Podcast: Data Visualization Accessibility for Employees, Teams, & Leaders
Future of Work Podcast, Episode 17
Doug Schepers of Fizz Studio discusses sonification and other innovative methods of making charts, graphs, and maps more interactive and accessible to all users, including people with disabilities.
This podcast is developed in partnership with Workology.com as part of PEAT's Future of Work series, which works to start conversations around how emerging workplace technology trends are impacting people with disabilities.
Intro: [00:00:00] Welcome to the workology podcast a podcast for the disruptive workplace leader. Join host Jessica Miller Merrill founder of work ology dot com as she sits down and gets to the bottom of trends tools and case studies for the business leader H.R. and recruiting professional who is tired of the status quo. Now here's Jessica with this episode of work ology.
Jessica: [00:00:25] I'm not ashamed to admit that I'm a data nerd and I absolutely love charts and graphs. They're a great way to analyze and assess the performance of a client recruitment campaign or in evaluating ecologies on digital performance and downloads Data is everything in business which is why it shouldn't be a surprise that accessible data visualization is also something we should be considering as businesses make strides in our workplaces as well as digital resources become more accessible. This podcast is sponsored by clear company and is part of our future work series in partnership with PEAT. The partnership unemployment and accessible technology. So we're going to be talking about data visualization and how to make it accessible. Today I'm joined by Doug Schaeffer's. He is the chief technologist at this studio which is a consultancy for creating a better way to create and explore information graphics. They create custom solutions for making data visualization accessible to all users including people with disabilities. Doug welcome to the work ology podcast.
Doug: [00:01:32] Hi Jessica. Thanks for having me. Can you tell us a little bit about your background. Sure. I start programming when I was twelve. I think on a TRW 80 and a Commodore 64 and then years later I picked it up again in the mid 90s when I realized that you didn't need formal training or math skills to be successful as a programmer. And when I was at a financial planning company I started working on my own charging library using SPG scalable vector graphics. This web vector graphics format that had just emerged and it from there I started a consultancy with a few other folks and then I joined a startup and then I got hired at W3 seed which is the Web store standards organization that defines team L and SPG and also look AG the accessibility guidelines. And so I went there for about a decade and I collaborated there with a lot of people who themselves had disabilities or who were accessibility experts. And when I was working on standards there that's when I got this idea for how we can make data visualizations accessible. And I started writing up standards for how to do that. And it wasn't really going anywhere. And I realized I was trying to push a rope. In other words standards is really driven by implementations by software. And I realized that if I wanted to make the succeed I couldn't just write a specification for how it should work. I need to write the software itself. And I've gotten more and more into that and I realized to help that there's a real need here. So I created a startup studio. And that's exactly what we do or we're aiming at making data visualizations accessible to everyone.
Jessica: [00:03:21] Awesome Well let's talk a little bit about the growth and demand for accessible data visualization. How important is this to business strategy and how does it overlap with where technology is headed in general.
Doug: [00:03:35] Yeah that's a great question. So. The rate of data collection is basically exponential it's doubling every two years. And what does that mean. It means that 90 percent of the data that's in the world today was generated over the last two years alone and that same trend has been going on for the last 20 years. There's just an enormous amount of data to deal with as a society right. How do we use that data. How we see patterns and trends in it. And one of the ways that we do that is we use data visualizations charts and graphs mostly bar charts line charts pie charts but also maps scatter plots parallel points. Many other chart types in. We use this to see these trends that we couldn't perceive in the data otherwise. You can see a trend much more easily in a line chart than you can in a spreadsheet. Some people have had this misperception that you can't make charts accessible but in fact it's just the opposite data visualizations are accessibility technology. Their cognitive accessibility technology they help us leverage one of our senses in this case site to enhance the limitations of our minds. Limitations in terms of memory in terms of pattern recognition and all sorts of other things. Basically data visualizations are inherently accessible more accessible for sighted people than data tables are. And so we just need to think about. OK so if we got this one since how can we use other senses to to also improve accessibility of this data it also going along with that improved information overload that we have.
Doug: [00:05:24] There's attention overload. It's difficult to get people's attention and to keep it long enough to impart a lasting message something that they understand quickly and that they will carry with them so eye catching charts and graphs are they provide a sort of novelty to help people meet that meet that goal. And so a lot of people a lot of businesses are increasingly using data visualizations for these reasons among others. And from a business perspective you can look at it two ways. One is how does that limit my own business. If if my charts aren't accessible how does it limit my own business in terms of meeting needs of my customers or my employees. What money am I leaving on the floor. What what workers Am I missing out on hiring. Keeping in mind that half of all blind people in the United States are unemployed because partly because they don't have the right tools at work to be able to perform the tasks they need to be able to perform. The second thing that you need to think about is what risks. In my opening my business up to by not making my content accessible so instead just people with disabilities that benefit from these charts we can touch on that later.
Jessica: [00:06:38] So what's involved in making data accessible. Can you kind of walk us through the process and what what it looks like.
Doug: [00:06:45] Sure. First off people aren't making their charts by hand anymore right. They used to they used to draw a map on graph paperwork graph paper with a ruler but nobody is doing that anymore. They're using software to make data visualization. So for the content creator or the developer the software developer the process is no different for making accessible charts than for making inaccessible charts. You're already using software. Just use software that makes charts accessible by default like. Ah the product we make is fizz charts great charts that are born accessible used. Add the data something but select the chart type you want select the colors style preferences and so on and the result is simply accessible outside of the box. No need for the user to do anything extra. I mean picking colors. Yeah. You do need to be careful about the colors you pick for color blindness but we even guide you through that process. As far as the technology aspect how do we do this. From a technology perspective first we embed the data directly into the file that's typically SPG on the Web. But you could theoretically embed the data in a P and G or JPEG file as metadata and this metadata doesn't only have to include the raw data but also like a textual descriptions that you want summaries human generated or auto generated.
Doug: [00:08:15] This technique by the way we call this deep graphics it's the technique of embedding the extended content as metadata it's deep graphics is a way of preserving context. It also improves your search engine optimization or SEO. It makes your data reusable people can extract the data from one chart and the data from another chart and add them together to make a new chart anyway. So once this data is in a file in a structured way the user can just navigate through each individual data point find out step by step what day by day whatever data point by data point what the values are they can explore the charter as much as they'd like but on top of that we extract that data and we add additional context like the ability to quickly and easily compared to different bars. We use natural language generation that create sentences and options and text and those can be voiced by screen meters or rendered in Braille readers.
Doug: [00:09:13] So from the enthusiast perspective making tracks accessible is is just sort of following the instructions following following your nose to navigate around the chart answers some options or select some options. And it's really intuitive experience for anybody who uses excessively technology like a screen there.
Jessica: [00:09:32] Can you share a case study or example of how accessible data visualization has maybe helped one of your clients.
Doug: [00:09:38] Sure. So one of the one of my clients for example they run excessively testing service and that's sort of it's a site that helps other people make their website accessible. It's a little bit meta so they use our charts to present the numbers of the checks that were performed on the site. They have a little dashboard that explains to you it shows you how many of your web pages were checked how many passed the rate of errors is it hopefully decreasing etc. So they wanted their number of users to have not only access to the day to day numbers like the individual numbers but they went to find out the trend as well. So you know is your site getting better or is it getting worse. This customer did that for their users and for credibility they need to do more than just check the bar for it definitely. So they used our charts and provides a much better experience than just a data table and there are lots of trying solutions out there that are starting to emerge. Any business should be able to find a charging package that does things excessively. When you talk about accessibility you should did a whole bunch of different considerations. It's not just blind people it's low vision people and SPG format we use is really great for that because it lets people zoom in but you still want to have the context of what you're zooming in on. You know just zooming in on a dot doesn't tell you anything you need to know the values you need to know the range and set and so on. So it's cognitive accessibility as well. There's a whole bunch of other types of accessibility that are beyond just blindness and so we try to take care take all of those into consideration.
Jessica: [00:11:12] Working with PEAT has been such an eye opener. As far as how or as much as how far we have to go when it comes to accessibility in our workplaces. In addition to helping people with low vision how can we take these complex charts graphs and images and make them easier for people who have cognitive disabilities.
Doug: [00:11:32] That's a great question. First off we all are sometimes challenged with cognitive limitations especially with the more stressed or busy or distracted. We're in a hurry or whatever data visualization is great for that because it can help communicate complex ideas quickly and there are times when you definitely need to use complex charts and images when the data are the relationships between factors are really complex like in some scientific papers you need to represent them accurately and that's also the case when you're doing exploratory data visualization when you're looking for patterns as opposed to explanatory data visualization where you're telling other people a story revealing to them a pattern you've already discovered most of the time people are doing explanatory data visualization. And when you're doing it you're better off refining the data to its to its core message and minimizing the complexity of the chart. I've seen some charts out there that are just you know I do data visualization for a living. And I sit there and look at that chart. I'm lost. I don't know what you're trying to tell me. And so oftentimes they're just adding too many extra details in there that I don't need to know in order to get the data that I'm in order to get the information. So I have a little mantra.
Doug: [00:12:50] You should when you're doing data visualization you should think of it as writing a poem not a novel. You need to communicate more complex ideas. You should consider representing it as multiple related simpler charts like a dashboard. Once you've done that once you've boiled the idea down to its essence you're probably meeting the needs of a broader audience including a lot of different types of cognitive disability. There are other things you can do for cognitive disability. For one thing data visualizations for some people are much easier to understand and for other people are very very difficult to understand they don't understand that the relationship of the visual to the visual to the data and by providing that the option to have it either as a textual description where you're describing what's going on the chart or a visual or a combination of both which is what our charts offer you let somebody figure out the modality that works for them rather than dictating one modality in particular cognitive disabilities a very broad subject and it's getting a lot more attention lately but I think we're just in the forefront of figuring out where data visualizations can help or where we need to augment the data visualizations help with cognitive disability. But it's definitely an area of increased interest.
Jessica: [00:14:10] I was just thinking about a podcast interview that I that I have where I'm talking with a recruiting and workplace leader who is really passionate about dyslexia and helping managers and employees be able to work together or you can work with peers who have dyslexia and get their brain thinks and how they process information is very different than than somebody who doesn't have dyslexia.
Doug: [00:14:35] Yeah I actually have a nephew who has profound dyslexia his dyslexia takes the form of my sister his mom told me was just she was floored when she was talking to him as he is a kid he said How do you get them to stop moving. For him it wasn't just the BP D Q thing right. The letters were actually moving around on the page and sometimes off the page. And so for him it was very difficult for him to do to find a way for it. He's a very intelligent guy and he ended up. Now he's a video editor in Hollywood.
Jessica: [00:15:17] That's exactly almost the same thing that Will talked about in the interview. So I think that it's important to be aware of our how our different abilities and disabilities and maybe how we learn and prefer to receive information.
Break: [00:15:38] Let's take a reset. This is Jessica Miller Merrill. And you're listening to the work ology podcast. Today we are talking with dog shoppers about accessible data visualization. This podcast is sponsored by Clear company and it is part of our future work series in partnership with the partnership on employment and accessible technology or PEAT podcast support provided by Claire company Claire company. It's a complete talent management platform used by human resources professionals to help them hire retain and engage players within a clear company dot.com to sign up for a free personalized demo.
Jessica: [00:16:16] I want to take a step back a bit and give some people some background. You mentioned this at the beginning of the interview about the standards and guidelines for work back and we've had Dan Ellerman who is the inclusion and diversity senior manager at Accenture on the work policy podcast previously and I'm going to link to dance interview and I think that Dan gave us a nice overview of the Web Content Accessibility Guidelines. Can you give us maybe some more background on that and how it covers data visualization.
Doug: [00:16:49] So right now look CAG two point one which is the most recent version it doesn't explicitly define any checkpoints around data visualizations separate from images in other words it doesn't have specific guidance that they have they have one requirement that's applied to all non text content It's section one point one point one non text content in what CAG to the point one and it just says provide a text alternative. Complex images they do go into a little more detail on some of their supporting documentation. They suggest that you for a data visualization you give a short description by default and then give an optional longer depth description. And then you provide like a data table giving access to the raw data. I think that that's lacking. Honestly I just don't feel it's good enough for a couple of reasons. One data table just isn't as accessible as a chart. We touched on this before. It's not the same experience to have a big spreadsheet of populations or whatever versus looking at a line chart and being able to just see trends right away doesn't give the same afford answers doesn't reveal the data in an equivalent way and it doesn't give the user the ability to perform tasks nearly as quickly or efficiently as a sighted user. There's a lot of cognitive overhead that goes along with that kind of description in the two textual descriptions textual summaries are not structured they're not exploring all data visualizations or kind of language they haven't nuance and expressively that you cannot capture with a sentence or even a paragraph of text so reflect this.
Doug: [00:18:35] I think that what CAG should add some additional success criteria for data visualizations apart from other image types and I think some people there are receptive to that. I think one of the barriers to that in the past has been that it is perceived that data visualizations can't be made accessible in my company and several other companies have showed that yes infected visualizations can be accessible. So there are some other things going on. So as you're working towards this goal I'm chairing a committee in the Bennett tech diagram center. It's aimed at defining how to make data visualization accessible in a more formal way. And we do plan to bring that to work to deputy we when we feel it's matured like we've got a good solid project and I'm all I'm I'm working on a chapter for a diagram Center report right now aimed at describing accessible data visualizations for teachers parents and people with disabilities serve in an education context and that should be published in the next few months. So people who are interested go to Bennett text diagram center. Could you can provide a link and they should look for more information there. But apart from that apart from all of that the ADA a Section 5 or 4 Section 5 or eight other legal requirements they're not strictly tied to work. AG There's a lot there's a large degree of interpretation there so this misperception that databases can't be made accessible and therefore being it's been given a free pass a free ride by even by some accessibility auditors that's shifting. Those of us making accessible charting software changing that perception and raising that bar so pretty soon formal standards or not auditors like accessibility auditors and courts are gonna start requiring that chart to be made accessible.
Doug: [00:20:28] So it's a good idea for companies to get ahead of that.
Jessica: [00:20:30] In our prep call and then the previous question you mentioned data visualization as a language. Can you talk a little bit more about what you mean by this.
Doug: [00:20:40] Yes sure. So there's a structure to Data visualization has got a distinct visual elements that have a specific meaning within a context that you can but can combine them together to communicate meaning like for example the height of a bar in a bar chart indicates the relative value within the range indicated by the axes. So that basic principle carries over to stacked bar charts to group bar charts the gadgets and so on. So just like we weren't born with natural language we're not born knowing how to read a chart or to create a chair for that matter. We have to be taught this. This is something it's a skill that we learn. But once we understand how to map values like size position color shape to two numerical values or to categorical values then we can figure out how to apply that to chart types. We've never even seen before because we rely on these visual these familiar elements in the data database this field it's common to say the data visualization field is common to say that data visualization. Is a story about data. I like to think that data visualization is a user interface to data. Each type of chart is specialized for accomplishing particular tasks. And so we look on our job as providing affordable sources that worked for multiple since is not just vision to provide equivalent ways to accomplish those tasks reach different chart.
Jessica: [00:22:03] One technology that is really changing things lately has been voice assistance like Siri or Alexa and I personally am increasingly relying on these to help me in everyday life. From adding calendar invites listening to my news and even voices tax hike. Alexa turned on my TV for me yesterday which I finally figured out how to do that and I was super excited kid.
Jessica: [00:22:27] Can you talk about how these voice assistance are impacting data visualization technology for both people with and without disabilities.
Doug: [00:22:37] That's a great question. Everything all is new again. We we think of voice interfaces as being this amazing new thing that never existed before. But it's really computers have finally caught up to so the primary thing that humans do which is which is to communicate information which is is voice is communication is conversation. So it's speech and conversation is the oldest method of communicating for Richard's older than touch interfaces it's over the keyboard and mouse. It's older than reading and writing. It only seems new to us because it's actually hard for computers to do it because our brains are really specialized at this. It seems easy to us because we're sort of built to do that and because we're built to do that it's really natural for us to use these conversational interfaces. I just think it's funny that you can even think of conversations that humans as conversational interfaces and I studied linguistics as a college so it's it's it's an interesting topic for me anyway all the biggest tech companies know this and they are pushing speech interfaces hard they all want us to reduce barriers for us to use their services right. So it's going to happen more and more and of course keep in mind that the technology aspect of this all has its roots in accessibility technology right. This is this is screen readers writ large screen readers where for not for sighted people I know a lot of people with disabilities and who think smart speakers have been a huge convenience for them be able to perform tasks control their environment without having to get up to a wall switch and get on their smartphone to navigate. It's made a big difference in their quality of life and many of them are already using Texas speech or voice interfaces of course. So it's a logical next step for us to provide a supplementary or even replacement interface for data visualizations. You think about the number of people who are trying to get information on the web or trying to information in general. And if they encounter say your Web site how what's their what's their experience of that on a smart speaker on using Siri. What's their experience of the information that you're providing on something on a device that either doesn't have a screen or they're not using the screen so or where the screen is really tiny like a smartwatch right. You want to provide to them some other way of doing this. Getting at your data people are already asking you Are you asking your smart speaker what's the weather gonna be like when you're getting ready for your day. So for many people would be really convenient to ask a smart smart speaker what are the sales figures for the week or compare this quarter's earnings to the same quarter last year while they're driving or getting dressed or exercising or in the shower or whatever. I don't think that speech interfaces these smart speakers are going to replace podcasts on your way to work but if you're rushing on your way to a meeting and you want to make sure you have your facts straight being able to just say to your to your device Hey give me this give me these figures give me this information that could be a real without having to look at the screen. That could be a real benefit for the user so we are we're looking at basically making smart speaker apps for your data as well.
Jessica: [00:26:06] I love that because I think that that's the world we're in right now. Like if I want to know recipe conversions or if I want to ask something really specific I would love to ask my website Hey tell me about what the definition of accessibility is. According to the work ology podcast I would love for it just to pull it up and say well this is what Doug said this means maybe that yeah that's where we're headed and I I really hope so.
Doug: [00:26:35] I think that you really touched on an interesting point there which is. It's not just that it's not a passive thing right. It's the smart speakers and not just passively listening. They're not audio books right. They're interactive interfaces. Right. You want to be able to basically make your information available to people in a way that they can extract it so that computer can extract it and present it to them in the way that they're asking for it to be presented. Right. And that's that's right at the core of speech interfaces and that's right at the core. Ultimately what we're doing with these charts right. Getting specific information answering specific questions not just pushing information at a person and details because when I think about all tagging right.
Jessica: [00:27:21] Like I can write information into the alt tag from an image that I put on my Web site. But it doesn't really tell the whole story and that's what I feel like you're trying to do is help tell the whole story and in different ways.
Doug: [00:27:35] If I had for example something that had all 50 states a bar chart that in all 50 states maybe I'm trying to tell you maybe I'm trying to compare I don't know New York and California. Right. And so somebody is. So I'm looking at a chart and I see New York in California and those two bars are colored I don't know orange in the wrestling Baja code blue whatever they're drawing attention to those two bars. But I've spent most of my life in Missouri and North Carolina. So I am probably more interested in those two states. So if the all text just talked about comparing California and New York that wouldn't be serving my particular interests. Right. So with with a with a bar chart right. I can personally go in and just look at New York of Missouri and California and North Carolina and get that information for myself. And so all Texas is not enough. You might not provide all the information and so that's what I try to do it lets them get access to the information that they want to get that chart not just information that you're providing.
Jessica: [00:28:42] I love it. Well Doug where can people go to learn more about you and the work that you're doing so they can always go to our Web site.
Doug: [00:28:49] We're revamping the site right now to put some more content links to videos conference presentations and so on. But in the meantime you can request the demo from our site. It's his studio with his dot studio and we're pretty new startup. So we've been concentrating on the software but not the site but that's all coming soon and you can always request the demo. I'd love to show a live demo of of what we do. I have a personal Twitter account at Shepazu.
Doug: [00:29:16] This s h e p a z u where I talk about tech and accessibility and databases but I also talk about politics and not business topics so. So follow me at your own risk and you can also go to the benefit Web site where we're providing some more information about this and hopefully soon you can. You can go to Debbie through C and find out more about it but that's maybe something for the future.
Jessica: [00:29:44] We'll link to Benatech and and hopefully that section I have a podcast interview with Jim. Also I've been attacked so it's a it's an ongoing topic and conversation and I appreciate your help in these areas.
Doug: [00:29:59] Thank you so much for having me.
Jessica: [00:30:01] This is a fascinating discussion and one that I think we are going to see a lot more of as our use of data to drive business decisions continues to grow. I'm also interested in seeing in the coming months and years how audio visual assistance like Siri Google Home and Alexa will continue to push forward how we use and add audio in our lives. You know I use audio for so many different things audio books I listen to the news. I send my text message via audio. It's such a great way to kind of like a productivity hack but I can just get things done. Certainly I'm invested in the portions of audio as someone who now has two different podcasts. The one you're listening to and a second we launched earlier this year called Work ology go podcast which is a five minute two times a week daily news podcast highlighting a single story in the H.R. recruiting community. Whatever happens I'm interested to see the rise of audio and how data certification and visualization will play into this. This Future of Work series is in partnership with PEAT and it is one of my favorites. Thank you to PEAT as well as our podcast sponsor clear company.
Closing: [00:31:14] The workology podcast Future of Work series is supported by PEAT the partnership on employment and accessible technology PEAT's initiative is to foster collaboration and action around accessible technology in the workplace. PEAT is funded by the U.S. Department of Labor's Office of Disability Employment Policy. O DEP. Learn more about PEAT and PEATworks Dot org. That's P E A T W O R.K S dot org.