reality couldn’t be further from the truth. He needed more awareness and education to be able to do the Japanese implementation into Vista. In hindsight, Thomas recommended asking more questions and trying to learn the problem.
Native Screen Readers for Operating Systems
A screen reader is what a blind person used to interact with the operating system. Beware that it’s not just blind people who use screen readers and not all of them use them. Anyway, Microsoft made the decision to not have a single screen reader for the operating system. At the time, Microsoft was worried about a consent decree. They were also concerned that if they made a screen reader that was good, it would put these other screen readers out of business.
Thomas disagreed because you need to have a good screen reader for your platform. Then, you should encourage people to innovate beyond what you provide. It turned out Microsoft Narrator turned out to be a bad product. People couldn’t use Narrator to access Windows.
This experience compelled Thomas to advocate that if you build a platform, you must have a screen reader that works. A better approach is to build a screen reader that’s part of the platform. People still have other options to use if they prefer them. Every operating system should have its own screen reader.
Working with Web Standards
Thomas worked on accessible rich internet applications (ARIA) 1.1. He personally believed it was important to work on these technical standards. His takeaway is that we need these standards, but we have to be careful not to overdesign or make things too complicated.
Desktop computers had sliders. During those days, if something worked on a desktop computer, then it should work on the web. There wasn’t a way to have a slider. Hence, one of ARIA’s solutions was it can help you make a slider that works on the web. They made it happen.
Still, consider all the different kinds of controls on a desktop. They had to document the most minute thing. This includes every type of control that used to exist on the desktop and put it on the web. What ended up happening was that none of the implementations worked or they’d only work in one browser.
If an idea doesn’t work for people with disabilities, then why do it? It took a long time to get a control implemented and it worked consistently. That’s really hard when you’re trying to tell someone to do it and you demonstrate that it actually works for someone.
Because it’s more theoretical. This, too, exists today in the work on standards. They should make sure they have the demo of it working to explain to someone why they should do it.
Thomas built a site called Queue Music and made toolbars. It was a way to queue up the Flash player and queue up songs on YouTube. It was lists, a toolbar, and a slider. Thomas built all these things to show that you can build this on the web and here’s how. Mozilla paid him to do that. What was the result? He got it to sort of work and only with Mozilla.
If you can’t show that a new standard works, then it’s hard to convince someone else to do it. A lot of this work relied on theories. It would be better if web standards were simple to understand and easy to implement.
The Real Value in Accessibility Audits
His time as CTO of HiSoftware informed how he set up his company Equal Entry to work differently by solving accessibility problems. At HiSoftware, he worked on big projects that used an automated scanning tool. It could scan a million web pages and identify the different problems.
Proctor and Gamble was one of their clients. They owned 65 brands and many different websites. They also worked with Veterans Affairs, which had many different regions. The company scanned all the client websites and all the VA regions. You think you’re making the world accessible.
They were producing reports with charts showing red for failed, yellow for warning, and green for passed. At the time, clients would be okay with more green than red and look at whether the green went up from the previous update. These charts didn’t mean anything because they didn’t indicate whether a person with a disability could use the site. Nonetheless, that was how the clients translated the reports. It shouldn’t work that way.
For example, one chart showed a passing rate of 73%. That sounds great, right? As long as every updated chart shows a higher passing rate than the previous chart, that’s good, right? However, the results show over 600 images with an empty alt attribute. These were not decorative images. Knowing this, can you say a 73% rate is great when someone has no idea what the images are about and can’t use the website?
The next scan may see the number dropped to 500. Does that mean someone can use the website? No. It was still a barrier. It didn’t make sense for businesses to look at only the passing rate. They need to be looking at whether someone can use the website. Can they log in? Can the purchase something? Can they accomplish the tasks they set out to do when visiting the website?
Thomas applied this learning in building Equal Entry’s business. He didn’t want clients to get excited because their passing rates climbed. Accessibility audits should focus on real problems with specific end results, not technical compliance. Equal Entry looks at whether people can use the website, app, or other digital product.
The overlay companies of today claim they’re doing something new, but HiSoftware was doing this in 2007. It had all the same problems as today’s overlay products. It didn’t work because it didn’t fix the problems. Years later, overlay companies claim you can automagically add one line of code and it’s magically accessible. It didn’t work then. It doesn’t work now.
Select any of the bullets to jump to the topic on the video.
- Creating accessible maps
- Learning about campus barriers
- Microsoft and test automation
- Working with web standards
- The value of accessibility audits
- Q&A with Thomas
Watch the Presentation
For 20 years, Thomas Logan has helped organizations create technology solutions that work for people with disabilities. Over his career, he has worked with startups to Fortune 500s including Microsoft, Google, Meta, Bank of America, New York Times, Princeton University, and Disney. He has also done projects for many federal, state, and local government agencies in the U.S.
He’s the owner of Equal Entry, whose mission is “contributing to a more accessible world.” He’s the founder of Accessibility Virtual Reality and co-founder of Accessibility New York and Accessibility Tokyo. These are Meetups related to accessibility and people with disabilities.