 Okay, so I mean, it's my it's me people seem to be coming in so I wasn't sure if we wanted to give it another minute Yeah, I would say give it another minute and then people still dribble in but it's up to you from now may you're on Watching for Kelly and I know I feel like it's week. We at least have to have all the presenters here before we start Kelly has just arrived I see Hello This is making a lot more sense now, so Kelly dropped in on our stand-up meeting earlier I saw a quick entry and a quick exit and I thought huh. I wonder who that was. I was just trying to make sure I could actually log on Perfect. There's Hadi as well. I think we all want to share screen, right? So we probably need to be promoted to Yeah, Matt are you taking care of that or Jean? So there's there's no promotion since this is just a standard zoom account So there's not a webinar for instance. You should have the ability but Tara Why don't you just give it a give it a try screen share and then we'll make sure it works Everyone joining us. We're just doing a quick check-in and then we'll get started As soon as we hit 11 I saw a flood of people so we decided to be given another minute. You're my entire desk That worked Okay, I think I'll go ahead and get started and so good morning everyone My name is Maggie Faber and I use she her pronouns I am the assessment and data visualization librarian here at UW and I am also a member of the Tableau user group steering committee I'm really excited for today's topic on accessibility This is a topic that I am personally interested in But more importantly, it was a topic that was suggested by a tug attendee in one of our post-meeting surveys This person wrote I am increasingly concerned about the accessibility of our visits because 99.9% of ours are for public consumption would be great to go over how to create accessible visits to be compatible with screen readers, etc And so we saw this comment and we were really excited to put together this session Just a note that we will be sending out another survey towards the end of this meeting So please know that we the steering committee are listening and following up on all of your suggestions We really wanted to devote some time to accessibility and improve our collective skills in this area So we expanded this to a 90-minute session today to give our speakers a substantial amount of time to Talk through the content Though you are welcome to leave if you have another meeting or to eat lunch or take a break and walk around as needed The session is being recorded so it will be posted to our canvas site if there's anything that you missed or want to catch up on or Want to share with anyone you work with And then I also just briefly wanted to mention that there are other opportunities to engage in this topic or expand your learning Our April tug session will be a show and tell with two dashboards that have spent a lot of time on Accessibility remediation, which is something that you'll be learning a little bit more about in this session today In May my colleague Nagin will be teaching another workshop on data visualization accessibility It won't be focused on Tableau access Tableau specifically But her last workshop had over 70 attendees a number of which were from this group So I think it's going to just get better and better And finally there is the accessibility the IT accessibility liaison teams that are here at UW They have an active listserv and training opportunities to learn more about accessibility broadly not just through the lens of data viz I am just dropping that link in the chat if you want to learn more Um, and so anyway without further ado. I am going to introduce our speakers for today We have Terrell Thompson and Hattie Rangan with UW IT accessible technology services and Kelly Gupton The product management senior director at Tableau software The three of them will be leading us in an exploration of how to ensure our data visualizations are accessible to people with disabilities If you have questions throughout the presentation, please drop them in the chat the presenters will address those as as It's possible in their presentation and I will be keeping track of any that are missed So we'll do a roundup at the end And if we run out of time in the presentation, even with the extra 30 minutes, we will follow up in an email or campus announcement Thank you all so much for being here and thank you, Terrell, Hattie and Kelly for taking the time to speak with us today. Thank you. You're welcome and thanks. Thank you, Maggie. I appreciate the invitation and this actually came on our radar. We being Hattie and I Are members of the IT accessibility team. I've managed that team, which is part of the UW IT accessible technology services group and In that role, we provide consultation and support on all things related to IT accessibility. So web accessibility documents, videos, software, we get involved in procurement a lot. So our our charge is to ensure that technology is accessible across all three campuses of the university. And we're using a lot of technology, obviously. And so that covers a lot of ground. And it's impossible for us to do that alone. There's so much technology and so many people involved that we really depend on you. And so a lot of our time has just spent providing support and consultation and education. To kind of help you, the campus to, you know, make decisions that have a positive effect on accessibility of technology and and to use the tools that are available to you in the best way possible. So anyway, in that role as we provide, you know, consultation on website accessibility and so forth, we've got a number of people come to us over the last, maybe six months or so, or year with questions about accessibility of their websites, where Tableau was involved. And so, so that really got us interested in exploring this a little bit further. I confess that I am just just getting my feet wet with Tableau behind the scenes. So, you know, I've experimented on the front end, you know, with what the user experiences like but I really don't know much about what's going on behind the scenes. So it actually has been really helpful for me to to interact with Kelly over the last couple of weeks as we've been preparing for this we have met three occasions I think now and looked at some dashboards together and and I feel like I'm developing a better understanding of how Tableau operates as a result of that so it's great that this, you know, this invitation to present has sort of kicked off this collaboration. So anyway I want to start with accessibility. In general just kind of very broadly looking at this what are we even talking about when we talk about accessibility. We're really talking about access to information. We've got some sort of digital information and you need to access that or interact with that. And kind of the traditional model is input that involves a keyboard and later a mouse, but those two often sort of, you know, combined to interact and put into a system where you can get data out. And that output then is often through a monitor, at least that has been the traditional model so this traditional model of monitor keyboard mouse is how things have sort of operated for a long time. But there, that really is only a small piece of the of the picture. And in fact it looks a little and indicated as we look at this now. There are lots of different devices people are using. Now this is an older slide says get some older models of cell phones but but you know people on handheld devices, lots of different shapes and sizes different platforms are interacting with digital devices on both input and output. We also have tablets a variety of shapes sizes and platforms. The choice of device has an influence on what the person actually perceives and what they're able to do and how they operate with it and interact with it. And we also have people who are not getting information visually, but are getting information audibly, and that as devices evolve. This is true and possible for everyone now. Historically has been the domain of people who have visual impairments or who are blind that they're using a screen reader, which audibly reads the content of the of the screen or provides the digital information. Using a synthesized voice and that that is how they get their output. They may still be using the keyboard probably not using mouse, but for output, they're getting that audibly. You also have people who are providing information providing input verbally so not using the keyboard or the mouse but using voice. This too is expanding now so lots of people are doing this, you know with a hands free sort of interaction that's possible on today's devices. But again historically this was the realm of people with disabilities who physically are unable to use a mouse or keyboard maybe they have limited mobility, maybe they have no hands at all. Each input is one viable way, among many others that they have for providing input into the system. They also have people who operate with a sense of touch. So screen readers as I mentioned for people who are blind can deliver their output audibly so they're listening to content. But they can also deliver content through a braille device which is what we see here on the slide. This is a refreshable braille device where you've got a row of braille dots that changes as a person explores and navigates through the computer. And so, so they can explore that way and this this particular braille device also has braille keys for input so somebody could do all of their interacting both input and output through this one device. Now I ran out of space on the slide. Otherwise I could have gone on and on and on about different ways that people use the computer. You know, even visually we have very different needs different screen resolutions different font sizes the browser allows us to change our font size and so that's going to impact our experience. So, bottom line is there's a lot of diversity in the way people interact with information. So it's not just, you know, this old school model of a monitor and keyboard and a mouse that's a great oversimplification that there's a lot more variety than that. Also, I want to point out what what you see now, assuming you have eyesight on the screen is a data visualization that, you know, this is all provided here so that you can better understand the landscape. But I also have been describing it all the while I told a little story here. So I just put up this slide, and I said nothing. Then, some of you might be able to sort of figure out what I'm getting at, but the deeper message of what this, what am I trying to communicate here exactly the deeper message probably would have been lost. And so, so this is a data visualization, but it's accompanied by description and I've done my best to sort of make this accessible for everyone. Here's another visualization this is ability on a continuum, we've got people who are able to do something people are not able to do something. And where that where a person falls on this continuum varies depending on what is we're measuring so maybe it's the ability to see some people have 2020 vision and don't require any assistive technology like glasses in order to attain that. Some of us fall further down on the spectrum, but with glasses, we move a little bit further up. Some people are not able to see it all. And most people sort of fall somewhere on that continuum it's not a binary thing where you've got people disabilities people without disabilities, you just got a lot of variety of human experiences depending on what it is you're measuring. Same thing with ability to hear ability to walk ability to read print ability to write with pen or pencil ability to communicate verbally ability to tune out this traction. All of these are variables that are extremely variable so some people are outstanding at their ability to do these things. Others are unable to do them at all, and most people fall somewhere in between. Also, there's the ability to process and understand numerical or tabular data that if we look at a large data set presented in a table. Some people can can handle that and they can understand the relationships between the parts, particularly if they can interact with that table and can sort the data and filter it. Then they work really well with that medium, but some people don't do so well with large data sets. And this is why we have data visualization that in a sense it is in order to overcome people's inherent limitation when it comes to being able to process and understand numerical and tabular data. So for those people. There, there is also there are data visualizations there's presenting this data and using graphs charts, you know, presenting in a visual way that helps some people then to better understand it. But that doesn't work for everyone and so again, lots of variety, some people do better with different types of information than with others and we're sort of all over the map when we measure our strengths and weaknesses in this regard. So I want to talk just in the way of overview about accessibility and I do want to talk about the law, I don't like to talk about the law much and I'm not a lawyer. So this is all from the perspective of an IT accessibility person. But, but we do have laws that we are required to meet and so this is an important part of the conversation. Specifically for us in higher education at a public university, we've got two laws that are sort of on the forefront. Section 504 of the rehabilitation act that dates all the way back to 1973. And then there have been things with disabilities act was 1990. And then there have been amendments since then, but basically at the heart of both of these laws is civil rights that prohibit discrimination against people with disabilities. So back in 73 certainly and even in 1990 technology was nowhere near what it is today, but it was just in general, all of our programs and services and resources must be accessible to individuals with disabilities, or we are discriminating against those individuals with disabilities. And so the difference in the laws is sort of a difference in scope, primarily that the the rehab act that applied to recipients of federal funding. And so that covers us, but Americans with Disabilities Act expanded that so it was more about society as a whole not just recipients of federal funding, but nevertheless it still applies to us so under both of these laws. We need to ensure that our programs and services and activities and resources are accessible. There, there have been hundreds of legal complaints filed against higher education institutions for having inaccessible it. And those have increased, if we were to actually show visualization of this there'd be a line that goes up dramatically over the last five to 10 years. And certainly those are complaints filed with the US Department Office of Civil Rights, and, and then there's some negotiation that happens and then there's a resolution agreement or a settlement that happens in which the university agrees to certain conditions like having a high level it accessibility policy, and various other things. Always, I think almost 100% of the time, maybe even 100% of the time, there is a requirement that the institution named in that case, comply with WCAG two dots, either 2.0 or 2.1 2.1 is the most recent version level to a. I'm talking a moment about what that is but just keep that in mind, WCAG 2.x level to a is our expected level of accessibility compliance. That message is reinforced over and over and over and over again in these legal complaints. Also, Washington states has a policy 188. This actually is an IT policy so it comes out of the CIO's office. But that says that all state agencies including higher education institutions are required to meet WCAG 2.1 level to a. So, so once again, I mean it's just reinforcing what has emerged from legal cases under federal law, but the state is trying to be proactive and just trying to raise awareness that you know this is a requirement US state agencies need to pay attention to it. In these OCR Department of Education resolutions, this kind of a template that they use and so we see the same language over and over again and their their resolution agreements, and that includes a definition of accessible, which I'll read here, it says accessible means a person with a disability is afforded the opportunity to acquire the same information, engage in the same interactions and enjoy the same services as a person without a disability in an equally effective and equally integrated manner with substantially equivalent ease of use. The person with a disability must be able to obtain the information as fully equally and independently as a person without a disability. So, and I highlighted, you know, this section of that the whole thing really sort of is equally important, but but it felt like I really wanted to stress in an equally effective and equally integrated manner with substantial equivalent ease of use, as we explored data visualizations. So this is a person who has no eyesight, or who is not able to use a mouse. You should be able to fully access the data. So, what is look at to 2.0 or 2.1 mentioned this a few times now this is our requirement this is what we are striving to meet. So this is an aspirational goal. You know, I'll be frank about that that we're far from meeting it, but we, we really need to demonstrate that we are making progress. It is the web content accessibility guidelines, affectionately called WCAG by its acronym is an international web accessibility standard published by the worldwide web consortium or W3C so that's the same group that brings us HTML and CSS. There are lots of other standards related to the web. The first version was published actually in the early days of the web in the 90s they began early 90s they began working on this. You know, soon after the web was invented by Tim Berners-Lee and soon after the W3C was formed. They began to work on accessibility. They published the first guidelines in 1998. And since then those guidelines have been updated twice 2.0 in 2008 and 2.1 is the latest version published in 2018. And they're continuing now actually to work on on more versions but 2.1 is the one that our policies are based on at this point. At its deepest level, it's divided into success criteria so these are the specific measurable things that say this is how we define web accessibility. And there are 78 of those so there are quite a few very specific things that must be done in order to ensure that websites are accessible, and they apply to a lot of other digital technologies as well. But one thing that a lot of people are familiar with is images on alt text or alt text on images. And so you've got an image. You know, maybe a logo University of Washington logo as an image a screen reader can't make any sense of that. So we put we add alt text to it it's text that is behind the scenes and the code that says University of Washington or maybe University of Washington logo. That then makes that image accessible to somebody who's using the screen reader. And so that that's critical. And it is a one of the success criteria of we'll get 2.1. And it's a level a success criteria because it's so important that without that alt text person using a stranger has no access to the content of that image. So the success criteria the 78 specific measurable things, each has a level associated with it so level a or the highest priority level to a or the second highest priority and level three a, or if you want maximum accessibility and you want to really ensure that absolutely everybody has, you know, equal access to your website then you go for level three a all 78 success criteria. I mentioned though the the bar has been set at level to a that we are expected to meet level a and level to a so that's a subset of 50 success criteria from from the work. So what what we how we do accessibility it accessibility at the UW, just so you sort of know how this all works here internally and where your supports lie. So here first of all there is an ADA compliance officer, Brie Callahan is that person currently and residing in the office of compliance. So she's kind of on a mission just to make sure that we are compliant and raising red flags whenever you know there are seem to be big, you know risks that people are taking related to accessibility. So they have disability resources for students. DRS, they provide individual accommodations to students with disability so student with a disability registers with their office has documentation of their disability and DRS then provides them the accommodations they need. Disability Services Office DSO does the same sort of thing for employees and for the public. So where we fit into that UW it accessible technology services works to support the UW community community as it strives to ensure it's it resources are accessible and so we're working sort of more proactively to build a fully accessible it infrastructure. So as these other groups on campus are working more so reactively to meet individual needs. We want to try to make things accessible by default, so that you know DRS and DSO and the ADA compliance officer don't have to do so much because retrofitting and making individual often is not not great it's you know last minute it's really hard to fix accessible software for example, you know, for an individual, if the software is inaccessible and so. So we're trying to tackle it up front, you know, let's make that software as accessible as possible. So our website. I didn't actually see, I'm not looking at chat now so I didn't see what was pasted in earlier but are the hub for our communicating with the campus community about it accessibility issues is UW.edu slash accessibility. So you can find out more about all the things we do there. And, and we do, as I mentioned provide free consulting and support to the community on in all these different areas if it has a user interface, then it has the potential to direct barriers for certain groups of people and so we want to make sure that anything with user interface is in fact usable by people with disabilities. So, so I want to then shift and we've talked about sort of broadly, you know all these different ways that people interact with technology. And the, the fact that we need, we have legal requirements that need to be met and their standards that we're striving to comply with that sort of define our legal responsibilities. But I want to shift them to applying all that information to data visualization itself. So, there are two purposes, you know as I see it and again I'm a, I'm a novice when it comes to data vis but I've been exploring this quite a bit over the last month. And it seems that the two purposes are either explanatory to make data easier to understand to tell a story, or exploratory to enable users to explore data and identify patterns and trends. So I've kind of just like the screenshot of an example that is from the accessibility world that there's an organization called WebAIM at Utah State University that does a every couple of years they do a survey of screen reader users to identify kind of trends and what tools they're using what browsers they're using and all sorts of other things about screen reader use. So this is from their results website, which is kind of a breakdown of the most popular screen readers, or all the screen readers are out there and, you know, kind of a market share visualization and so you've got in VDA and jaws in windows that have the largest chunk of the pie here. And so visibly I can see that that in VDA actually is the market leader at 40.6% and jaws is second close behind at 40.1%. Those have by far the largest chunks of the pie. And then there's another pretty good size wedge at 1212.9% voiceover. Everything else is just a tiny little slice of the pie. So, so I see that visually. But if I'm a person that processes information better through a data table. They also provide a data table. And so I can, I can explore the data that way. So, so this is a good example of providing a visualization that communicates the, the idea in different ways because different people access information in different ways. So what we want to do now that you've gotten that background is provide a few demos. And I'm going to take just an initial look at those really quickly just to sort of walk through what I see as a sighted person who can use a mouse. I also want to explore, are these accessible or how can we maximize their accessibility. Since this is the Tableau user group, we want to focus on Tableau to an extent, and Kelly is here to, to provide us that insight, you know, as the Tableau rep. And share, you know, strategies for maximizing accessibility of a Tableau dashboard. There are lots of things you can do to ensure your Tableau dashboard is accessible as it possibly can be. So the are going to explore an alternative example using high charts, just to kind of, you know, show what's what's possible and sort of contrast visualizations that were prepared were prepared with different tools. My colleague Hadi Rangan from Accessible Technology Services is a screen reader user and so he's going to give us some insights on what that experience is like as a screen reader user accessing these various demos. So, with that, let me just quickly show you the visualizations that we want to look at. I've got a browser here, this is a webpage that I created that we're going to start with. And one thing I want to point out I've got a tool here, a bookmarklet that says headings this is an accessibility bookmarklet it's a tool I use for checking accessibility. If I select that, then it identifies the HTML headings on the page it shows them clearly, you know that data visualization accessibility is an H1, then I've got three H2s that correspond with the different sections of the page. So this is a page that using HTML is very clear in the heading structure forms an outline of the page content. It's a really important concept when it comes to accessibility for web pages and PDFs and Word docs and everything. If this document has some heading structure, that heading structure needs to be communicated to users with disabilities and so Hadi later when he does a demo you'll see why this is particularly important. And from here, the examples I want to show that we want to show. First of all, there's this dashboard from the UW libraries, this is a Tableau dashboard. UW libraries profile by fiscal year. The things that jump out of me here are the quick figures first of all so you've got what I understand from Kelly are called big ass numbers bands. The numbers visually really jump out at the top of the page, but I also have individual visualizations that I can hone in on one that jumped out at me a circular circulation that just decided to refresh that I see that circulation has declined over the years. This is for all libraries and campuses, and I may be curious to know whether that is a reflection of Seattle more than the other campuses, or what exactly is going on there. And I can use that to change the visualization. So there's the Seattle campus libraries, it looks pretty much like the all libraries line. But if I select Tacoma, then I see a couple of interesting changes here to the overall pattern that there actually was an upward trend in the early days from 2013 to 2014. And then, more recently, another upward trend between 2018 and 2019. So I'm going to be curious to know, you know, what's going on there at Tacoma, but very quickly, you know, in 30 seconds to a minute here, I've been able to hone in on that and visually, and as a mouse user was able to access that interesting bit of information. So, so the question that is, can somebody that can't use a mouse or that can't see it, can they to access that we also are going to look at this is the high charts demo on population of seven Nordic countries, because I think the high charts is based in Norway so that's where we got this from. But, you know, there are lots of different ways that this data is presented so the population is color coded within this map I can see that Sweden is the highest because it's red. And I can see that Norway Denmark and Finland are a darker shade of purple, then Greenland and Iceland and the Faroe Islands. So therefore, you know, I get some relationships based on the color coding in the map. But all that data is also presented in this table beneath the map sorted by population so really easy to figure out. You know, which countries have the most population. And if I am interested in living somewhere that is both not too densely populated but also not growing. Then, you know, I may want to look at population history and see like Sweden is on an upward trend. But what about like Finland Finland I see kind of flattens out a little bit and so that's an interesting piece of information to be able to interact with this by just like on the phrase dashboard, selecting different pieces of information, and then, you know, following getting you know a change of information based on my selection. So, also, if we have time, there's also a sonification example that I would like to explore. This may carry over into that last 30 minutes but and that can be a significant way to expose people who can't see to data trends and high charts is able to do that. So, this may be the only opportunity we get to look at these two UW dashboards. But again, come back in April and you'll get both of these are actually going to be featured and so you'll learn more about what they did. But the fast facts. There's a lot of information here in this tableau dashboard, but as you'll see as we explore, both through Kelly's lens and hottie's lens. Tableau dashboards. There are some limitations there are things you can do to ensure this is as accessible as possible, but as accessible as possible is not perfectly accessible and so they're going to be some things that really need to be done through an alternative version and fast facts has provided actually got three versions to get the tableau version as well as a PDF version for printing and this text only version, which is not just text text with headings, with data tables that provides access to all of the same information but organized in a way that's accessible. It's somebody that can't access the tableau dashboard very easily. And the same thing. They actually did that. There was a pretty easy process I understand from them. Because it's static data, whereas the cove dashboard is dynamic this is changing all the time. I think they have daily updates. And so this was a bit more of an effort, which they can again they can talk about this in April, but you can once again view a text only version of the dashboard. And this is being generated programmatically from the same data source so they've got a Google sheet as the data source and PHP scripts that are drawing the data from that original data source and generating HTML tables from that data source. So they are dynamically creating an alternative version, but this is then a, you know, and more accessible presentation that accompanies the tableau version so so you can toggle back and forth with both of these examples between the two two versions. So, so I don't think we're going to look at these, you know, text only versions, but other than what I've just shown you, but the bottom line is, you know, you may be faced with, if you're your dashboard is not otherwise it going to be going to present some accessibility problems, then you need to be thinking about alternative versions as a means of providing access to to the information. So that said, I'm going to stop sharing and hand it off to Kelly to talk specifically about Tableau and what you can do to ensure your Tableau dashboard is as accessible as possible. Right. Thank you so much, Cheryl. Okay, can I get a confirmation that you can see my slides. Yep, we can see him. Great. Hello everybody. My name is Kelly Gupton. I'm a product management senior director in the development team at Tableau. And I'm going to be talking today about basically best practices for Tableau dashboard accessibility. Cheryl gave a great overview of the overall subject of accessibility, including the web content accessibility guidelines or WCAG. And what I've come up with are a set of best practices that basically help guide you to be able to apply the basic ideas and principles inherent in these WCAG guidelines in the context of Tableau dashboards. And as you noted, it doesn't create a perfectly accessible dashboard, but it does help you deal with a lot of the core accessibility issues that are in Tableau dashboards and it helps you achieve the best possible result. Creating accessible dashboards is not something people really try to do all that often in the Tableau world. There's been a lot more than over the last several years, but it's still a thing that is pretty few and far between. And this is really about achieving the best possible result that there is. So let's go ahead and get into these. So I have 10 best practices. I'm just going to go through each of these in detail. And they start really with kind of a more of a philosophical point. And Carol actually did mention this and talk about the difficulty of retrofitting accessibility into existing software. So the number one thing is to design for accessibility upfront. It is a design choice to design for accessibility and accessibility should be considered a requirement for what you're creating, just like any other requirement that you might have. So if you're thinking about the requirements of a dashboard, you're thinking about, you know, what is, who's the audience, what information are you trying to communicate, all of those sorts of things, and accessibility should be one of those requirements upfront. And part of the reason for this is, is that you can't really make any arbitrary dashboard accessible. There are some patterns and ways that people tend to use Tableau that just really don't work well for accessibility. In particular, a lot of the kind of fancy ways that people lay out dashboards in order to achieve a certain visual result, don't really work that well with assistive technology. And that's because assistive technology relies on the semantics of the underlying web page or dashboard. And when you try to assemble Tableau dashboards in a fancy way, you kind of do this in a very semantic free way that makes it difficult for assistive technology to work. So, so number one, make sure that you're designing for accessibility from the beginning and considering it upfront. And the second is to make sure that you're using the elements in the dashboard that we have worked to make accessible. So there are a lot of different kinds of objects that you can place in the dashboard. And we've, we've been on a accessibility journey for a few years now, and we're kind of working on the various objects that you can put in a dashboard to make them accessible over time. In the latest version of Tableau and this is true for about the last three versions of Tableau. And just so you know the latest version is 2021.1 we actually just realized on Friday. But overall, the things that you can put in an accessible dashboard are the tabs for a workbook, the titles of the dashboards views themselves. In particular, the title of the view, the caption for a view. And most importantly, what's called the view data window. So this is a window that you can open for any view and any dashboard that gives you an accessible table of the data that's underlying the visualization. And this is key for providing access to that data for users of assistive technology. There are also a few filter types that we've worked on the list and drop down list filters as well as the relative date filter. Categorical legends, the objects that you can put in dashboards like text objects, web page objects, images, buttons, etc. And then some of the other controls like parameters, set controls and the data highlighter. I'm going to end the presentation with pointing you at some resources online and where you can find a wealth of resources about accessibility, including this list that we keep maintained. So people know what all is out there that we've worked on. The third thing is to make sure that you have allowed the data behind the dashboard to be downloaded. The reason is is that this permission controls whether the view data page is enabled for any dashboard. The view data page again is how a assistive technology user is able to get access to the data behind the visualization. And since it's an HTML page, you can actually just copy and paste the data right out of that page and put it into another document. And that's why there's a permissions association between the ability to download the data and the ability to open the view data window. On Tableau Public, where most all of your visualizations that we've been looking at are stored. The option is basically to allow the workbook and its data to be downloaded. And so that option needs to be enabled in order for the view data window to be functional. Okay, now getting into some more specific techniques within Tableau. So the first thing is for any of the objects that are in your dashboard, you need to show the titles, and these titles should be useful they should be explanatory as to what the object is that's being titled. One of my titles are so important to show is that they actually show up to assistive technology as HTML headers, and as Terrell showed earlier, headers are a key way that assistive technology users use to navigate web pages. So in Tableau, if you've enabled the dashboard title, that will show up as a level one header. And the title for any views or visualizations in the dashboard will show up as a level two header. And then the titles for any filters, parameters, sets or legends will show up as level three header. And as our demo later will show screen readers and other assistive technology have keeps throw combinations and such which allow users to jump to various headers list off the headers in the page and so on and so forth. So good, clear, visible titles are essential in order to enable that sort of navigation for Tableau dashboards. The next thing to do is to show captions that describe the views in your dashboard. So if a caption is visible on the dashboard, a assistive technology like a screen reader will be able to read that caption. And captions in Tableau can actually even have some dynamic elements. And so what that means is for data that's in the visualization, you can actually put some of that data in the text of a caption and have that be dynamic. So if the data changes to caption will change. I'm going to have an image of a histogram from a dashboard that I created about SAT scores of students at a hypothetical university. In this particular case, the title of this view is called is number of students by SAT score. And I've created a caption that basically just gives a high level overview of what this, this chart contains what data the chart contains. The caption for this particular chart reads, number of students by SAT score is a histogram showing number of students by SAT score. SAT score is on the x axis, number of students is on the y axis, blue squares represent female students, orange square, orange circles represent male students, the reference line displays the average SAT score. And of course if I wanted to I could have put some data in there as well. But this, what will happen is if a screen reader user for example is interacting with this Tableau dashboard and they bring the keyboard focus to a visual this visualization on the dashboard. The screen reader will first read the title of this visualization, then it'll read the caption of the visualization, and then it'll give the keystroke combination for opening the view data window. And so the caption is very useful to provide an overview for a user so that they can determine whether or not the view data window or opening the view data window is something that they might want to do to examine the data in detail. I did mention using the keyboard to navigate the dashboard. So the next thing to do is to set a logical key focus order for the dashboard so that as a user moves around the dashboard using say the tab key or if they have assistive technology enabled arrow keys that things move in a logical way, typically kind of a top to bottom left to right type navigation. Unfortunately in Tableau today, setting that up is not typically or not very easy thing to do. The default keyboard focus order is a bit nonsensical in that it's determined by literally the order in which you drop items into a dashboard. Which is not typically from top to bottom and left to right. Right now we have a work around for doing that that involves editing the workbook XML for the workbook that the dashboard is in. And there's a topic on the Tableau community forums that I wrote basically giving instructions on how to edit the workbook XML in order to set the focus order correctly. This is something that we're working on. We're looking to ship later this year so that dashboards will by default have a logical top to bottom left to right focus order. Next is to make sure that you're using text colors with sufficient contrast. So this is one of the guidelines in the WCAG is to make sure that there's enough contrast between the foreground and the background of the text color that everyone who is, or people with issues discerning contrast are still able to read that text. So on the left of the slide I have the word accessibility printed in a light gray. This actually does not meet the standards. And on the right, it's a much darker gray that does meet the standards. There's a very useful tool out there called the color contrast analyzer, which basically allows you to sample the color of the foreground and background, and it will tell you whether or not the difference meets the standards for the WCAG, and it's quite useful. Another thing to do is in your visualizations, make sure not to use only color to convey information. So Tableau does have a color blind color palette, but it's specifically designed to account for red green color blindness, which while is the most common type of color vision deficiency, it's not the only type of color vision deficiency. And so if you're attempting to convey information and you convey it in a way that uses color only, then people with various color vision deficiencies may have difficulty understanding the chart that you're displaying. And so there are a number of different techniques that you can use and it's very dependent on the type of visualization. You might choose to use shapes rather than colors. You might use labels on top of bars in a bar chart, for example. But what I have here is an image of a line chart, where I have the line chart basically has different categories that are mapped to different colors. So you have orange lines and blue lines and gray lines. But, you know, if I left it that way, it would use only color to distinguish those categories. So what I did here is I used a dual axis chart in Tableau and on the secondary axis, basically plotted the same data using shape marks, which basically results in those shapes being overlaid on top of the lines. So that way someone could distinguish the lines by the lines with squares or the lines with triangles, lines with circles and so forth, so that we have an alternate method besides color to convey that information. Number nine is there are some of the objects that you can put on a Tableau dashboard allow you to set different text properties and these text properties can be read by assistive technology. So for image objects, on the left, I have a screenshot of the edit image object dialogue. So for images that you've put on a dashboard, there is an alt text field. And that alt text field basically allows you to intertext that a assistive technology device or screen reader will read when a user navigates using a keystroke combination to that image on the dashboard. For buttons, and this is both for navigation and for export buttons, there is a tooltip field, and that tooltip text is picked up by assistive technology to help a user understand what pressing that button will do or activating that button, which they can do using the keyboard. And then the last tip that I have here is to just use text objects on the dashboard to give instructions and context for that dashboard. It's usually useful for everyone to help give an overview of what information is in that dashboard, and it's particularly useful for users of assistive technology to help them orient themselves as to what might be the various filters in the dashboard, what views are in the dashboard, what data they contain and so forth. And you can even provide some more instructions for how to use assistive technology with those objects in the dashboard. And of course, assistive technology can read the text that's in a text object. And then finally, there are a couple of new guidelines in the WCAG 2.1, which are relevant to tablet dashboards. The first is that this version of the WCAG introduces an additional contrast requirement around non text contrast. So this is for user interface objects, marks and a data visualization, and so forth. Now, this is a bit of a tricky one to handle. It's quite difficult to create a complex data visualization that meets all of the various color contrast requirements along with the use of color requirements and so forth. So we need some more creative ways to help individual users, end users customize visualizations to meet their own needs. But for now, an enterprising member of the Tableau community created a set of colorblind custom color palettes that you can find on our community forum. It's a various sets of color palettes to adapt for different types of color vision deficiency, but all of them actually do meet the WCAG 2.1 non-text contrast guidelines so they can be useful for this. And then finally, the WCAG 2.1 also introduces new requirements around reflow to make sure that user interfaces can scale appropriately when a browser is zoomed in up to 400%. So zoom is very common for people with low vision to help them be able to see and discern the information in a web page, and a web page or anything in the web page should adapt correctly for that. And just a week before last, I published a blog post on our community about how to use the existing Tableau functionality around resizing layout containers and alternative dashboard layouts or device specific dashboard layouts in order to meet the WCAG 2.1 reflow criteria. It goes into a fair amount of detail about how to do this as it can be kind of tricky to make a dashboard scale and resize and relay out properly. But it's quite possible to do this for, you know, the kinds of dashboards that people typically create. Okay, so that's the end of the tips. I do want to point everyone at a very important page on our community forums called the Accessibility FAQ. We typically use that topic to pull together all of the resources that we have about accessibility in Tableau in one place. So there you can find our accessibility conformance reports, which is a way that we communicate our conformance with the guidelines. We have links to all of the product documentation for the creating accessible dashboards. I have numerous tips and tricks that I've created to help address some of the more common questions that people have come up with. There's some videos of me doing presentations at our conferences that go into even more detail on these things and how they relate to the various WCAG guidelines. And then there are some demos. In particular, there's one demo. It's a YouTube video showing how what it's like basically to use a Tableau dashboard using the keyboard and a screen reader. And so hopefully this presentation kind of piques your interest. And you can take it from here and go to the Accessibility FAQ and learn more about accessibility. Certainly feel free to ask any questions in the community. I monitor it for questions related to accessibility and usually pretty engaged there. And I hope that you're able to start creating more accessible content in Tableau now. And that is the end of my portion of things. So I'm going to stop my screen sharing as well. Of course, now I can actually see any comments in the chat. It was a question for both of us, Kelly, as to whether we're going to share our slides. I said I definitely intend to share mine. I assume you are. Yes, as well. Yeah, I'll send off a PDF of the slides back to the organizers. Good morning. There is one more minute to the top of the hour before it becomes officially noon. So now we have gone through the boring part of the presentation. Now we are getting to the exciting part of the presentation. Thank you. Thank you Kelly for the portion. This is a lot of good information about accessibility and thank you Kelly for the sharing the tips. And of course, thanks to the team for inviting us to share our to an opportunity to discuss accessibility of this vision, the data visualization too. So, I have been told that typically it is meeting is for an hour. Hopefully, if I can stay for another 30 minutes to see them for the exciting part of the presentation. So my name is Hadi Rangi a member of it accessibility team. If you are with you that you probably in the past year I have been involved in some of the accessibility project with you and then had the chance to work with you in one way or another. My primary responsibility is that to make sure that the software that we develop here on campus or the purchase are accessible. So, we are involved in a lot of accessibility testing and evaluation and collaboration with the vendors, as well as with our local designer and developers. I'm going to share my screen. I mean, it happens that I am blind and I use a screen reader and then again, this is one of the topics that I found it extremely extremely difficult to understand and read and then discover. You know when we as a screen reader user when you as a non blind or a screen reader user look at the interface within a fraction of second you can identify the major object on the page that the design you can see what is what and then depending on what why you are on that page you can narrow down your focus and then see and then dive into that section that you came for for a screen reader user it is very different. We do not see as a pick the page as a whole we see only one element at the time so we see for example one heading. We see one piece of a static text at the time we see one form control like a button or like a text box link, all as a independent and individual elements, and then we do not see the relationship the visual relationship between these elements. Yes, you guys provide this list that there are structures behind those elements. So, let me share my screen, I hope. Okay, I am now with a lot by by popping my screen to the system. So, can you tell me this is the volume level for this community that is the data is the media controls window. It could be just a little bit louder perhaps. Okay, so, so make sure everybody is awake. Here on this landing page for the examples that we are going to look into. You know I'm not going to really dive into very much details but one of the basic elements that we see in every page as a guidance for navigation is the heading structure. It's also important that really if you are not there we need to read everything for every page from top to bottom to understand what the page is about. And then it's interesting link when you read the entire page then you see that it is not relevant. So it is a perfect waste of time. And security that provide a lot of means to help people blind people to get some information about their page one of them is the headings. And they have a lot of shortcut keys that there are a lot of short that keys that we use to navigate with the page. So, for anything that I interact in this mode. My screen reader is intercepting that and then tries to interpret it as a command. For example, if I type letter H. So it took me to the heading one which is a case of the data visualization. So that is good the pages start with the heading one so I know that the what the page is about that is what the what the purpose of the heading one is. So I moved to the next setting main landmark table heading level two. Then it is about table. It's another heading which is a heading two. So I know that it is a major section of the page Microsoft power by heading level. And then the next heading is about the Microsoft power BI. And then the next one is about the high charts. So, this heading mechanism is not only helps me to understand the structure behind the page on the page it also helps me to navigate to the desired section so so important and this is an important that Kelly also mentioned. And both tell you all and Kelly mentioned that, and then I do not know at this time it happens automatically automatically in in tableau or not, but it is so critical that otherwise it will be a total failure. So, let's navigate to some of these pages, some of the examples that you want to look into. I think we decided to showcase this example. I'm going to open it back. Actually, if you go back there's there are two UW libraries links. I apologize that was the one. Yeah, that's the one. Okay, a library's profile Google Chrome, a library's profile document busy blank, a library reference Santiago has left the meeting alert, a library's profile document. If you are leaving, if you are leaving my screen reader with let me know. So, stay there. Okay, I am on that page. Consider me on this page. I mean, I have had the chance to meet with to talk about this page and this example for several hours of material and Kelly. I have some idea what is on the page, but discovering the functionality, and all those information of the page, it is so cumbersome so difficult and sometimes inaccessible. So on this page for example, I think Kelly is referring to them as a bands. I think information that you see there. I completely miss it. When I didn't know that I mean the event area I didn't know that I don't do not see them. But this is for example one of the biggest problem that we do not see. In the workaround solution, Kelly mentioned that I think Terry also emphasized that it is always good to provide your message, I mean the author provides the message of the information of the of the of these charts or of these visualization to explain that in plain English, what the page is about. And what are the highlights of the. The common practice that we also recommend to the old professors and only instructors who create the course material, saying that you know, do not expect that you put a chart there and everybody understands that. So if we also always recommend to provide the data table that the chart is driven from as plus the summary of the message what is on that chart or what user should see and understand. Okay here but not available and do but not available redo but not available. Okay, I do not data visualization. Since interacting with this will be in school that we usually have to mode reading mode, which we call that they kind of also discovery mode to just see that what is around. So, in that mode, we see practically every object that it is exposed to screen reader that bands that that that are on the top of the page, they are not exposed to screen reader so we do not see. And then the other elements, depending on what the order of the of the linear of those elements are. Good varies. Kelly mentioned that you know some of this tab order might not be correct. So if it's not correct then we do not we do not see them in in in context so we see some information mixed up, and then we do not necessarily can understand based on what we see that because they are not in the right order. So let's interact here in this page, or let me go first try to, but not available and do but understand the heading elements list dialogue, three of you. But not available application data visualization graph data visual is applicable data visual. I'm trying to find my way to elements list dialogue, three of you. These are the headings used on this page, which is really important. There is a some problem with the order of the heading three and holding two, but you know, hopefully the tableau can manage that and fix that circulation to a five level zero. And then these are the subject. I still I cannot say the list of the headings that we see here. It or are representing the headings of the page properly or not. There's no way for me to determine that, unless I checked with somebody who can see that and then we can goes one by one through this heading and tells me this is in line with what you see that but we wanted to see that. I think the circulation that terriel mentioned that and see that how much I understand from the circulation type group cancel button. Clickable circulation that in level two, I thought to the circulation section, and then I am going to interactive mode circulation graphic process to clear any mark selections to open the view data window press control shift enter. A lot of orbit, but it didn't just it is a graphic and it told me that how you can see the data behind that chart. So data visualization graphic process to clear any mark selections to open the view data window press control shift enter. That is another visualization data that I have no idea it is a difference between what was it that is a different between the previous one and this one research consultations graphic and then we go to the next topic. So, and the filter, unfortunately, is not in the neighborhood. It is some way farther down where I can select the, you know, the campus. So if terriel or Kelly didn't tell me that there is a filter there I wouldn't be able to discover that at all. But let's go to want to see the one of these chart how much I see that. The chart that we want to look at. Yep, you're on the circulation visualization now. Okay. I have been told, you know, press control shift enter to open that in a data. I am opening that in level one new data. So what you see, it is a data associated with that data visualization that I see that completely in a static environment. So main landmark showing for seven rows. So it is it is nice. I mean the information has been organized in a table summary table with eight rows and two columns summary. And then I can read the table, not in a table cell row one accurate row two fight when some count edge of table academic edge of row three fight 2014 some count column two three hundred ninety one thousand two hundred thirty two. So I see this data, but in a in a summary form, but there is also a full data. I can find it out of table showing for seven apps selected summary type full data, full data, full selected. Now, the table main landmark table with 201 rows at 201 rows. Now, consider if somebody wants to read the data, and then make the conclusion that terrier was able to see within maybe a few seconds that you know that the trend of circulation has been changing since 2013 or whatever for 14. So I have to go through the data row to Seattle campus filter campus group column three Seattle campus libraries row three Seattle campus library four. row eight Seattle camp, campus filter column four true. Cirque type column five initial Cirque. Count column six, 2,000 edge of table. And the Cirque type column five row nine. So consider understanding this huge table and determining what you guys see that immediately here on this page. And if I want to filter that, then as I said, first filter was not in immediate neighborhood. Data of research. I have to go. I have to go. Data visual, data visual, data visual, data visual, data visual, data visual, data visual, data visual, data visual, reference questions. Ask question, put form landmark. Select campus all library. Now I have access to the form. Yeah. Pool bar, go to tableau. I'm sorry. Application, form landmark. Select campus all libraries and campuses. Okay. Pool bar, go to tableau public button. Question point, not selected one. So I lost my... Reference questions, ask question, form landmark. Select campus Seattle, campus libraries. Select campus Seattle, campus libraries. Select campus Tacoma, campus library. Okay, now I can access it. So I selected Tacoma, but it even doesn't tell me the graph has changed. Select campus Tacoma, campus library. And now, if I want to see the Tacoma data, I have to go back. Data visual, data visual, data visual, data visual, data visual, data visual, data visual, data visual, instruction sessions, research consult, date, date, circulation, traffic, process, the clear end, data visualization, traffic, circulation, traffic, data visualization, traffic. I guess this is the, this one. Terrell, can you verify that, please? Actually, you've bypassed it, but that a couple of comments, and this is, we have met, Hadi and I and Kelly have all met and explored this dashboard and some others over the last week and learned a lot from each other, I think. But, and I just learned this actually, that what you seem to be on now is one of those bands. So they do show up, each one is an individual visualization. So right now, the quick figures, there are 83 instruction sessions. And so that's showing up as its own visualization. And you could then access the number 83 by doing control shift enter now and going into view data mode and seeing a table of, it would be a table of one row and one or two columns. So that's a lot, a lot of steps to get that one data point. But these are accessible via that means. The other thing is circulation, since there is some heading structure here, you could probably use the H key to get back to the circulation heading. I was there. I was, yeah, I was circulation and then I was forward. No, I have to, again, I have to move, switch back and forth between reading mode and interactive mode. Circulation, graphic process, the clear image mode. This one? Yeah. Because since they have the same label, there's no way for me to say that. All of them is, so that's the visualization, something like that. That's very same there. So then I can open that selection because I chose Tacoma now in a static way. So as you see, these are the very cumbersome and then despite all those accessibility effort that has gone into that is semi-accessible and then definitely not easy, definitely not easy to access that. But this is a way how Tableau does that. And then we are hoping that in collaboration, cooperation with Kelly and then Tableau, we can improve the level of accessibility and ease the communication between application and then a browser. So screen reader can obtain the information. As Teriol mentioned, there are some other tools in the market that we have been working with. Highcharts is one of them. And then I had the chance to work with them in the past. I really didn't work with them in the past recently, but I'm very delighted to know that they have also taken accessibility seriously. And they have, we contacted them last Friday and they created a few examples for us. A library's profiled data visualization accessibility SAT entry, list with one, list with four items, iCart's accessibility demos, visited link, population of seven more. This one that Teriol told us that we would like to, that we look into then. And you all. Link graphic to get missed toolbar, view site information menu button secure. Address and search bar at JS Fiddle document. Link graphic to get missing image descriptions, open the context JS Fiddle document. Link graphic to get missing image descriptions, main landmark, edit. You went in to edit the code, Hardy. As opposed to running the Fiddle. I didn't decide, I didn't want it. Out of edit, multi-line. Out of edit, clickable one. Fill down pointing, small triangle. So do you need it? Yeah, there you go. Link graphic visited link editor in JS Fiddle. Clickable run, this Fiddle run code. Now, this is probably not as complex as the Tableau examples that we had, but it is showing that, you know, the population growth in Nordic country. Frame region world bank.org link. Figure population of seven Nordic countries. ICARTS interactive chart, region clickable Sweden. So this is the stuff that, you know, they have created. I mean, the problem is that what we would like is we would like to see the interaction in the same manner, same place as others. So we cannot, we should not study the data offline. We want to see, I mean, that's the reason why we make it interactive chart or... And then when we make it, this is a purpose for that. So we need to see the same interaction level right in that chart, not in a different page. So that is what I would like to see that in Tableau, that we have that interactivity right in the chart, not that we are redirected to a different page and study just the data. Data is wonderful to have the associated data table, but it should be, one should be able to study the table and then interact with the table, with the visualization table, with the dashboard, right landing page. So view chart menu button. If I... World Bank clicking on countries in the map will display population details for that country in the adjacent. So that is the interesting information that Kelly had also in her tips that you convey the message, you tell the highlights of the visualization that everybody can see that. And it seems that Tableau has the mechanism to provide that summary for description. Figure population of seven Nordic countries, interact interactive chart. Let me now be in interactive mode. Interact with the World Bank.org link figure. Population of seven Nordic countries, Denmark, value 5,818,553 population, graphic. So as I move with the keyboard, I'm... Nicholas Simon has left the meeting alert. I am using the keyboard. Finland, value 5,520,314 population, graphic. It told me that the population of Finland, but if I want to go to... Norway. Norway, value 5,347,896 population, and then I'm pressing tab key. Zoom out button. Zoom out button. View chart menu button collapsed. I'm sorry, hold on. Selected country, combo box, Sweden collapsed. Showing details for Denmark, then Finland, Norway. Showing details for... You know, as soon as the data loads, it tells me that the data is ready. That is a wonderful thing that we need such into feedback, real time feedback, also in Tableau, that we know that the chart has been loaded and it's ready. Figure, population history, Norway, ICARTs interactive chart, region, interactive chart, graphic. One, 1984, 4,140,099, graphic. So I am in this table and I'm using just arrow key. I'm using the arrow key to the next value. Two, 1989, 4,200,000... There's meaning there for 1999. Three, 1994, 4,394, that was 89. Four, 1999, 4,5, 2004, 4,591,910, graphic. And then if I want to see, for example, some of the examples of the cities that have grown. Figure, most populated cities, ICARTs... Most populated city, and... Oslo, 1,467 population, graphic. Probably there are more interactivity on the dashboard that I don't have access, but it is a project in progress because they really created our weekend for us. And then... But this shows that it demonstrates that what an accessible visualization tool should be able to do that. So what I am emphasizing there, the past should be to provide interactivity, accessible interactivity model right on the chart and do not necessarily redirect them to just study the data table and then determine from the huge table to make your determination. Yes, researchers might need that, but not every visitor is a researcher. So back to you, too. With five minutes left, I guess we're gonna open up to any Q&A. Nothing new, there have been a lot of messages of appreciation in the chat, but if anybody has questions, feel free to ask. Otherwise, Hadi, I wonder if you might also look at the sonification example, just real quickly in a couple of minutes, see what sonification is all about. I think that's really interesting. Sonification... Data visualization accessibility San Francisco Art Soundification Demo Document Busy Blank Soundification Icarz Document Again, this is also one way of doing that. It is really interesting. I am not a person that who personally do not use it because I'm really not sure that I can understand the sonification, but I think with some practice, people can do that, can understand that, give us some idea. Skip navigation, let me... Main navigation. List products, link submenu products. List Icarz, link Icarz. Icarz stock, link Icarz stock. Icarz maps, link Icarz scan, link Icarz scan. Icarz editor, link Icarz set. Icarz wrappers and add ons, link wrappers and add ons. Demo, link submenu demo. List Icarz demos, link Icarz demos. You're in the navigation menu currently, Hadi. So, I haven't, since I was not, I didn't prefer, I didn't know where it is, okay? So, let me do that, let me see. Main landmark sonification number, adding level three. Is it here? In the main menu? Yeah. Number link, direct link to adding, adding level three. Icarz example frame code, pen link. JS fiddle link. Figure, play, chart, sound. Icarz interaction. Which one, this one? Then, can you tell me? Yep, you're on the, one of the lines now in the two line chart. Toggle series visibility region. Show San Diego toggle button, chart menu region. View chart menu button collapsed. Play button. Courtney Berger Levinson has left the meeting alert. Let me, I'm just playing two graphs at the same time. Figure, play chart, sound, chart menu list. Level one view in full screen, one of nine. No. Toggle series visibility region. Show New York toggle button pressed, not pressed. Chart menu region, view chart menu button collapsed. Toggle series visibility region. Show San Diego toggle button pressed, chart menu region. View chart menu button collapsed. Chart menu chart menu region. View chart menu button collapsed. Toggle series visibility region. Show San Diego toggle chart menu region. View chart menu button collapsed. Tell me, where's the play button? Try tapping it again, I think, one more time. Toggle series visibility region. Show San Diego toggle button pressed. No, it doesn't go. interactive chart graphic, San Diego line one of two. Either tab or shift tab, whichever I'm not sure it's direction. No, don't say tab forward. Keep tabs forward. OK. OK. iPart's interactive chart. Toggle series visibility interactive chart graphic. I think it's going to give you a piece by piece. But if you can see, because I didn't really look into that, if you want to play at your end, feel free to do it. OK. Well, we're pretty much out of time. I think folks probably got the idea from that one example. And listening to two lines at the same time would be a real cognitive load for most people, though some might be able to handle that. But I think, you know, I imagine like the library circulation line, you know, where there was a little blip in the Tacoma timeline. And so, you know, if you were to listen to Seattle and Boothal and Tacoma, you would hear that Tacoma has some unique patterns. And particularly for really large data sets, we're going item by item by item through the line is going to be prohibitive. To be able to listen to a much larger data set, it could be really valuable. And I know we actually, there was a conference we were at a couple of years ago that had a keynote Wanda Diaz-Mercede, who is an astronomer who is blind. And she makes a lot of use of sonification to explore the universe and really huge data sets and has made some really important contributions to the field by accessing data in this way. So it does have a place, I think, in the data visualization. You think of eyesight, but it really is more than that. You can visualize, in a sense, with sound and by exploring with keyboard and lots of different ways to visualize data. So we're right at 12.30. And I do appreciate everybody sticking around. And hopefully this was meaningful. I know that I've learned a lot, both from Hadi and from Kelly and from all the exploring that we've done. And again, those are the two example sites from the UW where they, because of the limitations of Tableau, they had to go the alternative version route in order to make the data exploreable as well as tell their story. That's all going to be on tap for the April Tug meeting. So I encourage everybody to come back and to hear some more about what those folks did. But also, thank you, everyone, for coming. And then I appreciate the user group for inviting us. Indeed. Thank you so much. Yeah, this is really excellent. Maggie, do you want to wrap things up for us? Just thanks, everyone, who stuck around this long. Thanks to the presenters. And we'll see you all next month. Thanks very much. Good to talk to you. Sorry, Bloom.