Document Accessibility Unraveled
Chad Chelius and Dax Castro during an Accessibility Podcast with the Chax Chat Logo between them.

Insights from Adobe on improved mobile accessibility

Accessibility Podcast Topic Links

Accessibility Podcast Transcript

Dax Castro
Welcome to another episode of Chax Chat. Join Chad Chelius and me Dax Castro, where each week we wax poetic about document accessibility topics, tips, and the struggle of remediation and compliance. So sit back, grab your favorite mug of whatever, and let’s get started.

Chad Chelius 
Welcome, everyone. Today’s podcast is sponsored by AbleDocs, makers of axesWord, axesPDF, document remediation, as well as website accessibility, auditing and monitoring services. So we want to thank them for being our sponsor once again on today’s podcast. My name is Chad Chelius. I’m an Adobe Certified Instructor and Accessible Document Specialist, as well as a consultant.

Dax Castro 
And my name is Dax Castro. I am an Adobe-Certified PDF Accessibility Trainer, as well as certified as an Accessible Document Specialist by the International Association of Accessibility Professionals.

Chad Chelius 
So Dax, I’m really excited to be here for another podcast. And I’m even more excited that we have an esteemed guest on our show today.

Dax Castro 
Awesome.

Chad Chelius 
We have Matthew Hardy from Adobe. And Matthew is a Director of Engineering in Document Cloud with the particular responsibility for liquid mode in Acrobat mobile.

Dax Castro 
That’s right. We had Leonard Rosenthol just a couple of podcasts episodes ago. And so for those of you who aren’t familiar with liquid mode, you can go listen to that podcast, as well as this one. So awesome.

Chad Chelius 
So Matthew, how are you doing today?

Matthew Hardy 
I’m doing good. Thank you excited to be here.

Liquid Mode for Adobe Acrobat Accessible PDFs

Dax Castro 
Well, we’re glad to have you absolutely. Liquid mode is such a great tool. I knew it had been coming out for quite a while. And I just wasn’t aware it had been implemented. When Chad told me, “Hey Dax, liquid mode is in there.” I was like, “What!” And so I went and tried it. And I was just blown away how great it works on quite a few different documents.

Matthew Hardy 
I’m glad to hear that. Yeah, liquid mode shipped, actually, about a year and a quarter ago now. Officially, although, we had various phases of beta leading up to that. And we’ve been really excited to release that on both our Adobe Acrobat mobile products for iOS and Android. and get that out to our customers.

Chad Chelius 
Yeah. You know, we love to kind of promote the idea that accessibility benefits everyone. And anybody who’s ever had to read a PDF on a mobile device knows the pain and suffering that goes along with that. Like, a great [story] I was in my backyard trying to fix my lawnmower one day, and I have the PDF manual on my phone and I’m pinching and zooming and trying to [read], and I tap the liquid mode button and keep in mind that doesn’t work on every PDF, but I tapped the liquid mode button, and I was like, “Wow!” You know, all the sudden I could scroll and I could access what I wanted. You can search, you can… It’s just a really… It’s very similar… You guys have heard me talk about the pdfGoHTML plugin that will render a PDF as HTML, that’s really what liquid mode does. It kind of renders your PDF document as… Is that officially what it’s doing Matthew? Is it rendering it as HTML?

Matthew Hardy 
It is. Yes, you effectively are building an HTML application that helps you consume that document.

Dax Castro 
One of the nice things that I liked about it was the fact that it had the header, the heading tag at the top of the page, so as you were scrolling, you kind of knew what section you were in. And I was like, “Wow, that’s pretty cool to have that persistent header in there.” Somebody definitely thinking ahead. So I was excited. Now if we could just get reflow mode inside the regular Acrobat to follow suit, that would be amazing, that would be great.

Matthew Hardy 
And hopefully, in not too distant future, we can definitely look at how we can improve that space as well.

Dax Castro 
Right. I think the biggest problem with the reflow is just when you have white text in your document on – let’s say – over a photo or over some colored background, and you go into reflow mode, that text stays white instead of defaulting to black, and so you end up with missing text. But you know, everything is in evolution, right?

Matthew Hardy 
And that was something we definitely had to think about with liquid mode when is it appropriate to keep the color of the content, whereas when do you have to switch it to black, so you can see it if it moves away from the content behind it. And a lot of thought went into understanding those relationships between content and allowing us to produce a really powerful experience based on where things we float and what needed to reflow with each other. And I think, you know, that’s one of the things we put a lot of energy and effort into building that feature.

Dax Castro 
Awesome. Well, it definitely shows. I mean, I really thought the product did really well. I think some of the things like when there were watermarks or there were bigger images, it has issue, but I think for the most part though, the benefits that it gives you be on such a small device are far outweigh the few times where it might have a hiccup. And we’re using our eyes to digest that information. And so things like a watermark or large image or something that might break up content in a different way, don’t affect people who are using assistive technology on a mobile device. So for them, it’s still as perfect as it needs to be. So I think it’s definitely a great product. I’ve been thrilled to be able to use it. And I’ve used it several times. A lot of times, the small print for me is difficult. And so, like Chad said, having to pinch and zoom is just a pain. So it’s definitely a benefit.

Chad Chelius 
So I’m really impressed with how you implemented liquid mode. So we talked with Leonard about this. And I was slightly disappointed to hear that if we give you a tagged PDF, currently, you’re ignoring that. But that being said, I’ve been impressed with the mechanism that you’re using to determine the heading structure in the document. Because when you look at a technology such as autotag in Acrobat, we know that autotag is a very very rudimentary algorithm. Like it says, “If text is between 12 and 18 point, it’s got to be an H3.” It’s not smart enough to say, “This is the biggest text in the document, make it an H1.” And that’s what you guys seem to have been able to do. You kind of analyze and say, “This is the biggest text. You’re an h1. This text is a little bit smaller. This is an H2.” And it seems very, very good. Before, we hopped on this session, I was curious. And I’m like, “Huh, I wonder how liquid mode works with voiceover on a mobile device.” And I was incredibly impressed. I mean, I was navigating the headings, I was in a table, I was reading the cells. And I just have to give you guys a lot of kudos because it was a very well implemented feature. So really great job with what you’ve done.

Liquid Mode Accessibility Benefits

Matthew Hardy 
I’m really glad it works for you guys. That was definitely one of the key things that we wanted to address. You said earlier that liquid mode [is] a mainstream feature. It benefits everyone. And that includes users who rely on AT and other assistive technologies to consume the documents. But it is obviously something that’s beneficial to many people. In all walks of life, people need accessibility for different purposes and at different times in their lives. Maybe, it’s that I just need hands free reading in a car. Maybe, it’s that when I’m, sort of, in a low light situation, I want to bump the size of the text and personalize it. As I get older, I’m starting to find my eyesight is not as good as it was a few years ago. And I’m starting to have to increase the font size just to read content. And that’s where PDF has challenges. If you don’t mind my digressing and just talk a little bit about the history of PDF and how that played into liquid mode, PDF is nearing 30 years of age now. It will be 30 years old in two years. And the situation that we’re addressing is Adobe and the PDF try to address was fairly different.

Matthew Hardy
On computers, all look very similar. We’re had lots of problems with visual fidelity across these systems. And so PDF was very much centered on the idea of digital paper, perfect visual final form fidelity, so to speak. And that was obviously a hard problem when we started and then took that on. Now, when we look today, the world has changed so much. Internally, at Adobe, we talked about from watch to wall as a concept. And we have… You know, I’m sitting here with a watch on my wrist that’s got a screen and I want to view content. I’m getting messages. I’m getting content. But I might go something that’s a massive size screen. And I’ve got AR and VR these days. There’s so many different ways of accessing material. And here’s this PDF that’s designed really as a canvas with layout glyphs and vector art and images to produce a perfect print reproduction. So PDF has been incredibly successful at addressing that need for perfect reproduction. But where it historically had trouble is when you want to scale it to different media. And that’s why we came in with liquid mode.

We wanted to understand, “Can we address more screens?” And when we thought about where’s the biggest challenge people face today, and if I want to work everywhere I am…

Matthew Hardy

Matthew Hardy
If I want to be able to work when I’m out and about when I have just my phone or maybe even just my watch, how can we make a compelling experience that basically works for users in those situations? The mainstream use case for us was that for mobile. It targets for phones and tablets. They’ve allowed us to really think about how we could bring a compelling experience. Tagged PDF was actually introduced not that long after the introduction of PDF in 2001. So that’s already 20 years old. But the reality is that it’s only used about 20% of documents in the market today as we’ve discovered in our own investigations. And even then, it’s not usually high quality tagging. It’s the exception to the rule to have a really high quality tech PDF. And so when we thought about how we make a great experience for mobile, we realized that we had technology at Adobe that had been around for a long time, but then it wasn’t really living up to the needs of the modern environment. So we took AI techniques, machine learning to

Dax Castro 
Sensei technology?

Matthew Hardy 
Exactly! Yes, we took Adobe Sensei and the concept of using machine learning and modern deep networks. And just for the record, I’m not an expert on these technologies. So I talk…

Dax Castro 
That’s okay. We won’t grill you.

Matthew Hardy 
But basically, we can bring this technology to play combining it with our existing technology stacks to produce really rich, deep understandings documents. We do things like we recognize the text, the Unicode, we identify the word breaks, we identify the language of the content. And then we go further. We start thinking about reading order, the logical structure and semantics of the content. And beyond even that, the layout characteristics of that content, so that we can understand fundamentally, what do we have in this document. And then from there, we can take that and transform it into, as you said earlier, an HTML application effectively. We build a reflowable dynamic presentation of this. And then we layer on top this additional information, whether it’s that outline mode, where you can do nonlinear navigation of the document. You’re moving around a document, getting a sense of the content that’s in there, keeping that heading floating at the top, as you said earlier Dax, where you can basically get a better sense for where you are in the document at any given time. And by bringing all this together, you have a mode that you can personalize. If you want to increase the font size, and if you need to increase the Word Spacing, the line spacing, we allow that personalization to make it a more readable experience for everyone. And then using this exact same semantic understanding allows us to drive accessibility and great interaction of voiceover and other technologies. And that’s how we bring a liquid mode together to produce a compelling experience for, hopefully, everyone who needs PDF on mobile.

WCAG Minimum Font Size for Accessibility

Dax Castro 
Well, it’s interesting. You know, there is a WCAG principle. And a lot of people… You know, one of the questions we get is, what is the minimum font size for accessibility? And the truth is, there isn’t anything in the Web Content Accessibility Guidelines that says what the minimum font size is, however, there is one rule that talks about the technology must have the ability to increase the letting and the kerning and the font size to these certain minimums. And that’s what you’re addressing. I mean, you’re giving them that ability, the end user the ability to have that visual experience that’s customized to their specific reading ability. I mean, you can’t do that in a PDF. You can’t… On a Windows machine or inside Acrobat, you don’t have that ability to… You know, reflow has some certain capabilities, but inherently, the document is what the document is. And that’s the whole idea of portable document format is. And you’d said before, an accurate representation, a one to one of what it was in the design, in the source file. So I think, the ability to have that flexibility inside the mobile and tablet device, really, like you said, expands the usability of the document for a wide variety of users, not just from a screen reader standpoint, but from people who have the visual impairments.

PDF Association

Matthew Hardy 
Definitely. Like when we talk about this… So, I’m also heavily involved in the development of accessibility standards with Adobe. And we through to the ISO committees of the PDF Association. And we spend a lot of time thinking about what’s really important to the PDF. And more importantly, how do we address this broad spectrum of accessibility needs that people might have. And we think about things like color contrast and things like WCAG minimum color contrast, that’s really important. But for some people who have issues with color contrast, maybe they need lower contrast, some bright higher contrast. As you said, some people want larger font sizes and some people want smaller font size. And when we looked at what enables readability of the document – there’s teams who look at this stuff – basically, there’s no one size fits all solution for reading, even if you can’t choose that there’s a common principle or the lowest common denominator that we can target. Having that adaptability being able to customize the content, potentially switch into things like a dark mode for some people who want different levels of contrast, we have that adaptability and that ability to cater to you the individual, while broadening this to bring PDF to everyone, hopefully.

Dax Castro 
Well, I’ve been in some presentation… I’ve presented some presentations where my slide deck has been black on white, and someone has said, “Look, that’s too much contrast for me.” So now I only use like a 10% Black. It’s like just a light gray on a black, and maybe I don’t use full black, I use 90% black, so that there’s not that 100% to 0% kind of contrast. But you know, it’s like you said. One person’s accessibility is another person’s barrier. And just for our listeners, the success criteria which I was just mentioning is 1.4.12, which is part of the new WCAG 2.1. And it has to do with text spacing. And a lot of times people think that that applies to an acrobat document, but there’s actually a statute in here that says, “The current success criteria does not apply to PDFs as it’s not implemented using markup.” Meaning, a PDF is a flat document that uses non-flexible content. So therefore, it is exempt from these text spacing and things like that. But hopefully, in the future, your team could develop a kind of take reflow to the next level and integrate that liquid mode. It would be great if liquid mode was the reflow that’s inside Acrobat. That would be outstanding.

Matthew Hardy 
So we would love to see that. Obviously, we definitely wanted to address the key problem of mobile first. But I would definitely love to see liquid mode continue to expand as a platform to enable access to content wherever you need it.

Understanding PDF/UA

Dax Castro 
Hey Chad, do you want to ask the question about PDF/UA? I’d love to get Matthew’s voice on the acceptance of PDF/UA and PDF/UA 2.0 as a standard. I think one of the things that we struggle with, obviously, is being advocates for PDF/UA versus WCAG. Because most people are just trying to check the box. They’re just… What is my section 508 requirement? Well, it’s WCAG, not PDF/UA, so why should I have to do extra. So it’d be nice to have that conversation about the benefits of having code compliant content, because it allows people like Matthew to have the flexibility to take that code compliant content and repurpose it in ways that you couldn’t, if it’s just simply WCAG.

Chad Chelius 
Yeah. Matthew, I mean, you had brought up the fact that the PDF format is 30 years old at this point. And you know, I’m old enough to remember those early days. But what’s really interesting is that the early days of PDF and using Adobe Acrobat, you could not edit a PDF. Like, there literally were no editing features in Acrobat in those early days. And I’m talking… I’m gonna misspeak here. But I think it was like version two, version three of Acrobat or whatever. And they’re literally… You know, people would say, “Can you edit this?” I’m like, “Well, no. I gotta go back to the source file and make a new PDF.” But it’s a great example of how user demand dictates how the application evolves. Because a few years later, all of a sudden, there was, in fact, an editing tool in Acrobat that allowed us to make those changes. And I just think it’s an interesting perspective as to how Acrobat has evolved over the years. And I think liquid mode is a great example of that how demand kind of forced the evolution of the product. And I think that’s really need.

The History of PDF/UA

Matthew Hardy 
Yeah. And when you look at the history of PDF, it’s definitely been something that has been a growing standard for a long time. When PDF was first introduced in ’93, it was pretty simple. It offered text and vectors and images. It didn’t even have color fidelity across different devices at the time. It was trying to meet that basic demand of visual fidelity of systems. And that had to evolve over time. It started off completely static. And then over time, things like forms and links and video and audio are added to it. So it’s been an evolving standard. And as that interactivity has increased, some of the use cases of workflows. It’s also worth mentioning that the mechanisms that actually drive accessibility, content, reuse, extraction, all these other things that allow us to repurpose and benefit from the information within the PDF actually started in 1996 with the introduction of mark content as metadata to allow people to embed information into the flow of display information to help them repurpose that content on the inside. And that slowly evolved into what we now know as type PDF logical structure and such and so.

Matthew Hardy
Similarly, as users engage with content, we’re very much listening to what our customers need. What do people want to achieve with PDF? It’s what you want to accomplish in your workflow on that document. And of course, creating content. Starting with simple, just want to create one character. I misspelled a word to potentially anything paragraphs and whole pages and even floating across pages. This has just been an evolution. Because PDF is intrinsically a digital paper, it’s effectively something you print to, quite traditionally, it does mean that there’s a lot of post processing needed to figure out how to achieve this piece of functionality. And as processes have got faster, as we’ve had more time to develop these features, as Acrobat has matured as a product, we’ve obviously expanded these capabilities over time and PS is so far great container for content, but also data and rich information. The more we can encode that into a PDF, the best place to do that is authoring time when you’re creating the document as a republication system. But we do know that a lot of systems just “Print to PDF.” When you “Print to PDF”, you’ll lose a lot of that rich information. So although, we do encourage tooling that does produce high quality tagged PDF out of the door, we obviously have a lot of systems now to understand PDF on the other side, and to try to put that information back in to get yourself.

Read Order and Tag Order

Dax Castro 
You say “Print to PDF”, and my heart literally shrinks two sizes. So it’s interesting, you talk about the evolution of the PDF document itself. And Leonard touched on this just a bit when we talked with him about the evolution of the read order and the tag order panel, that originally the read order panel was the only way to really set the order of the content. And as evolution kind of built itself out, the tags tree became that… Can you talk a little bit about that? And are there plans in the future, maybe, to kind of homogenize those two things?

Matthew Hardy 
I don’t know about plans to homogenize them, but it’s definitely the case that with the integration of logical structure, you have a separate means of determining your reading order for a document. Traditionally, PDF is about laying down content onto a canvas. And the concept of the order or the ability to kind of have overlaps and then transparency and interactions between content is very much based on the idea that you paint in an order. And as you lead that order up, it produces the right appearance. And so content order, so to speak, has always been about visual fidelity. And of course, you know, back to either the dawn of time and where people had to do gutter hopping to align printer lines and such, and so many things, you know, maybe optimized printers for font usage to use one font before use another font. And so order of the content stream was very much about optimizing for printing and visual fidelity, and not so much about how am I going to get this content back out and read it somewhere else. Incomes logical structure, and now we have a separate, basically, tree that allows us to say this is the logical arguments. Meaning, this is the flow that I would expect to read as I go down columns and across columns with the tables, lists, paragraphs. You know that this drives the order. And fundamentally, those don’t have to be fully aligned. But it’s really important for, obviously, processing these documents that they do align.

Matthew Hardy 
The requirement and things like PDF/UA that basically the document logical order is correct, meaning that it matches what you would expect it to be as a consumer of that document. And then if I process it, using that logical order, I get the right thing. But a lot of processes don’t even look straight to treat to do basic things like even reflow and reading. And that’s why you run into a lot of these problems, where you really do need that reading order in both places. It’s also much more efficient when you perceive a document if you’re not jumping all over the content stream trying to compose a big string of text to read your interface. And so basically, we do try to align those, whether we are going to be able to ever fully achieve that because there are some places where transparency means you have to lay things in a certain way to produce an exact visual effect. And so we try to make it as congruent between the two as possible without it forcing them to be identical, if that makes sense.

Dax Castro 
Well, but that’s the content panel. So for those of our listeners, we have that. We’re back to the age old argument. We’ve had quite a few times on the show, where we have the tags tree. PDF/UA says, “Tags tree is the king.” It is what sets the read order. But then you have the content order, which is what we’ve been talking about. Matthew has been talking about, the Z order, that the layer order in which objects are built. And then we have the read order or the order panel which it’s now been renamed. And that is this other order that, I call the ACL, the architectural content order for the document and these three things don’t always align. Now, it’s only okay that the content panel doesn’t align with the tags tree. But most people, HHS being one of them, higher-ed, a lot of the time is another that they want the read order, the order panel to match the tags tree, because they’re not using JAWS or NVDA, which we know do look at the tags tree, but they’re using something like read and write… Text helps read and write, which just looks at the order panel. And so we’re constantly having to have this teaching moment with people that tell them the difference between these three different things. And we’d love to see Adobe take a hard look at either making the order panel go away, and just redirecting. It’s almost like a 301 redirect. We just want you to 301 redirect to the tags tree. You know, when in doubt to the tag tree. And that would really help. There are so many users that spend hours and hours of time. One of the key questions we have for clients, a lot of the time when we get a document is, do you want the read order or the order panel to match the tags tree, because it can add 30% to 50% more time to a document to have to go in and do that.

Chad Chelius 
You know, where PDF/UA focuses strictly on the tag tree. WCAG, however, does, in fact, maybe not directly referenced read order, but does indicate that it needs to be read using all technologies.

Dax Castro 
Right. And that’s the reflow. That’s the new version. That’s WCAG 2.1 reflow statute that has been implemented. That is a 1.4.10, which says, “Content can be presented without loss of information and functionality, and without scrolling requirements in two directions.” In other words, that the document should be able to be reflowed without loss of order.

Chad Chelius 
Right. And I mean, as much as we talk about it being a pain, when Matthew explains it, it really does kind of make sense. Because one is the stacking order of the elements on the page, where tags are… I always say like tags are imaginary. They’re not physical things on a page. They’re just identifying the content. And you could delete all the tags in the document. And it doesn’t affect anything in the document visually, where the tags are – I like to say – somewhat imaginary. You know what I mean.

Dax Castro 
We’d love a button that says, match the read order to the tags tree. Boom! You hit it, and it just magically… Right now, what we teach people is don’t touch the order panel ever. Go into the content panel, and adjust the content panel to get the read order to reflect the right thing. But the moment you start messing with the order panel, it starts changing your tags tree. And if you spent all this time getting your tags to be perfect, and then you’re like, “Oh, my read order is out of whack.” You just move a few things, all of a sudden, your tags are combined in ways you didn’t want it to. Playing with the content panel avoids all that.

Matthew Hardy 
Definitely. You know, obviously, we’ve been dealing with a lot of PDFs that doesn’t even have a logical tags tree today. And so thinking about the order and the flow of order and how to control. You know, within your sub-content, the things like paragraph, the text within them. That’s something that the tags panel has traditionally never dealt with. The content panel does, the audit panel does to some extent. And so it is a… I do think now that we have the ability to bring these AI techniques to understand this content, we have a tag tree for every document. It just potentially open up new avenues to explore in terms of how do we produce better interfaces, how do we do a better job of exposing this to customers.

Which Order to use?

Dax Castro 
Well, and I think that part of it is you get some people saying, “You know PDF/UA says we don’t care about the order panel. But HHS says order panel absolutely has to be set.” Most Canadian organizations are demanding that the order panel to be set as well. So we’re in this fight where it’s like, “Okay. This standard says no. This standard says yes.” And then somewhere in between, we’ve got people who aren’t really trained on accessibility very well. And they’re trying to make sense of it. And they come to us as instructors, and we’re like, “Yes, sort of, and yes, be sometimes, and yes, whenever you know.” And so I talked to the people at text help. I said, “What’s going on? I want to see. And I’m like, “Okay. Let me corner you and give you… I need to know why are you guys using the read order panel?” And their answer was, “Look, we built our entire company around this read order panel 20 years ago, when it was the only thing. We’re not going to completely scrap everything we’ve ever done and develop a new learning, a new program that’s going to reads the tags tree. So it is what it is. And that was their answer.

Matthew Hardy 
Yeah, it’s a difficult one actually. Because, you know, fundamentally, even before you get into UA, when you just look at ISO 32000, the PDF standard, it does strongly encourage a lot of alignment between content orders, the content stream, and the stretch tree, and trying to make sure that those are as aligned as possible. While acknowledging that there are going to be situations for visual fidelity that you just cannot have them 100% White. I think that that’s one of the big challenges that people faces. How are we expose this in a way that makes it intuitive to build this? And ideally, the two aren’t in contradiction to each other. They’re facilitating each other. They’re complement each other. And that’s what we want, eventually, is to have that complimentary view. I do think that as we build a strong future for PDF, PDF, 2.0, PDF/UA 2.0, we are going to double down on that struggle. We are, obviously, still really, we do want to think about content and order and such. And I think what you’re telling me here is we… In particularly, in UA committees, we need to do a good job of thinking about this and understanding the customer’s needs and such around this, and the confusion that it causes when you have so many different ways to do things.

Dax Castro 
And one of the other things that would really solve it, if the read aloud would just follow the tags tree, because we get so many clients, they don’t use JAWS or NVDA or any other technology, we give them a document, we say, it’s compliant. We give them the PAC 3 report, they open it up, and they want to listen. They want to know that you’ve done it right. And what do they do? Well, they open up the tool they have, which is Acrobat’s read aloud. And, of course, read aloud uses the order panel and not the tags tree. And so you could literally overnight make the order panel obsolete if you just told the read aloud to read the tags tree. And then it wouldn’t be an issue. Because then none of the lay people who don’t know or don’t want to have that 15 minute conversation about order content and tags can just feel good that it reads the right thing.

Matthew Hardy 
Yeah, definitely.

Chad Chelius 
And I mean, it’s in the standard. It’s just the adoption is not there. I mean, and that’s the point of PDF/UA, right Matthew? I mean, the point of PDF/UA is to create that standard that everybody follows, developers, software manufacturers, users all. And you know, we just don’t have that adoption yet.

Why PDF/UA as a Standard for Accessibility?

Matthew Hardy 
So let’s just talk for a second about PDF/UA. So why do we do PDF/UA? And I actually came into the first UA standard partway through. So this has a long history, a rich history, and a lot of backing behind why it was needed. But while PDF itself delivers the mechanisms that allow tagging and accessibility, there’s a lot of open endedness in the spec that says, these are capabilities, but we don’t really tell you how to use these to produce a good result. So when we looked at PDF/UA, particularly, we’re thinking, “Well, what are the syntactic requirements for those who are actually building PDF using the creation software or manipulation software to actually control the syntax within the PDF? How do we actually put the right information in, have the right properties to enable all of these downstream accessibility workflows? And that includes things we just talked about with reading order. And the various structure tree, logical order, and how those relate. But we also thought about two other things which were processor and ET requirements. Because, again, that’s where PDF itself doesn’t say very much at all. It just tells you here are these capabilities. And so the intent was to try to help people producing PDF software, whether it’s direct renders, like Adobe Acrobat or whether it’s something like AT that’s trying to consume that content out of those systems. We wanted to make sure that there’s better consistency. And so when you think about PDF/UA, it’s both of file formats in tech spec, but it’s also a guidance to your processes and at to help have more consistency. I think, as those evolve, going forward, they will help us to have more consistency in how they’re used and reduce user confusion.

Dax Castro 
Do you have JAWS or NVDA? Do you have any of the screen readers as members of the PDF/UA organization?

Matthew Hardy 
Not to day, we don’t. And that’s something we definitely want to change. Having the people who are actually going to use these specs be fundamentally part of developing them is really important. There are definitely investigations going on. I serve under the PDF association to understand how to better engage the audience PDF/UA, whether it’s, as I say, the creation software, whether it’s the vendors have to implement it, to make sure they’re in the loop. So we do have a lot of people who are experts and build software around remediation, around building good high quality type PDF, but there is a gap today in terms of having the vendors who built the AT the consumer. And that’s something we do need to address.

ISO 32000-1: What is it?

Dax Castro 
Well, we’re actually just to give our listeners a little sneak peek. I have… In a couple of weeks, I’m going to be having a meeting with someone from one of those AT companies to see if they’ll come on as a podcast guest. So we’re hoping to maybe further that connection a little bit and get them a little bit more interested in. And I know it’s difficult on both ends. Writing the standard and creating a product that’s usable for everybody else, it’s a big lift. And I can tell you, I am one, probably, only one of a very select few who’s actually read every single page of the chapter 14 inside the ISO 32000 about what PDF/UA has to say on document accessibility, and the 14 to 89 chapter seven that… I tell our audience, guys, you really… Lots of times, we get the question of how do I structure this or what does this really mean? What is the OT? You know, what is the object tag mean inside a PDF? And what are my possibilities? That’s where the ISO comes in. And I can go into a specific table and look at here are the properties I can choose from. Here’s what they mean and what they affect. And I think that really helps you expand what your can do, or even just your understanding around what it is you’re trying to do inside an Acrobat document. The ISO 32000 is almost 800 pages. So it’s definitely not like reading… But like you said, Matthew, most of it is about creating, like from scratch, from 08 from a code standpoint, creating a PDF document. Whereas chapter 14, document interchange, really focuses on the accessibility, functionality, and code standard for the tags and tags trees and a lot of the tag properties to be more specific.

Chad Chelius 
Dax, I’ll have you know, I also read the spec. Unfortunately, I never get past like half a page because that’s how I fall asleep every night. But I’m trying. I only get like half page at a time, so it’s taken me a while.

Matthew Hardy 
And it certainly is a dense reading material. And I think that is why we’ve tried to help to facilitate access to these documents, particularly with the PDF association through things like Matterhorn as a way of looking at UA and breaking it down into really easily understandable checkpoints that people can use to do that. Various guides that help to understand the intent of these standards. So that it can be lighter reading, because sometimes looking at a table 365, And reading through the properties of that table, it is pretty dense. And it’s pretty hard to tie it all together into how these things used to actually produce them. I am sure, Dax is an expert on this. You’ve read this enough to tie this all together. But it is very dense reading as someone who works in the space and has actually written several parts of 14.8 in the new ISO 32000 part 2. There’s a lot of material in there to figure out, to understand how it relates to other parts of that material.

Dax Castro 
And for those of the geeks in our audience, table 365 deals with entries in and output intent dictionary. I random, you know, fact. You know, it’s interesting if… For those of our listeners who are interested in more information, we did a podcast several episodes ago on the Matterhorn protocol where we walk through the principles of the Matterhorn protocol and talked about what the standards were for them. So definitely in your folder on your computer, you should have the ISO 32000, 14289, the Matterhorn protocol, and the best practices tags syntax guide. All of those should be in your wheelhouse and you can use all of them as references. So many times, we get the question of how do I structure a TOC when I’ve got this situation. And I always go back to the best practices index guide, because it’s got a little section in there that shows you, here’s how you would structure, here’s a couple examples. Matterhorn protocol as well has a couple of sample structures. And we know unfortunately, one of the things about the tags structure in a TOCI is when you have a reference tag and then a link tag inside the person using the AT link because of the double and we talked about that actually in our PDF Accessibility Facebook group. We answered kind of why that was and what are some solutions around it. So definitely, these standards are really a great tool. So all right. Well, Matthew, it’s been amazing having you on. If you could impart any thoughts or wisdom to those people out there who are interested in getting more active in accessibility guidelines or understanding the more, what are some good courses of action they can take?

Getting involved with PDF Association

Matthew Hardy 
Yeah, that’s a great question. So the first thing I’ll say is, you know, a lot of the development today of the accessibility standards around PDF have come through the PDF Association. It’s a vendor body. It’s neutral. It’s not a sort of one company. And it allows many different people to come together and discuss and evolve these standards. They’re the groups that are basically built most of the documents you described with the, you know, in relationship to ISO, whether it’s best practices on tagging, whether it’s Matterhorn. And we do a lot of work. I’m actually the co-chair of the PDF/UA technical Working Group. Addition to just the technical working group that looks at the spec and things like that, we have marketing working groups that look at things about how to get information out there, we have a Liaison Working Group at the moment that’s working on improving the workout techniques for PDF to try to make it do future things.

Dax Castro 
Awesome.

Matthew Hardy 
There’s plenty of ways to get involved in. It’s a great place to come together and have those discussions. And the success of the standards is very much about all of the various people within the industry coming together and discussing these things and understanding them. And that will have greater success with greater participation. And so I strongly encourage people to join them.

Chad Chelius 
So yeah, and if any of our listeners is are interested, just head on over to PDFA.org, and go ahead and sign up there. There’s a lot of good information over there. There’s actually a conference coming up called PDFA, that is going to be full of a lot of really interesting information and discussions.

Matthew Hardy 
Well, and one of the other things too is that if you’re a member of PDFA, you get to be part of these working groups, you get to see these drafts evolve and be able to voice your… You know, having a voice in what happens. And one of the great things of… You know, it’s one thing to be at the end of the chain and complain about what can and can’t do, what the standard does to that standard or what those idiosyncrasies are, but it’s another thing to be part of the change. Being an effective contributor, I think is a great thing. And not all of us have the time or the capability or honestly the interest to do that, but there are those of us out there – and thankfully, PDFA is full of those people – that are interested in. So if any of our listeners are passionate about effecting change in the standard, by all means PDFA.org, please.

Chad Chelius 
Alright guys. Well that brings us to the end of another awesome podcast. Matthew, thank you once again for being our guest today. And we also want to thank AbleDocs for being our sponsor, AbleDocs, once again is the makers of axesWord, axesPDF , document remediation services, as well as website auditing and testing. So thanks again for doing that. My name is Chad Chelius.

Dax Castro 
And my name is Dax Castro, where each week we unravel accessibility for you.

Chad Chelius 
Thanks, guys.

No Comments

You can be the first one to leave a comment.

Leave a Reply

Your email address will not be published. Required fields are marked *