On January 7, 2020, Accessibility New York City hosted a Meetup entitled “EdTech and HigherEd: Considerations for Digital Accessibility.” Presenter Zarina Mustapha, Senior Front-end Developer and Accessibility Lead at the Center for Teaching and Learning (CTL), Columbia University, outlined the importance of thoughtful design for educational technologies for accessibility and inclusion. She shared her experiences, challenges, and solutions in integrating learning design frameworks into the making of accessible educational technologies.
Here are some of our highlights from the presentation:
- Accessibility In Itself is Not Sufficient
- How Tech has Changed the Meaning of “Course” in Higher Ed
- Accessibility and the Science of Learning
Transcript: EdTech and HigherEd: Considerations for Digital Accessibility - Zarina Mustapha
ZARINA: So tonight I'm gonna be talking about ed tech in higher ed. The considerations that have to be taken into account when you're working on digital accessibility of content or digital accessibility of educational technology in classrooms. So as Cameron had said, my name is Zarina Mustapha, I'm accessibility developer for the accessibility technologies group for the Center for Teaching and Learning at Columbia University. There's multiple layers here, to show you the levels of distance between the university and the department and the group within the department.
So for tonight, I am going to be talking about... First I'm gonna go have an overview about: What is EdTech? And what is higher ed? In the context of my talk tonight.
And why accessibility is so critical in higher ed, and from that evaluation, of accessibility in higher ed, we'll start to consider in my experience what we do with software and media -- looking into universal design for learning, and how it maps into accessibility. And I'm gonna demonstrate a couple of our works in software. And also media on how we apply these principles in order to make all of the tools that we make for university use more accessible.
And what we've learned, or at least what I have learned from making all these software and being a consultant to the development of media in classroom materials, on what are the lessons learned from making all these things accessible. So the CTL that Cameron mentioned earlier is the Center for Teaching and Learning. And our mission is basically to support excellence and innovation in the culture and practice of teaching and learning. So teaching and learning is education, but they're two sort of counterparts that move together.
We're a department of 40 people, and we offer services in consultation to Columbia faculty and graduate instructors and lecturers, and that's how we basically advocate the practice and culture of teaching and learning. And the group that I'm in is called instructional technologies group. Two of my colleagues are here. Alexis from the media team and Nick from the software development team.
We make software and we make media. Sometimes from scratch. Sometimes it's an evaluation of other tools that we put together with the needs and use of a particular classroom or a particular project at the Center. So how do I get involved as a frontend developer in accessibility?
If you are a frontend person, it's about user experience design. And in the things that we make, there are basically software and web application tools. There are users, and our users are students. And these tools we make sometimes from scratch.
So they're in-house. And it's being used in the university. So it is educational technology. And in the process of making more and more tools, at least about four to five years ago, we started to migrate from old technology to new technology, old content to new content, just basically updating the applications, and starting to notice that they're not very responsive with all the technologies that we have right now.
In terms of tablets or smartphones. And then of course, they're not accessible, because accessibility wasn't a huge consideration ten years ago, because things were moving so rapidly, and now that's how I got into looking into accessibility. Some of our projects are funded by government grants, and so accessibility is built into the grant, and we have to consider what that means. And then...
So in making all this educational technology, software and media, I realized that in the accessibility -- in itself -- is not sufficient. There is another component to it, where you do have to consider how people take information, how people process the information, apply the information, and turn it into knowledge. Information without contact doesn't lend itself as knowledge. So the software and the architecture of how we actually are going to craft of the content or the functionality of the tool has to comply with some form of science of learning.
This is not... Okay. Oh, okay. So this is basically...
I didn't update the slides. I'm sorry. But you get the idea. And so we're going into what is educational technology.
To put it simply, educational technology is technology that promotes teaching and learning. Microsoft Word is educational technology if you use it to type your papers and deliver it to your professor. However, it's a little bit deeper than that. Technology is software, media, hardware, the theories and practice of all these tools, how effective is it, how does it interfere, how do people use it, how do people process the information through all this technology, how do we integrate all this technology into the curriculum, into the syllabus, into the classroom, into the wider university system?
And how does that facilitate -- how does that advance, how does that improve teaching and learning? So yes, the statement is indeed technology that supports or promotes teaching and learning, but all together, it's actually a discipline that people usually refer to as educational technology. So the context of higher ed. In my talk here tonight, higher ed I define -- based on the definition by the National Center for Educational Statistics -- is postsecondary.
It's after graduation from high school. So it starts usually with a bachelor's degree. And beyond. Masters degree.
Et cetera, et cetera. It can be a two-year, like a community college, or a traditional program of a four year-plus span. The age range -- the youngest typically for an undergrad is 18 years old. More and more people are enrolling as a first year at the age of 19.
They take a break. Or even at age 20. And in higher ed, students do have a higher independence. The classes are -- the schedule, the calendar for their classes are basically a little bit more flexible than, say, secondary or primary schools.
They are now... They have to manage their own time. They have to manage their own expectation, and they have to manage their own discipline. So it's a lot of work to become a college student.
And by the way, this is a definition of higher ed in the United States. It's a little different in other places. So I don't think the principle is terribly different, but there are some... What do you call it?
Differences. So let's take a look at the youngest enrollment, as a freshman or as a first year student. Who is enrolled in fall of 2019. Okay?
This person is born in 2001. And Wikipedia was introduced in 2001. So you want to feel old? This is the session.
I'm gonna make you feel old. Facebook. Became public in 2004. And then two years later, Twitter becomes noisy.
And then when the kid is six years old, Steve Jobs walked onstage and introduced the iPhone. And then a year later, Android followed suit by introducing other smartphones. And then three years later, the iPad came to be. And now here we are.
After the kid is six years old, she or he can play with the iPhone, any smartphone, play with the iPad, play with the computer, everything. By the way, when was Google introduced? Anybody know? 1996?
1998 is the formal date. So this is three years before this undergraduate was born. So yeah, we feel old now. So you can see by six years old, they're very familiar with all these tools, these portable tools, mobile tools.
And information now is very, very much readily available to them. It changes the way people learn. It changes the way people teach. So technology has changed the landscape of what is the meaning of a course.
A course is traditionally a face-to-face thing. You come -- we are doing a face-to-face, in-class meeting, in-class learning materials. But then there's distance learning, or in this case, the blended learning, where it is livesteamed, where someone is actually watching right now with us. Now, face-to-face with web -- you started giving some of the learning material in LMS, is learning management systems.
So this platform where people can -- where faculty can put up their syllabus and learning materials, videos, et cetera. So you still have the in-class meeting. But all these materials are probably uploaded in the LMS. Or even the faculty's website.
Now, the blended learning is a little bit more separated from the face-to-face classroom. So you still have face-to-face meeting. But a lot of the learning instructions are on the LMS. A lot of the materials, learning materials, are on the learning management system.
And the discussion boards are also probably online. And so on and so forth. And then we have online. Which is distance learning and also the massive open online courses.
The MOOCs. Everything is completely online. The learning materials are online. The meeting is actually discussion boards, also online, and the instructions are online.
The faculty is completely absent. And the connection between the learners are through discussion boards. You don't meet anybody. Unless you actually have a video conference.
Which is also online. So you start to see how technology, as it improves -- we're becoming more and more diverse in how we define a course, who can take these courses, and how these are available to people. So you can see now where accessibility is becoming more and more important. So accessibility in the year of 2015 -- this is according to the National Center for Education Statistics -- 19% of undergraduates in 2015-2016 reported having a disability.
Now, notice the word "reported". What does that mean? That means these students actually approached the Office of Disability Services and requested support and services in the classroom. So it's self-disclosed.
There are many students who do not do that, for whatever reason. And we don't know how many people who need... Who are disabled and of that, 11.9% are postbaccalaureate. So masters degree, PhD.
Existing support and services tend to be very reactive rather than proactive. So I can give you an example. At Columbia University, we do have the Office of Disability Services. A student will approach, and I guess put an application for support.
And then the ODS will give either technical support or personal support, or they can sit with the faculty member and request that the faculty do some adjustments. And the faculty will do these adjustments. But this is all reactive. And it relies so heavily on self-disclosure.
So we don't know who isn't getting the materials or the learning or the teaching. So what are we doing here? So here is a typical faculty at a university like Columbia. Because I'm from Columbia.
That's my experience. A faculty in a course will have a syllabus. And the faculty will either upload it or hand it to the students. If it's uploaded, it's digital.
And if it's online, it is usually on a learning management system. Or on a particular website by a department. And it can come in the form of Microsoft doc or Google Docs or a spreadsheet or something that is either online... It's still digital.
A PDF. Media, meaning video or any kind of interactivity, any online tools like Poll Everywhere. Like Google Docs is a form of a tool. Any external resources like libraries.
Linking to the New York Public Library. Resources that are available over there. And in terms of communications. Email.
Skype. Zoom. Phone calls. You name it.
So you can see that some of these materials are either generated by the faculty, generated by consultants that the faculty are working with, like -- what are they called? The instructional designers. They work with the faculty to design the course. And the LMS, for example, is a form of a third party tool that is supported by the central university.
So the university makes an effort to give formal support. But then there are unsupported third party content tools. We download apps all the time. So who's gonna be responsible for whether or not the tool is accessible?
So you can see this is just one course. If any of these materials are not accessible, it is a form of discrimination. You're not gonna be able to access a syllabus, which is to start your course. If the syllabus is not accessible.
You're not gonna be able to access a reading material if it's a PDF that's OCR in the form of an image. So then you become... What are you going to do? And so you seek support from the ODS.
That doesn't sound quite right to me. Because the student will have to do the work in order to learn. I think that's not ideal. I don't know.
If anybody else disagrees with me... Fight me. So it is not enough also that it is just about compliance. That's something that I have realized.
That compliance is one thing. But the context of all those materials, of all those objectives and activities and videos, whatever. You can have a picture of a cat with alternative tags, equals dog, and so there is no knowledge that I am actually viewing a picture of a cat. So the information that I'm receiving is dog.
But it doesn't match. So the knowledge is not formed to look at this particular picture of an animal, to identify what species that is. So we need a little bit more than just compliance. Compliance is the easiest part.
You could run through tech -- field validation. Done. You follow the rules of the web content accessibility guidelines. You get it.
But it's not enough. The science of learning. How people take information, process information, formulate new ideas from that information, and give it back to the community in the classroom -- that is the science of learning. How do people use technology to access the information for that science of learning to happen?
Plus compliance. Is what I call accessibility experience. If anyone remembers, if anybody had attended a talk by... I think it was the person from the museum...
I think her name is... I'm blanking out. But... Aurelia Flori?
She created "accessible experience" as an alternative form to art. But it's actually exploring the art. So it's not just compliance. It's not saying that this is an image of Van Gogh.
But it is an experience of the painting by Van Gogh. The different strokes that they use, that he used in his painting. So that is an experience that we have to create in order to make things not just compliant, but actually useful and learnable. Are there any frameworks that will help us with this?
Anybody know? So I got my WCAG thing. And now what? It's the science of learning.
It's what we call the universal design for learning. In October of last year, Professor Regina Gilbert gave a talk about universal design, which is a concept that was coined by Ronald Mayes, an architect, to describe the concept of designing buildings and environments to be accessible to all people. And so Professor Gilbert took those principles, the seven principles of UDL, and applied it to web design. So how do you design a site or any kind of product with UD, universal design?
Universal design for learning is actually universal design plus the science of learning. And it's developed by the Center for Applied Special Technology, CAST. That's the acronym. It was developed in the '90s.
And that's the website, if anybody would like to look at it later on after the talk. So there's basically three principles in UDL. One is about engagement. The student engagement to the particular learning material or the classroom.
Representation. How are these materials being presented? How many different ways are they being presented? What kind of language is being used?
Are there any kinds of cues in order to translate certain things that are not translatable? And action and expression -- how do you take all this knowledge and formulate it and put it back into what we understand as learning and contribute back to the classroom? So this is great. This is wonderful.
This is learning about how people learn. How does that apply to accessibility? The Web Content Accessibility Guidelines or the Web Accessibility Initiative Group -- there are four principles to accessibility. Either the material or the tool has to be perceivable.
They have to be operable. Understandable and robust in any platform. Assistive technology, tablet, smartphones, laptop, screen readers. Everything.
So these things do relate to these three things. So let me take, for example, perceivable. One of the WCAG guidelines is that every video has to have a caption. And so we have a caption here.
And that is a form of representation of the material that I'm handing to you. So this is what I'm giving you, and this is what we're listening to, and this is what we're looking at in different modalities. Now, we look at operable in terms of, like, how do you operate a website? What is the navigation?
Is there a wayfinder? Is there a way for me to get to... Can I navigate through the website or the tool using keyboard? How does that work with screen readers?
Different kinds of screen readers. That is a form of action and expression. How do I get to my content? How do I get to my discussion board?
How do I access the text block that is at the bottom of the page, in order for me to write my thoughts? And that is -- they are very comparable. And so we see that -- and understandable, of course. The engagement.
How do you get the students to engage with the content of your course? How do we minimize distraction? How do we make things readable? How do we make things legible?
Is 14 point okay? 18 point better? How about if I put a red text on a blue background? Would it work for people with colorblindness?
What if I use, for example, a symbol in my course, red as danger, green as okay, and yellow somewhere in between? If I put that, do you think that will be a problem? Right. Because of colorblindness.
So they are comparable. In terms of where we put accessibility in UDL. Even better is UDL if you do it right and well. Accessibility will be there.
So it reduces the barrier to access the learning materials, or just education in general. So let's see. I'm gonna give you a few examples on how we apply these principles in the work that we do at the Center for Teaching and Learning at Columbia University. I'll see the mouse here.
Attaining higher ed is... Actually a massive open online course on the EDX platform. It is a course designed for people, student learners, who are in active duty, or who are veterans, who want to get back into higher ed. The course teaches -- guides the learner on how to find information, on what universities are more supportive of veterans, and where do they get support, what universities are close to them, what kind of services are available to them, how to apply financial aid, et cetera, et cetera.
So that is the course. And the companion side is... So it is not an undergraduate course. It is a little bit more open to the world.
But the principles that we used in order to build this particular site can be applied to a lot of the educational technologies that we build for undergrads. So the principles are the same. The solutions are different. But we learn from each project and we apply it across projects.
So here is a map. This is the companion site to the MOOC. So I'm going to go to the map. I'm gonna search.
Maps are tricky. Because maps are visual. It gives you a position relative to another position. And in this case, it's a visual representation of data.
This is how we approach it. And so the first question we ask is: What exactly do you want to do with this data? Not the map. The data.
You have a list of things you want to put on the map. Why do you want to put it on the map? So the question we ask is basically: The learning goals and the outcomes. What do you want to teach?
What do you expect the student to learn from this? And from there, we work together with the instructional designers, the faculty member who is in charge of the course, the developer who is actually building the backend, and the frontend developer or the frontend designer, to actually design on how it should look. And then if there's media, then we will work with the media team as well. And it's the same conversation.
Who are the people that you are trying to deliver this product or this information? Are there any limitations? Should we consider the limitations? Or we don't care, because we want to build for everybody?
So here there is... You can navigate by keyboard. So I'm gonna say I want to do a four-year program. And so you can see that the list of data that matches the filter is being displayed here.
And you see yourself in the map. Okay. Some of these universities are close to me and some are not. But this is a visual representation.
I mean, a screen reader reads it fine. But the screen reader will not read the map fine. So what do we do? There is a comparable listing.
Because it's just data. And so it's the same search interface. It's the same filter interface, that will narrow down the data, but we listed it, rather than actually having it positioned on the map. And this is printable, downloadable, which is great, because people can carry it with them.
And the other thing is also responsive. If I could shrink this... Bear with me. I can't see.
It's responsive. So a lot of people probably don't want to use this on the laptop. They may not have a laptop readily available. They might just want to search in on the web.
I mean on their smartphones. And we make sure that that happens. And this is... We do the checking with DQ's aXe, which is a plugin for an accessibility checker, and we feed it back to the faculty.
We say: Is this all right? Is this something that you think your students will benefit from? So this is a back and forth discussion between developer and faculty. We all sit in a room and talk about it.
And that's usually the best way to conduct any kind of development of software products. So that's the Attaining Higher Ed. Now, this one here is Project OHCOE. It stands for Oral Health Care in the Opioid Epidemic.
This is for the Dental School, up at Columbia Medical School. This is Professor Burt Edelstein, the faculty involved in this particular project, so what that basically is is to teach future dentists who are now enrolled at Columbia dental school on how to identify from their patient's teeth -- how to identify the patients who are at risk for opioid use. Now that you know who they are, what do you do? Where do you find help for them?
This particular curriculum, again, is designed not from the screen. It's designed from interaction. So I am a student. I am a dentist to be.
And this is the scenario that I'm going to be facing. So that was where we had the instructional designer, learning designer, talk together with the faculty. And the graduate student. Graduate student?
To figure out... What are we building? How does it work? Does this work on all platforms?
In classroom? Offclassroom? Anywhere? What are the other materials?
We have videos. Okay. Now we're gonna talk to my friend Alexis here. A video has to be accessible.
It has to have captioning. It has to have transcript. So how do we deal with these kinds of things? So now that we actually have the form of architecture, of what we're building, then the developers will have to know what kind of backend to build.
Because once you start building for accessibility, with accessibility in mind, if there's any changes that you need to do, the developer is already involved, it is a lot easier than having to do that retrospectively, or... So you're becoming more proactive from the very beginning. And that is actually... It sounds like it's a lot of work in the beginning.
But it saves a lot of time in the end. So that's the UDLcast website I was showing earlier. So going back to this slide... That's the opioid.
So this is the video I'm talking about here. We have a MOOC. Massive open online course. On inclusive teaching in the classrooms in higher ed.
These are two of our colleagues. Dr. Jundles and Dr. Patel are the faculty for this MOOC.
So the setup is very typical. You have people talking and you have their title. And then we turn on the caption. We can't read it, because the caption is over the content.
So when I got the invitation to give a talk tonight, Stacy Smith from Equal Entry said: Make sure that a quarter of your slides are empty, because we want to put the caption. So there! It's the same idea. You need a space for a certain thing.
But you will not know things like this happening until it happens to you. And it did happen to us. And then we realized that... Oh, okay.
So Alexis said... So next time, when we do videos, make sure that... Boom. There's a certain space for captioning.
And we don't have to worry about it, because we always have captioning. It is accessibility. You know? So there are a couple other things also.
Before you start shooting a video, you need a script. A script is a story. The story has a teaching and learning component related to it. And so you need to...
So we started to work with the two professors there. On how we apply the principles of universal design for learning in order to make the script a little bit more accessible. If I needed to describe something on the screen, I can't just point at things. Look at things over there.
So you might say... Okay. Look at the top right corner of the screen, et cetera, et cetera. The ball that I'm holding is blue in color.
And it's on my left hand or my right hand, and so on and so forth. And you see a lot of cooking instructions, they just put all the materials in the pan and you don't know what it is. And if you do have a clear script, and it helps even people... Everybody.
You know what it is. You know what they mean. It's very clear. And then...
So that's just a script. So again, revision of the script, even before we set up the camera. Then we start to set up the camera. The improvement of transcript with multiple people.
When you have multiple people talking, how do you represent who's talking and what? So that's another thing that Alexis encountered. That we actually have to have a way to represent these different people talking in the transcript, or even in the captioning.
If you go back to your previous slide, there's an example of that. You see on the left side. You put Dr. Patel...
ZARINA: Right, okay. So the person who is talking, Dr. Patel, is talking right now. And so the cue is right there that she is talking.
She's beginning to talk. As opposed to when it switches back to Dr. Jungels, then you see the captioning changes. So that's something that we need to start thinking about, when we're doing captioning.
And we're doing transcripts. So in the considerations for the purpose of B roll and animation, it's similar to any images or any kind of external material that you want to put on your website. If it is an image, does it serve a purpose? If it is a B-roll that just shows a picture of New York, does it have a purpose?
Or do you really want to say... It's in New York. Then you have to say it out loud. It's in New York.
So those are, like, three examples that I would like to show to you tonight. That we've developed by applying the principles of accessibility and universal design for learning. So the lessons that I've learned from the years that I have been working on these tools is that Accessibility Experience, and I'm gonna use it very deliberately right now, instead of compliance -- because it means compliance and learner experience, the processing of information, the understanding of information, the access to information...
They do take time. And I think if anybody here in the room have done it, we know that. It takes time. It takes a lot of work to understand how these frameworks work.
How the WCAG works. How a screen reader works. How PDF works. And accessibility should be integrated into the making of the process.
It cannot be and it should not be an appendage. It has to be proactive. It can't be reactive. And I'll tell you why later on.
And then also another thing that really helped is communities like this. Where you have a community around what I do. At Columbia, we have a community in accessibility where we do talk about what we've done, the problems we've solved, because not everything is just about compliance. It's about problem solving.
There are a lot of things like drag and drop -- it's a famous thing that people keep talking about. At least, two, three years ago. But how do you make it accessible? How do you solve drag and drop?
There is nothing in WCAG that tells you how to solve the problem of drag and drop. The problem is: How do you make an alternative to drag and drop? Is it even necessary to drag one term to the next term and make a matching exercise whatsoever? That's the question.
It's not about the interaction. The interaction comes after that question. So yes, it takes time. Yes, it's hard work.
Yes, it's mind bending and brain hurting. But the outcome, the end product that we develop, we find that it is so much more together, it's so much more complete, and doesn't feel like it's going to fall apart. And with that, it is a better experience for the students. For every student.
For any student. Because we have lowered the barrier to access to education. And there will be some need for assistive technology. But it's not...
We have lowered the barrier, and it's a little bit easier now. And I think that it is a work in progress. We're continuing to do things like the problem with the map. We have another project also with the map.
And we're trying to figure out... How do we present data on a map with relations to each other? Not on a map? It's a hard problem.
But it's an exciting problem, I think. So this is basically what I have on my wall. You can ask my colleagues. I can't take credit for this.
I saw it in a tweet. Now I can't remember the person who tweeted it. So ASPIRE is an acronym for anything that we build - accessible, secure, performant, inclusive, responsive, and ethical. That's that last part.
I think is the most important. POUR is the WCAG principles. It's perceivable, operable, understandable, and robust. So I have this on my wall, as "I ASPIRE to make POUR choices".
With a positive connotation. Not negative. So I mentioned a couple of things in there. The Web Accessibility Initiative is a group that came out with WCAG.
And does anyone here not know what WCAG is? Everybody knows? The WCAG? Okay, good.
So the universal design is from the source itself. Which is CAST.org. The UDLguidelines.cast.org. And DO-IT, which is a site of resources by the University of Washington, is a very good site, because it gathers information and resources -- particularly for the purpose and use of higher ed.
Not that you can't use it for not higher ed. But it's a different set of culture that we need to address quite specifically. Thank you. If there are any questions...
(applause)
CAMERON: Thank you, Zarina. As Zarina made the observation, we're gonna take Q and A. I have a mic here. And again, because we have the captions remotely, if you have a question, please raise your hand, and I'll bring you the mic.
Don't just holler out. Please.
Hi. Thank you so much. I had a question. So you showed how you improved the video that had the lower third too low.
And on the right was the corrected version. Can you think of any other examples like this? Something that you learned from an experience that happened and had to be improved later?
ZARINA: Sure, definitely. I'm gonna refer that to my colleague here, Alexis, because she is the person who has the experience, in terms of accessibility and video.
Thanks, Zarina. And great job. Your talk was awesome. I learned a lot.
Is your question in reference specifically to video? Or generally? Okay. So I come from a video background.
So I'll answer in that context, and maybe if Zarina might do a development context. So I would also say that this one on the right is not actually... One of the things we learned is the one on the right is what is called open captions, or burned-in captions. WURPBD in during post.
The ones on the left, the captions on the left, are called closed captions. And that's what you get when you maybe submit to a professional captioning service like at Columbia we use rev.com, but there are a lot of other different services out there that have benefits and disadvantages. The difference between those two things that we learned from doing this MOOC experience is that closed captions -- they provide you with a transcript that you can then make downloadable.
We use the platform EdX and Coursera to host a lot of our MOOCs. And those platforms allow you to have closed captioning, which is the captioning that shows up in a downloadable transcript. But we ran into the issue of... What if we're promoting on Twitter or Instagram or things like that?
They don't have those available yet. So what we started doing is whenever we wanted to put something on Twitter, we decided we were gonna put in burned in captions. This is much harder if you have a very long video. But if it's for social media, it's like a minute long, two minutes long, it's much more doable for a member of our team to just go into our editing program and then add those captions in there.
So that's one thing I would say that we learned.
ZARINA: Does that answer your question? Okay, good.
You said you used open captions for social media? Okay. So the burned in captions were only for the social media stuff?
Yeah. Let me unpack it just a little bit more. In case, you know, I jargon about it all the time. So basically the difference between something being burned in -- it would be...
It's not fully accessible. Let's just say that. It's not fully accessible if it's burned in. This is our way of making it a little bit more accessible to somebody.
Because if you're on Twitter, and you're looking, you still have to be able to see it. There's not any sort of text that a computer can read or that a screen reader would be able to read. And so we chose to try to make the best out of a situation that I think these large companies like Twitter are going to get on the accessibility train pretty quickly. As quick as they can.
Hopefully. That's what you really hope. But this was a workaround that we found. Is that by putting these burned in open captions in our social media videos, that at least people who can read and who don't need a screen reader -- if they're deaf, they can still understand what's happening in the video.
ZARINA: By the way, captioning and transcription, according to the WCAG 2.1, if you have one or the other, then the video is considered not accessible. You kind of need to have both. And it is a big issue. In 2015, EDX, MIT, and Harvard faced a class action lawsuit filed by the National Association of the Deaf, because their captioning -- there were no transcripts at that time -- the captioning was inaccurate or unintelligible.
And EDX settled for that. And so the EDX platform has improved dramatically. But there is another lawsuit also for Harvard by also the National Association for the Deaf. It's that the content of all the videos that they have produced do not have captioning.
Same issue. And in November of last year, they settled. Harvard settled. And so in March...
No, May of last year, Harvard put out a very comprehensive -- and very good, actually, digital accessibility policy for the university, in terms of their content. In anything that the university produced. Lawsuits are terrible, but they're kind of great at the same time. Accessibility should not be...
Personally, I think that accessibility should not be associated with punitive measures. Accessibility should be about empathy and facility, and not fear and liability. If we keep talking about lawsuits in order to frame accessibility, we're gonna have resistance. It is for everyone.
It is for everyone. There's no harm in doing it. It is just a little bit more work. But it's okay.
JOLY: I'll follow up on that with... So last year in July, I did the partnership for the progress on the digital divide conference at George Washington University. And because of the lawsuit, they said: You can't livestream this unless you have captions. And we got into a whole thing of like...
We had to find out what level of captions were actually acceptable, that we would be permitted. And nobody seemed to know where you drew the line, how accurate the captions had to be. It was something that was... You know, in the end, we sort of went away and fortunately Vint Cerf and Google coughed up the money so we were able to live caption the whole event.
But coming back to your point about burned in captions, with this Meetup, you know, we burn in the captions on the live stream. And then we make a transcript and put it afterwards, and we were uploading that transcript to YouTube and YouTube would sync it. So then what you'd have is on YouTube, when you went back to watch it, you would have two sets of captions, where you would have one set of captions overlaying the other, which is not good.
So only since the last two Meetups, I've changed, so I now make a recording without the captions and upload that to YouTube, and then upload the transcript, so YouTube has closed captions afterwards. And you want to have those closed captions, because people can configure those captions however they like. They can make them as big or whatever, as readable as they like. They can autotranslate them into other languages.
So that's a thing. But the live closed captions is a difficult thing to achieve. YouTube does do it a little bit now. But...
That's not so easy to get.
ZARINA: She has a question.
While we're on the topic of video, you were saying that captions and transcripts are both required by WCAG, but another thing that's also required by WCAG is descriptions. So captions and transcripts are both for people who can't hear the content. So I'm wondering if you've already started doing work for people who can't see the content as well for these videos.
ZARINA: Definitely. Part of the writing of the script that we were talking about earlier -- it's something that came to us only after we started seeing it. It's funny how like -- when you start working on accessibility, when you start using assistive technology, again, empathy is about being in somebody else's shoes. When you start looking like...
I can't tell what this person is doing... Just by reading the transcript. And so one example is the opening title for... So in the MOOC, we have different videos for different modules.
And in the beginning of the module is a title card. What this video is about. What the title is. So traditionally, it's just a card.
So it says there: You know, this module is module 4. It's about accessibility. In the classroom. But if you don't have an access to that visual, you don't know what this module is about.
So the solution that we came up -- after doing that whole reel -- is to come back and actually have someone read module 4. Accessibility. So that's, like, a very small example. But we realized that it is important to actually have it there.
And we started looking into, like, when there are graphics in there, like the graphics of a glossary or... So a glossary is just a bunch of text. It's an image. But then the professor who is being interviewed is talking over it.
So how do we do this? And then we ended up making the decision saying like... The professor is actually explaining that particular term, so there is no need to describe what is being written on the card. So it is a case by case, as we go along.
And we are aware that there are situations like that, and we're becoming more and more aware every time you start making new videos and new MOOCs and new tools... You know, new promotional video, or... Et cetera, et cetera, for any kind of platform that we're using in a classroom or social media.
Can I just add really quickly to the end of that? I just wanted to add really quickly to the end of that. Because we also had -- to your question, ma'am -- is we had a talk the other day too with the head of -- one of our bosses. About...
Sometimes, right, if you have a lot of videos, it's hard to pinpoint what the titles are gonna be, because the faculty might not be able to make up their mind right away. And so it really messes with the workflow. And so one accessible workaround that we came up with is: You can go in... Usually what the captions are output as is either an SRT file or a VTT file, and you can go in and edit those.
And so it does take about... It takes time to go and voice record every title. But by doing that and uploading your video for captioning, it will automatically, as Zarina said, make that description. It'll be in the title.
So they'll know they're in module 4, section 2, accessibility practices, yadda-yadda. But that takes a lot of time. And time is money. And so one thing is that you can go in and you can edit those titles yourself.
So people will have the visual on the video, but it will show up in the closed captioning. And therefore it will be accessible via the transcript. That's one quicker workaround that we very recently talked about.
ZARINA: Yeah, it's very case by case basis.
One more question after this.
So I was wondering if you would be able to talk a little bit about ownership in terms of applying some of the UDL principles to curriculum. Because for me... I'm an upcoming educator, and I'm familiar with the UDL principles, but I think for me, the idea of figuring out and actually producing a lot of multiple means for my materials and assignments is a really daunting process. And I was wondering if you could talk a little bit more about the ownership of that.
Does that fall on the instructor? Does that fall on, like, a technology team like yourself? Or does that fall on something like the Office of Disability and Inclusion? Or is it some sort of collaborative process between all three?
ZARINA: I think for us, it's more collaborative. Because again, if we're not exposed to just accessibility in general or Universal Design for Learning in general, it doesn't cross your mind. I gave this presentation in front of my team. Earlier today.
And I made a lot of assumptions. Just because I don't remember or it's something I'm not aware of. But through collaboration with my team, I started to make changes. And that's the issue in terms of...
Advocacy of accessibility of content, that the faculty will put up on any kind of learning management system. We need to start introducing these concepts of making it accessible or making it more clear. But UDL doesn't talk about accessibility, because I know when people mention accessibility, and everybody is like... That's another thing that we have to worry about...
But you could couch accessibility into UDL, and just like -- this is the teaching practice. This is how to make people understand better. And this is how to... Like, I don't understand this concept.
It is a collaboration typically in our department. It's faculty and the learning designers, instructional designers, who help faculty look at their syllabus, and how they put together a syllabus in the class, and whichever ones fit. The blended learning or web augmented classes. But you can't do it alone, really.
It's a little hard. Not a little hard. It's hard. I'm not going into...
It is hard, but not in a negative way. It's just... Anything is hard. I don't know how to ride a bike, so it's hard for me.
But, like, it's doable. It's just... Little by little, small practices, and then you start... Once you do the small practices, you fold it back into your process, it becomes a habit.
And then you do a little bit more. You fold it back into your process. It's a cycle. It's an iterative cycle for ourselves.
For yourself. And then this is where the community sometimes is useful. Because we bounce ideas with each other. And how do we make it better?
We're not going to do this alone. That's impossible to do. Definitely. I hope you will find a community that will fit what you need.
Oh, we do have a MOOC. Right there! What is it called?
Inclusive teaching. We have to Google it.
Inclusive teaching. Supporting all students...
ZARINA: So this is an inclusive teaching -- this is EDX. It is... Freeeee? Yes, it is free.
There it is. Actually the only one.
ZARINA: So if you go to EDX.org, and you search for "inclusive teaching", that is the course that talks about... Well, there are many other things about inclusive teaching. But one of them is: Like... UDL is a module...
Module 4? But there are other parts in this course that are not UDL. But it's actually part of UDL. You just don't know it.
It's broken down and being analyzed differently. So this is free. It's a six-week course. And yeah.
We worked hard for it. And it's available to anybody. You don't have to... If you are enrolled...
It says there "free". But if you want a certificate, then it's an EDX model that you will actually have to sign up for.
I just wanted to add on to that. Sort of -- the particular sort of situation at Columbia. A large pipeline of our work is through an RFP process with the provost's office. And so we look at faculty and projects that are sort of exemplars of...
Faculty that are really engaged, who want to put in the time and effort. And I find that my experience, working with these faculty, once you have that interaction, you go through a whole project with them. That's something that -- they've learned a lot from us, and continued forward, and I've seen faculty who have offices next to each other, and the ideas seem to seep through the walls. So yeah.
Columbia is a big place. But we found that in these incremental steps, we can start making a change. One last question.
This is a quick question. You mentioned something called DQ's aXe? It's an accessible plugin?
ZARINA: Yes. So the company... Well, DeQue. That's the company.
I tweeted them. I asked them: How do you pronounce your company's name? Is it "deck"? Or "deek"?
They said "it's DQ, like Dairy Queen." So it's DeQue. And they offer services -- this is actually in competition with Level Access, right?
It's okay.
ZARINA: It's all good. And DeQue is... Let's see. Good grief.
The mouse. So the tools... Overview... So yeah, aXe is the tool, aXe is also the engine behind Lighthouse, which is the tool that Google used to check the compliance of your website.
Lighthouse gives out a report. I think it's still using DeQue aXe. And Microsoft accessibility also uses DeQue's aXe. And there's a lot of -- if you're a developer, there are a lot of modules that have been written in GitHub about integration of accessibility checking.
When you're doing your development. Continuous integration. It checks as you push. Which is great.
Because I don't have to do it line by line.
CAMERON: Okay. Well, I wanted to just highlight your last note on community. And collaboration. I think this is an excellent example of people with amazing expertise and experience, contributing back to the community.
So I wanted to say thank you again to Zarina and her colleagues. So thank you so much. (applause)