Discovering Accessible Learning Resources with Benetech Labs—Part 1

The digital revolution and ongoing advances in technology have made it possible to get more content, in more ways, to more people. This could truly be a golden age for people with print disabilities, such as those who are blind or have dyslexia, who cannot access or efficiently use books, apps and other information in most standard formats. But what good is accessible content if no one can find it?

Imagine yourself in a situation where you might need accessible content—say, a science textbook with image descriptions, because you were a student with low vision. Or, say you needed an audio format of a textbook, because you happened to have dyslexia. If you searched for these online accessible resources in alternative formats, you’d quickly realize that it’s very difficult to find them. In fact, oftentimes they wouldn’t show up in your search results at all, although they may as well exist. Why is this the case, at a time when accessible digital content in alternative formats is becoming increasingly more available?

The answer is: the lack of information about the accessibility features of applications and digital content—features like supported accessibility APIs, keyboard accessibility, captions, image descriptions or other specialized formats. This information is called “accessibility metadata.”

At Benetech, we believe that as accessible digital content and applications increasingly become more available, they must also be easily discoverable. That’s exactly what we set out to do through the Accessibility Metadata Project, one of our Benetech Labs projects.

Before I describe how we’re addressing the lack of accessibility metadata, let me first explain better how metadata relates to the World Wide Web and why it’s especially important in the case of learning resources.

Metadata in the context of the World Wide Web is information webmasters use to markup their pages in ways recognized by the major Internet search engines, such as Google, Bing or Yahoo! This information (webmasters also call it Schema.org rich snippets or microdata tags) is important because it allows these search engines to effectively comb the vast online content and return narrower results based on defined categories, instead of thousands of hits.

Now let’s take a short virtual trip into the world of educational content. Here, as textbooks, exercises or labs are becoming more granular and therefore increasingly available by the chapter or concept, it’s already difficult to create a focused search to effectively find these learning resources on the web. That’s why a group called the Learning Resources Metadata Initiative (LRMI) has created specifications for educational content metadata. The problem is that this group’s work hasn’t included accessibility metadata—that is, information about the accessibility properties of educational content.

This lack of accessibility metadata means that if you search for, say, a diagram of the heart on Google, you’d have no easy way to filter those diagrams that contain accessibility properties like captions, image descriptions, audio or other specialized formats. You’d need to know how to search for each content type in multiple specific locations and even if you do have that knowledge, your search results might still not reflect what’s available, because the available content isn’t encoded with metadata for accessibility properties. The paradox, then, is that while these accessibility features are becoming more widely available, it’s getting ever more difficult to find the right accessible learning resources.

To address this challenge, Benetech is leading a group effort by experts in the field of educational resources and accessibility. Together, and with support by the Bill & Melinda Gates Foundation, we’re developing the accessibility components of educational metadata so as to extend the LRMI. These added metadata items will make it possible for students, teachers and parents to discover, evaluate and use the educational resources that best fit their specific learning needs for maximum success in the classroom and beyond. Moreover, the accessibility metadata we develop will be applicable to all creative works, such as blogs, books, audio, video and more.

Our goal is to make accessibility metadata tagging a standard part of the publication process. We see this step towards making accessible content discoverable as key part of the shift into a world where people can widely find and use accessible online materials. We call our vision of this brave new world “Born Accessible”: a world in which all content born digital is made accessible—and discoverable—from the outset.

In October of this year, I joined the World Wide Web Consortium (W3C) to deepen Benetech’s participation in the evolution of standards that enable educational content to be born accessible. Our first opportunity for deep engagement came earlier this month with the W3C Technical Plenary and Advisory Council (TPAC) meeting in Shenzhen, China. You can read more about the latest developments advancing accessibility in digital publishing in my update on W3C’s blog.

Meanwhile, we’re also making great progress with the Accessibility Metadata Project. Here’s a taste of what we’re up to:

Earlier this year, we submitted to Schema.org—an initiative supported by the major search engines—our group proposal for accessibility metadata that would make it widely possible to find online content with accessible features. Today we have big news to share: the Schema.org team has just announced it agreed to adopt the accessibility properties we proposed! This is a tremendous milestone in the collaborative journey towards enabling a born accessible future and reaping its benefits.

To get a better sense of what such a future could be like, consider this implementation case of our accessibility properties:

Over the summer, we incorporated a draft accessibility metadata specification into Bookshare, tagging our entire collection of over 220,000 titles with accessibility metadata. This means that information about Bookshare titles—from content type, to age ranges, to available media formats—is now available in a manner that makes its discoverable by search engines. For example, you can find textbooks in Bookshare with image descriptions via a Google Custom Search query.

We have a number of other cool project implementations in progress, which I’m presenting these days to key stakeholders in the education technology and accessibility community at various conferences and workshops. I’m glad to say that there’s enthusiastic interest in the outcomes of our work and their implications.

I look forward to sharing insights from our test project implementations in a future blog post. For more updates on our work, please check back the Accessibility Metadata Project website.