Compliance and PDF Accessibility

Okay, so it is two o'clock here on my clock, so  I think we should get started. Welcome,   everyone. Thank you for joining another edition  of our Equidox Webinar Wednesdays. Today,   we're just going to be talking a bit about  compliance and PDF accessibility, so just   appreciate everyone for joining. My name is Dan  Tuleta. I'm the Senior Sales Engineer here at   Equidox, so I'll be running us through the webinar  today. Just a quick rundown of the agenda: first,   we're just going to briefly talk about who Equidox  is, who we are, and what we do. We're also going   to talk about PDF accessibility challenges  and our solution to that challenge. Also,   we're going to be discussing why we need to make  PDFs accessible. We're going to be talking a bit   about the legislative and lawsuit side of  things as well, and we will spend a good   chunk of time doing an Equidox demo. So anyone  who hasn't seen our Equidox software before,   we'll give you a rundown of how that works  and how it might apply to your PDFs and   your documents and your accessibility challenges. Okay, so Equidox Software Company, our mission is   to enable PDF accessibility through intelligent  automated solutions. So we are a leader in PDF   accessibility. We have two main core components  to our business. We sell our Equidox software,   so it's a software-as-a-service, a web-based  application, which we'll be going through today   in the demonstration. It is intended for quick  and easy efficient remediation of PDF documents.   We also offer an AI-based solution, so Equidox AI  is more for those recurring type use cases where   you have a very high volume of documents  that have a lot of similarities in them.   So think of bank statements or directories or just  documents where you're constantly updating them,   or they're constantly changing, or you are  producing many thousands or millions of them on   a recurring basis. Obviously, that's just a volume  of documents and pages that no human could handle   remediating on their own. So we can use Equidox  AI to train artificial intelligence models to be   able to automatically apply the ACC and compliant  tags to those types of documents. So we won't be   talking too much about Equidox AI today. That's  another webinar for another day. It's a totally   different type of tool for a different use case,  but if you have any questions about Equidox AI   and how it might apply to your documents, you  can definitely reach out to us. We'd be happy   to kind of understand your situation and give  you a more personalized demonstration of it.  So let's talk a little bit about the market  challenges. The challenge of PDF accessibility   is no secret to anyone who is familiar with  traditional methods of tagging PDF documents. The   process can be extremely manual and time-consuming  for people who are using tools like Adobe Acrobat,   for example, which is really synonymous with all  things related to PDFs. I like to refer to Adobe   Acrobat as a tool that kind of resembles a Swiss  Army knife; you can do just about anything with a   PDF document, but it is not built with the ease of  use or the ability to do things quickly in mind.   So it's just unsustainable as a tool that you have  to use for large numbers of PDF documents because   it just takes too long and it's very technical  and requires a lot of experience and expertise.  So this really leaves organizations behind the  eight ball when it comes to getting caught up   with their backlogs of PDF documents that are  still out there in circulation and also on top   of making sure that all of the new content moving  forward is tagged and compliant. Because of how   technical and tedious Acrobat is for tagging PDFs,  this often forces the remediation responsibility   into the lap of just a small set of experts who  are often unable to keep up with these backlogs   and the constant flow of new documents and content  being produced and updated week over week, month   over month. This will usually lead to issues with  quality because the amount of work that's required   is simply unsustainable if you're having to use  these traditional manual tools to tag the content.  So why are we making PDFs accessible? Let's kind  of talk about why this is important. Many people   have probably heard of laws like the Americans  with Disabilities Act and different compliance   standards like Section 508 or WCAG, which  stands for the Web Content Accessibility   Guidelines. But at a high level, our world  is becoming so much more reliant on digital   information. And just as organizations have  to comply with the ADA by offering wheelchair   ramps or handicap parking spaces and Braille  signage for their buildings, they also now   must ensure that their digital content that they  are distributing to their customers and people   is accessible to people with disabilities. So these articles that you see here on the   slide deck, they are linked. So when we send out  the slide deck afterwards, if you'd like to read   in more detail about the ins and outs of Section  508 and the Web Content Accessibility Guidelines,   feel free. We won't go too deep into the  weeds on that today because I'm not a lawyer,   so I don't want to stand here and preach about  the exact letters of the law. But the fact   of the matter is that all PDFs and all digital  content must be made accessible as we become such   a more technologically based and digital world. So again, this is another article here that will   be linked if you'd like to read more on the  topic. But recently, the Attorney General has   signed a final rule on digital accessibility  under Title II of the ADA, mandating that all   state and local governments must adhere to WCAG,  those Web Content Accessibility Guidelines 2.1   Level AA for content and mobile applications,  and this of course includes PDF documents.  So exceptions are not a free pass. There are  many instances where PDFs still need to be   accessible. So the rule of thumb is, if a PDF  document is still actively in use or relevant   to multiple individuals, it likely needs to  be made accessible. And if it's your document,   it's your responsibility to make  sure that it's accessible. The   Department of Justice has provided a  Fact Sheet, which is again linked here,   citing numerous examples when exceptions do not  apply. We can also talk to you if you'd like to   have more of a one-on-one discussion to discuss  your use case and your PDF accessibility backlogs   to kind of help you put together a prioritization  of these documents. So if you're staring at a   pile of hundreds or thousands of documents, it  can be quite an intimidating thing to look at.  So how do you get started? Our general rule of  thumb would be to start with the newest content,   the stuff that is going to be the most up-to-date  and most relevant, then work towards the easiest   content. So when you can get through all of  the content that you can remediate quickly,   then start thinking more about the older,  10-year-old types of reports that are probably   not being interacted with as often, and also  the very complicated stuff, which could just   be more of a strain on your resources in terms  of time because it's just very, very difficult   to remediate certain types of those documents. We  can talk in more detail if you'd like to have more   of a one-on-one consultation after the fact,  but just keep some of those things in mind.  Okay, so to help quantify the scale of this  problem, Equidox conducted a survey in part   with the National Federation of the Blind. And  what we found after surveying about 250 blind   and low vision people is that they're finding that  two-thirds of their PDFs are not accessible. So   just imagine the frustration that you would have  if everything that you were interacting with on   a daily basis, you couldn't read two-thirds  of it. It would be extremely frustrating,   and you would just be at a huge disadvantage  in terms of how you go about navigating the   world. So that's what we're trying to change.  We're trying to make sure that we are closing   the gap here. We don't want to see 67% of PDFs  inaccessible and unusable to a large section of   the population. We want to make sure that we are  empowering organizations to do the right thing   and get these documents up to the right level  of standards so that they can be compliant.  Okay, so again, here is just some more data about  the rising number of lawsuits that organizations   are facing over digital accessibility, which  again includes PDFs. So the theme is very clear:   we are a highly digital society, and organizations  have a responsibility to ensure that their content   can be interacted with by everyone, not only  fully sighted people. These increasing numbers   of lawsuits are not going to stop increasing  until these organizations make the necessary   changes and improvements to their digital content.  So again, it's just not slowing down. The ADA   is not going anywhere, and people are going  to continue to become more and more reliant   on digital information. So we have to start  working through our backlogs and PDF problems.  Okay, so when we are going through the  demonstration today, which we're about to move   into in just a moment, I'm going to throw around  the word “tags” pretty often. So just anyone   that's new to PDF accessibility, there are these  little elements in a document called tags that we   are using Equidox to create or fix if the document  was already tagged. The tags are basically   these little digital markers that are used to  identify and organize the content on the page. So   there are different types of tags: you have tags  for elements like text, images, headings, links,   lists, tables. The tags must be in the correct  order so that the PDF will be read in the correct   order because obviously you can't have content  that is just being kind of scrambled around the   page and read in any random order; it could render  an entire document useless if that were the case.   So that's really what we're going to be doing  inside of Equidox. We are going to be organizing   these tags to make sure that this document is able  to be interacted with using assistive technology,   a screen reader for example, and that the user  who is using that screen reader can extract the   exact same information that a sighted user  can by visually looking at it. So different   elements must be read in a different way,  and depending on what type of element it is,   that's going to determine the type of tag that we  apply to it. So just to give everyone a heads-up,   when we're throwing around the term  “tags,” this is what we're referring to.  Okay, so we are going to jump into the Equidox  demo now, and the demo itself will be, of course,   recorded. So we will link it here inside of the  slide deck. If you want to share this amongst   your organization, show it to different  people, feel free to send that along,   and we'll insert the link to the video as well. Okay, so let's jump over here to Equidox now.   Equidox is—this is my Equidox application. If  you notice here, I'm operating directly in my   browser. So what's nice about operating in the  browser is I'm not tied to any one individual   computer. Equidox is a web-based application; you  can interact with it from literally any computer,   assuming you have an internet connection, of  course. We also work with a concurrent user   licensing model. So what that allows for is if  you, let's say, have 10 concurrent licenses as   an organization, that means any 10 people can use  Equidox at the same time. Maybe you have 20 or   30 or 40 people that theoretically might be using  it for a project that they're working on, but you   don't necessarily expect them to be sitting there  for 40 hours a week remediating PDFs. Oftentimes,   people just need to log in for 10 or 15 minutes  and work on a document. When they log out,   that seat that they were using is now  available for someone else to log in.  So it's very flexible in how you go about  deploying it to a larger group of users   because you don't have to purchase an annual  subscription to anything for a user that just   might need to use it once a week or once a month,  very infrequently. So you're not having to install   and update and maintain software across every  one of these individual computers. The other   nice thing about being web-based is that Equidox  is naturally collaborative. So what that means is   you can share documents with other users in your  account. So if you needed help on a document,   you could ask a subject matter expert because  maybe there are diagrams or charts that you're not   really sure how to describe with alternative text,  and you don't want to make assumptions or guess   at what you should be typing in for alt text. So  you'd rather share that with the person who might   know more about that subject. You can also work  together on the same document at the same time as   another user. So if you have a very large document  with a tight deadline, maybe, let's just say it's   a 500-page-long document, that's a very daunting  task for any one person to just kind of tackle   500 pages on their own. But maybe you've got five  really strong remediators and everyone can take   a 100-page section of it. Now you're getting that  project done in 20% of the time because you've got   all hands-on deck working through that really  large document that has that tight deadline.  Okay, so what we're going to do is, I'm actually  going to import a document for us to work on   today. What we're looking at now is just kind  of a list of documents that will populate as you   start to import PDFs. Then your five most recent  documents will show up here across the top. If I   go to the Import Document tab, though, it will  take me to this screen. I can either click to   open up the folders on my hard drive, or if I  have my folders already open, I can just drag   and drop a document or multiple documents into  this gray area. That document will quickly upload   to the cloud, and then this blue Import button  becomes available. When I press the Import button,   this will start an automated process where  Equidox is now evaluating this document for   things like existing tag structure, which may or  may not be there—we don't quite know yet. So if   the document has been previously tagged, you  can use that existing tag structure and kind   of clean things up and make adjustments to what's  there. Or in some cases, you might prefer to just   start over from scratch. That's the thing with  PDFs; you don't always know what you're going to   get because you might not have had anything to  do with designing this document. It might have   been published 15 years ago from some old,  deprecated software, and you don't know what   you're going to get into until you start working  on that document. That's the challenge of PDFs;   they're all unique. They come from different  tools, different designers. Some are new,   some are old. So it's a very broad spectrum of  what you can get inside of a PDF, and that's   really what makes the remediation side of things  kind of challenging but also kind of fun because   it's kind of like putting together a puzzle. In  Equidox, think of Equidox as a productivity tool   that helps you solve that puzzle a lot faster  than you can with other, more manual tools.  Okay, so if I click here on the thumbnail  for this document that I just imported,   it will take me into the Document Detail page.  Inside of the Document Detail page, I can see some   thumbnails here. So I have two thumbnails for the  two pages in this PDF. I have some basic features   up here to, like, search for pages. If I need to  go to page 453, I can just jump straight to it. I   don't have to scroll through all those thumbnails.  I have buttons here to export this back to a PDF,   which I'm not really ready to do yet. I  could share this document with other users,   of course. So if I needed to ask for help on this  PDF for whatever reason, I can. And I, of course,   can delete the document, which I'm not going to do  yet either because we haven't done anything to it.  Down below, there are some properties of  the document. So this is the file name here,   and if you notice, in this case, the file  name and the title are identical. You can   always update the document title if one is  not provided. Equidox, at a bare minimum,   will fill it in for you with whatever the file  name is because that's an often overlooked part   of PDF accessibility. People will forget to add  a title, and then that's an automatic failure   from an accessibility standpoint. So if you  want to, you can always adjust your title to   make sure that it's something that's plain  English and easily understood. You can add   or edit your author if you want. So if you have  an author already there, it will be in there,   or if you want to add one, you can. And then one  other property would be the language attribute,   which Equidox will set to English for you,  just assuming that the document is in English,   which most documents are for most of our  customers. But if you hit the drop-down menu,   you can change the language attribute to whatever  language matches the content of the document. So,   for example, if you have, let's say, a Polish  document, you don't want an English screen reader   trying to read Polish words; that would sound very  silly to a Polish speaker. So that's how you can   adjust the language attribute, and you can always  save your properties if you make any updates.  One other important thing is the Images tab.  Now, this document only contains a couple of   images in it, but what's nice about the Images  tab is you get this sort of consolidated list   of the images together. So if you know exactly  what these images represent, you can quickly   streamline your approach to writing the alt  text. Or, you might prefer to actually get into   the document itself and make those decisions  about alternative text once you're in there.  So we'll jump into the first page now. And when I  get into the first page here, what I immediately   notice about the PDF is suddenly there are these  yellow boxes that are kind of like covering up the   content. These yellow boxes are what we refer  to as reading zones. The reading zones are,   again, like similar to the tags. These are  what's going to create the tags when we   export the PDF. So we are going to use these  reading zones here and organize all of this   information on the page to make sure that it  is going to line up with how a screen reader   user would be interacting with this content. If I don't like the initial detection that   I've been given, so for example, in this case,  I don't really want every line of text to be   in its own individual zone, I can move this  Zone Detection slider from left to right,   and you'll see how it kind of changes the  granularity of those reading zones. As you   bring it way over to the right, you get kind of  like a zoomed-out view of all of the data on the   page. Whereas if you bring it further to the left,  it gets much more microscopic and focused on the   data. You're typically going to find somewhere  in the middle that seems to work pretty well for   whatever page you're working on. In this case  here, the detection level of eight seems to be   fine. So now I've got all of my zones in the right  location, I have the right number of zones, and I   can start making changes to these individual zones  to make sure that they're all organized correctly.  One other thing you can do is you can always press  the Reorder button as well. Reading order is a   critical part of PDF accessibility. If content  is not read in the correct order, it can render   everything useless. Just think of like a newspaper  article with three columns. If the screen reader   is just reading clear across the page, you're  going to hear just a bunch of fragmented sentences   from totally separate columns, which would make  no sense to anyone. It would be impossible to   understand that document. So reordering the  content to make sure that everything is going   to be read either in a one-column, left-to-right,  top-to-bottom layout, or a multi-column option,   depending on the layout of the page, is  another critical part of PDF accessibility.  One other thing before we start making any  changes to the structure of our tags here:   if I press this button on my—the button here that  looks kind of like a computer monitor—it will open   up a separate tab in your browser. In this browser  preview, what you get is this HTML rendering of   the page that you're currently working on. The  reason that this HTML is important is because   instead of having to interact and understand  the complexities of the tag tree, you just get   this very simple linear representation of how  would a screen reader read this page if I were   to just stop working and export the document.  This shows you the reading order. Of course,   if you have tables, lists, or links, or anything  that is different in the document besides just   standard text, you'll see all of that proof  here in the HTML preview. So we'll be going   back and forth from the preview several times  just so you can kind of see how things change.  If I go back to the PDF, I'm going to start making  some changes here. One of the important parts of   PDF accessibility is heading structure. Heading  structure is a way that screen reader users can   navigate the document. They can jump from section  to section and drill down into a subsection of a   section, and they can find that information that  they're looking for in a much quicker way. If   everything is just tagged as text, there's really  no navigation help with a million little P tags   or text zones. In this case, when the user  is trying to navigate through that content,   they get very lost and they can't find that  specific section that might be in Chapter 13   of that book. So the heading structure is  what will allow them to quickly find that.  Now, headings can be set very quickly in  Equidox. You can see over here that this   is a text zone by default, but if I hit  the dropdown menu, I can select Heading,   and it will change to a Heading Level One.  However, I prefer to use keyboard shortcuts.   So when I want to set a Heading Level One, I just  tap one on my keyboard, and it instantly changes   to a Heading Level One. I'm going to set this as  an H2 and I'll set these as H3s. I'm just hitting   the corresponding number on my keyboard, and what  we'll see changing in the HTML preview is the   font size now gives us that visual confirmation  that we've actually set that element as an H1,   and these are—that's an H2, and then we obviously  have our H3s here with the bold font that kind of   stands out compared to the regular text that  we have. We also have a couple of lists on   this page. So these lists right now, they are not  tagged as lists, but you can understand that list   structure is very important because all of these  list items have a relationship with each other,   and they are nested inside of another  list item, which also has a relationship   to the list item it is partnered with. So, the structure of a list is critical. If   you don't tag it as a list, it will just read  these elements as kind of like strange run-on   sentences with no punctuation. You can imagine  the confusion that could cause. So we want to   make sure that we're handling our lists properly. Now, by default, it's a text zone, but all I have   to do is hit "L" on my keyboard and nudge my list  detection slider. You'll see a theme of sliders   today to make this a lot easier and faster.  Instantly, I've detected those five list items,   and if I go back to my preview, I will see  the proof of that. Instead of having just a   big weird sentence, I now have five distinct list  items that are delineated by these bullet points.  This is a nested list, which is exponentially  more complicated to tag manually in Adobe Acrobat.   You'll hear horror stories of Acrobat users  fighting with nested lists because there are   so many tags that you have to build, and they all  have to be placed in the correct order. There's a   million ways to do it wrong and only one way to  do it right, so it's very tedious and slow. But   with Equidox, we just hit "L" and nudge our  slider, letting the artificial intelligence   quickly figure out that okay, that's a list, and  inside of that first list, we have another list,   and then we even have a third layer to that list.  So very quickly, if we go and look at our preview,   we're able to build that nested list structure,  and all of those tags will be automatically   produced for us when we export the PDF. We also have a couple of images on this page. Now,   images, of course, need a description or they  need to be artifacted. For this image here,   it's the first iteration of the logo. As a  general rule of thumb, I might give a logo   a description one time in a document. If this  logo keeps repeating itself throughout, I'm   not going to keep describing it. That's just too  much work to keep typing in the same alt texts,   but also a screen reader user doesn't need to hear  it page after page. They don't need to be reminded   that there's a logo there. So I'll simply type in  "Equid logo" for now, and that will give me my ALT   description that's required for that image. Then down here, we have another image where,   again, it's another iteration of our logo,  but this one has a dog in the picture. This   is really just kind of a decorative image; it's  just there for visual aesthetics. It's not really   adding any new or unique information to the PDF  itself. So if you want to get rid of an image,   you can just hit backspace on your keyboard. The  zone itself will disappear. We're not visually   removing the image from the document; we're just  making sure that it's artifacted so that a screen   reader will pass right over it instead of stopping  to read just redundant alt text. And again, in the   preview, that image is now gone, and we're just  left with the text-based content with the lists,   the headings, and of course, the image  at the very top of the reading order.  Okay, so we'll jump over to the next page  here. On this page, we have a couple of   tables. Tables are another one of those things  in PDFs that are extremely time-consuming,   but I'll show you how quickly we can handle  tables inside of Equidox. All I'm going to   do is put a zone by clicking and dragging around  the table. I'm going to hit "T" on the keyboard to   change it to a table and I'll go into the table  editor. Blink and you miss it, but if I just   nudge these detection sliders back and forth,  you can see that it will instantly figure out   the structure of this table. All of my cells have  been identified, and all I now have to do is press   this checkbox to give myself a table summary. So when I go back to my PDF, all of those little   text zones that were there before have now been  replaced by a single table zone. And if I go into   my preview, we will see that we have a nice  clean HTML table compared to the other table,   which we haven't yet addressed. You can imagine  how useless this would be to hear through a   screen reader. None of these numbers would make  any sense if they're not structured as a table.  Again, we will solve this table very easily by  just drawing a zone, hitting "T" on our keyboard,   and going into the table editor.  Again, if you blink, you'll miss it,   but you just have to nudge the sliders  to wake up the artificial intelligence,   and all of the cells have been identified. The only other thing with this table—well,   there are two small things. The years 2024 and  2025 are actually column headers that straddle   four separate columns, so we just want to make  sure that we are spanning by holding shift on   our keyboard and selecting above Q1, and just  holding shift and then selecting above Q4. We   can then either hit the span button or  just press "S" on our keyboard. Again,   I'll do the same thing here for 2025. So what that allows for is for the year to   just be a column header that has four sub-columns  underneath it. The other thing is that this table   is somewhat unique in that it actually has a  second level of headers. So not only is 2024 a   column header, but also the individual quarters  within the year need to be set as headers as   well. All we have to do is hit the up arrow  to change the column header from one to two,   and then this will indicate to Equidox you  want the top two rows to be set as your column   headers. And here you can see that bold font  that reflects that change that you've just made.  So, the logic here is that if a user is inside  this data cell, they are aware that they’re in   the sales row for the year 2024, and more  specifically, they’re in the fourth quarter   of 2024. All of that structure is built for them  so that they can freely navigate this table and   always understand exactly which columns  and rows they’re associated with. Again,   hit the checkbox here, and Equidox will  programmatically write a table summary   for you. It’s one less thing for you to do. If I save the table and close out, I'm left with   the single table zone. I realize we are right  up against time here. I haven’t covered form   fields in detail, but at a high level, for those  of you dealing with form documents, the main thing   is to make sure that you're adding tooltips. I can  enter tooltips here, such as “Enter today’s date,”   “Enter first name,” “Enter last name,” “Enter date  of birth,” and “Select if US citizen” or “Select   if non-US citizen.” Adding tooltips is  as simple as adding an ALT description,   but you can build these tooltips really quickly. Now, our page is more or less handled. We could   probably make some small adjustments to  the reading order of these form fields,   but in the interest of time, we can always just  press the reorder button to consolidate things,   and that will correct most issues. So, let’s back out of the document and   go to the Export tab. All I need to do is hit  “Generate PDF,” and Equidox will produce this   brand-new document for me. It will be the  exact same document that we started with,   so you’ll see nothing has changed about our  PDF—it’s still the same. However, everything we   built in Equidox, and what we validated with the  HTML preview, is going to appear in the tag tree.  If I download this document and save it to  my desktop with a suffix to indicate it’s the   remediated version, I can open it in Adobe  Acrobat. The original document was completely   untagged, but the one we just built has everything  tagged—our lists, tables, form fields, and more.   This document will pass the accessibility checker  if that’s your goal. You’ll only see two warnings:   one about reading order and another about color  contrast. These warnings are not issues but   notifications that you need to handle reading  order and color contrast in a different tool.  That said, I know that was a lot of information,  and we’d be more than happy to discuss Equidox   as a solution with you and your organization on  a more one-to-one basis. If you have any sample   documents or questions about the tool, please  feel free to reach out. You can email us at   EquidoxSales@Equidox.co, call us at 216-529-3030,  or visit our website at www.Equidox.co.   We’re also on various social media platforms. Thank you all for attending today. I apologize   for being long-winded and going a few  minutes over, but we appreciate your   participation. Please feel free to reach  out if you have any questions at all.