IGF 2018 - Day 2 - Salle III - WS45 1.3 Billion Reasons for Making Technology Accessible

The following are the outputs of the real-time captioning taken during the Thirteenth Annual Meeting of the Internet Governance Forum (IGF) in Paris, France, from 12 to 14 November 2018. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the event, but should not be treated as an authoritative record. 



>> GERRY ELLIS: Good morning, thanks for coming back, those who came back. All of the new attendees who came in who weren't here before. I'm Gerry Ellis, I'm from Dublin. I'm here with the accessibility called 1.3 billion reasons for making your technology accessible.

      I'm going do a quick introduction and then ask each of our five speakers to say two sentences to introduce themselves. So, each person, just two sentences. That will do two things. It will introduce and it will also make sure we can hear the remote people properly. So, they're working at two birds at one stone. And each person can do a presentation and we'll have time at the end for questions and answers. Why 1.3 billion you say? There are 1.3 billion people with disabilities in the world according to a global economic report on disability. They control 1.2 trillion dollars of disposable income annually.

      And, of course, no person with a disability is an island. They have family, friends, people who are emotionally attached to them. And if you add that, that's another 2.2 billion people who control $8 trillion worth of disposable income every year. That's a huge potential market, apart from the inclusion or whatever that accessibility gives. So, what we're talking about today is why make people -- why make technology? Why make society accessible? And we're going to look at some of the policies. We're going to look at some of the standards. We have a wonderful set of speakers that are going to give us a broad view of what we're talking about.

      So, first, Gunela, can you give us one, maybe two sentences about yourself?

     >> GUNELA ASTBRINK: My name is Gunela Astbrink. And we have a big echo here. So, I'm from Australia. I am a member of women with disabilities Australia. I'm a member of the Australian chapter of the internet society and also the disability chapter of the internet society. And have been involved with disability issues, as someone with disabilities around 25 years, thank you.

     >> GERRY ELLIS: Shadi, two sentences. You know what's more important about you than I do. So you tell us. Shadi?

    >> SHADI ABOU-ZAHRA: Hello, I'm Shadi Abou-Zahra. I am taking from the captions that it's my turn. So, I work for the World Wide Web consortium, W3C, an international standards body that develops standards for the world-wide web. And I focus on making the web accessible to people with different disability since -- also for nearly two decades now.

     >> GERRY ELLIS: Shabbir, give us a quick introduction to yourself. Very quick?

     >> MUHAMMAD SHABBIR: Hello. Since I've been unmuted, I can tell it's my turn. I'm from Pakistan. By profession, I'm with international coalitions and I'm a member of the internet societies chapter. The accessibility field, been working for the last five or six years or so. Thank you.

     >> GERRY ELLIS: Thank you. For our remote moderator, it seems we can hear our speakers but they can't hear us if we can fix that. Derrick on my right-hand side?

     >> DERRICK COGBURN: Hi, I'm Derrick Cogburn. I'm a professor and the school of international service and focusing on international communication and information technology and analytics and executive director of our institute on disability and public policy and co-director of our internet governance lab.

     >> GERRY ELLIS: Wonderful, last but not least, Anthony?

     >> ANTHONY GIANNOUMIS: Hi, I'm Anthony Giannoumis and I do teaching in Universal design in ICT at the Department of Computer Science at Oslo Metropolitan University.

     >> GERRY ELLIS: You can hear from that such a variety of top-quality speakers. Looking forward to hearing this today. So, Gunela, let's go back to Gunela and give us your presentation, please? Thank you.

     >> GUNELA ASTBRINK: Thank you very much, Gerry. And I will -- very brief presentation. And -- (broken audio) I still have a very bad echo, so I'm going to try and --

      So, the internet -- what -- and called IRT for short. Budget concerns that -- (indiscernible) internet and these are -- (indiscernible) exchange for example, the internet society mentions that shoes can track your heart beat and alert you to health problems, TV can record your movements and conversations which is a bit scary for privacy reasons, and smart transport systems both public and private as some examples.

      So, we've already -- I've already mentioned that with the record evolution of IRT, there's a lot of challenges to -- to dealing with privacy, security, and also interoperability. And these are main stream issues are just as relevant to accessibility to IRT devices to people with disability. And mainly talk about IRT in the smart home environment. In the past, people with disability, particularly physical disability, have used ambient assistive technology --

     >> GERRY ELLIS: I wonder if Gunela's internet gone? Gunela? Gunela Astbrink I'm presuming you're not hearing -- the echo.

     >> GERRY ELLIS: No, we're not hearing you well now. Your internet connection probably. If you're on wifi. The audio is being very difficult to take. Gunela?

     >> GUNELA ASTBRINK: Should I continue?

     >> GERRY ELLIS: Your audio is very blurry. Sorry. You need to have a better internet connection? Part of your intervention was well heard. But if you could maybe try to write it. But maybe if you can a little bit more. Otherwise she might talk a little later if you manage to have a better internet connection? Maybe we'll move on to Shadi and come back to you after Shabbir, give you a chance to redial in.

     >> GUNELA ASTBRINK: Yes, it's okay.

     >> GERRY ELLIS: The trials and tribulations of international remote participation, ladies and gentlemen. So, I apologize to that. Shadi, can we move on to you and come back to Gunela in a few minutes. Shadi?

    >> SHADI ABOU-ZAHRA: Yes, hello? So, I think I think the room needs -- yeah, thank you, for muting. I think that sometimes interfere s with the speaker. I was actually hearing Gunela speak. So, but anyway.

      Thank you, Gerry, for this opportunity. The title is 1.3 Billion Reasons for Making Technology Accessible. I think there may be more reasons than that. I think these are the 15%, 20% of the population that actually need accessibility. But I think we at W3C look at accessibility more broadly. It's a requirement essential for people with disabilities but also beneficial for many more. And there are many examples. I would like to compare that. As a person with disability using a wheelchair myself, I like to use the built environment as an analogy because sometimes that makes it a bit more tangible for people to understand and see what we mean.

      Now, everybody uses an elevator. But nobody thinks of it as an assistive technology. But this is actually very similar to what we have in the mobile -- in the digital world. We have accessibility features and functionality such as text-to-speech, voice input, key board navigation, all -- all of the like, being able to customize the presentation of text making it more readable, more understandable. And these are all essential for people with disabilities. Without these, people with disabilities are essentially excluded from accessing technology from being able to use it effectively. But at the same time, these features also benefit everyone. And I think similar to an elevator, we have to see this as the standard way of building technology.

      Very often, we hear the question about cost. How much is this going to cost us and how many wheelchair users are there to build an elevator. If you take only this cost measure, it's a very skewed way of looking at it. If you look at the costs, you need to look at the benefit as well. You need to not only calculate the 15% to 20% of the population that needs this accessibility, but then maybe 2/3 of the population in some studies that benefit from accessibility features. 

      What I'm referring to here is actually a study made by Microsoft. Already back in 2003, they commissioned Forrester research to look at why they need to access accessibility for the Windows operating system. They came to the conclusion that 15% to 20% of the -- of the American workforce, it was back then a study local to the U.S., that they require accessibility features in order to be able to use the computer. But that nearly 2/3 of the population actually benefit from the accessibility features. Again, these are things like captioning, like being able to increase the font size, being able to scan the headings or use text-to-speech to interact with the computer. And since then, they've actually changed the phrasing instead of using -- calling it accessibility, they call it now ease of use because this is what it is. It makes technology easier to use for everyone and more accessible specifically for people with disabilities.

      So, in -- in that -- in -- in that approach, accessibility standards are important tool. In order to guide you to what features when you're designing or a service, what are the requirements? What considerations do you need to have in -- in your products and services from the beginning and follow that through so that at the end you have the an accessible, an inclusive product and service.

      Specifically, I wanted to mention the web content accessibility guidelines, WCAG is a standard for web content accessibility. With web content, we really mean the broad definition of anything on the web, that means videos, audios, forms, texts, images, all of these things on the web, increasingly, these are accessed using mobile apps. Also, these are part of the web in our view increasingly even home appliances, digital televisions, all these things are now becoming web-enabled and part of the world-wide web. And part of the internet of things so to say.

      Yeah, in addition to providing these core technical standards for accessibility, W3C also provides a lot of educational resources on how to make -- how to apply these standards in practice. Because part of the big issue that we think is part of the -- you know, the barrier to accessibility is not only the technical standards, but also the lack of awareness, the lack of knowledge and skills amongst many of the designers and the developers and people who are responsible so the project managers and the executives who are not aware of what the accessibility requirements are and how to apply them in practice.

      So, this is part of our mission. We're more coming from the technical angle and we try to explain these standards to the widest audience possible. And in order to ensure that accessibility is included in all web products and services.  Thank you.

     >> GERRY ELLIS: Thank you, Shadi. Very interesting. I think the question of standards is so true. We need something to measure it against. If you can't measure it, you can't manage it. Shabbir, I think we'll move on to Shabbir now. And can you give us your presentation, please? Thank you.

     >> MUHAMMAD SHABBIR: Yes. Thank you. Detail in the responsibility and importance and significance of the standards. I will be talking about the challenges in -- in the domain of accessibility. And what are the differences, particularly, as developed and developing countries?

      A small disclaimer here that I am speaking on my personal behalf and not on the behalf of the institution of the university that I work for. So, this is my personal opinion and might -- and the speech that I am going to give here is my personal observations. So, I may be wrong and the list and challenges that I'm going to highlight here is not exhaustive. It's unending and anyone can add to this -- to these challenges, but as I see there are multiple challenges, both related to policy, awareness, and technical domains.

      First, a little bit. I will talk about the accessibility in Pakistan. As I -- the country I hail from and I come from. Before I talk about the accessibility in Pakistan, when Shadi was talking about the -- I will just refer to one thing. When Shadi was talking about the standards and the -- and the importance of following standards, I'll give you just one personal example. That here at IGF, I've been using this on-line tool. But it is not possible. It is not even accessible, rather, I should say. For me to mute or unmute myself or to raise hands to ask for the floor from the remote moderator.

      So, this is just one example. And as some would say, these are small things. But in the -- in the longer run, these small things do matter. And if the standards are followed, then we also need to have some audits as well. Something about the accessibility in Pakistan. There have been a number of efforts starting in -- starting in 1980s when a decade for accessibility for a person with disabilities at that level was celebrated from 1983 to '92. Mainly the efforts for persons with disabilities started at that time.

      From then, we have a long journey. In 2009, Pakistan signed a United Nations convention on the rights of persons with disabilities. And in 2011, July 6, to be exact, Pakistan, government of Pakistan ratified it. Since then, there have been a number of efforts. If I give you one example of challenges between the developed and developing countries, before 2014, for persons with visual impairment in Pakistan, it was not possible. And even banks, private institutions, or the government banks would not allow a person with -- a person with visual impairment to carry debit or credit cards or use e-services of the banks.

      But in 2014, thanks to the government -- some government institutions, statements of Pakistan particularly and some other organizations, we negotiated with the state bank of Pakistan, it drafted a complete policy and now banks are bound to give all of the electronic services, including ATM, debit cards, credit cards, and all electronic facilities to the visually impaired customers as they would issue these to any other customer. So, this is just one example. If I lift these challenges of accessibility for persons with disabilities, I see these are main four challenges. There may be, again, I would say multiple issues and can -- and people can -- we can discuss this in the interactive session exhaustively.

      The -- the limited low-income versus high cost of the technology, with that much of the cost of the technology and the -- in the developed world as well. But when we come to the developing countries where GDP at issue is already low, the technology versus cost of technology is very high.

      Second is the since the accessibility policy or there is a gap between what policy is stated and what is practiced. So, the absence of the policy means, as I mentioned, that in -- before 2014, banks even would not allow a person with visual impairment in my country to carry the ATM or the debit cards. But now there is a policy, but there are a number of things to -- to work on. And if I understand it right, Bangladesh is still one country where banks are -- banks do not allow persons with visual impairment to carry or use these services. And Afghanistan is one of those too.

      So, this is -- this is some kind of a gap which needs to be bridged. And then there's a gap between what policy states and what is practiced. Sometimes it happens that policies are made for persons with disabilities, but the difference between developed and developing countries is that the developed countries, they have strong legal and institutional norms. And the institutions, they -- they made the companies and the people to -- to follow up on the accessibility rights.

      For instance, if there is some issue that comes under the -- under the section 508 in the United States, the companies and even the government institutions can be dragged into the court. So, people have been fighting cases on those as well.

      But, in certain developing countries, it becomes difficult. Then there's lack of awareness about accessibility, sometimes by developers. Then the service providers and persons with disabilities themselves. Developers talk about in detail, service providers sometimes like banks and other institutions, they do not including some technology for their newly installed buildings or infrastructures. They do not -- sometimes they have a lack of awareness about the accessibility features and what accessibility they can use. And this creates -- this causes a problem as well.

      Lastly, persons with disabilities themselves, sometimes the technology is available. But persons with disabilities, they are not aware of the -- of the technology, which is available for them and to make the -- make the life easier for them.   So, it becomes an issue as well. Lack of accessibility training. This is an issue. I have been talking about it here in Pakistan and we're hoping that a couple of universities have started now including in the BS programs.

     So, on the developer side, what we are doing here is we are trying to add some courses in the BS software designs or other link so that a who are doing the advanced software engineering development and design, etc., they should be aware of the accessibility and its needs. Because what happens is if you -- if you teach a student for four years, just the design for the businesses then it becomes difficult for them to follow the accessibility standards and to relearn from the -- from the -- to relearn so as to include the accessibility issues and standards from the designing on the development side.

      So, my effort here has been to encourage students and facilities to have courses which make the students about awareness and significance of the persons with disabilities and their issues.

      I talk about the policies, talking about the example of Pakistan last year, the policy was on ICT, which is here in Pakistan, digital Pakistan policy. The policy carries a certain guideline that address the issues of persons with disabilities and as we speak, a -- an action plan is being conducted to implement that policy. So, the policy has the guidelines, but let's see, we are still working on that as well. That whether those policy guidelines -- those action plans were how those guidelines would be implemented and what kind of instructions are there in the action plan to implement the provisions of the accessibility for persons with disabilities. And the thing was that the guidelines while being drafting the policy, the ID and the other global institutions, they involve the persons with disabilities and ask for their input as well.

      The best approach, if we want to make the accessibility -- the digital accessibility and the digital accessible for persons with disabilities, we also need to -- to follow a policy which includes the input by the persons who are actually suffering from the disability because it is them who will have -- who lives with it themselves. And often -- more often than not developing it happens that policy considers themselves all knowing and say whatever we do or whatever waiting is the best option.

      This, perhaps, is the wrong way we somehow need to change this approach. Lastly, I will -- (indiscernible) talks about 1.3 billion reasons so this is not a -- (indiscernible) 1.8. We all know that circumstances -- please forgive me for being -- but we all know what -- (indiscernible) tomorrow. So -- tomorrow -- (indiscernible) we also need these kinds of -- (indiscernible) so from a -- the disability whenever we make the environment accessible. So we need it, we also may -- (indiscernible) thank you.

     >> GERRY ELLIS: Thank you, Shabbir. Ladies and gentlemen, we're going to try Gunela again, all the way to Australia. Gunela, can you hear us?

     >> GUNELA ASTBRINK: I can hear you loud and clear. And thank you, Gerry, for giving me another chance. I have reduced my volume, so hopefully it works. And it sounds like it does. I started my presentation. I'm a bit reluctant to start from the beginning, Gerry, so maybe I can continue from where we were cut off. Is that acceptable?

     >> GERRY ELLIS: perfect. You were saying security is just as important for the people of the rest of society, not just people with disabilities. So, if you want to carry on, please do.

     >> GUNELA ASTBRINK: Okay. We were talking about people with physical disabilities using ambient assistive technology that use to be called environmental design units. They support living, the entertainment and security systems. And we're integrated through accessible interfaces such as voice command or other control systems. And these assistive technologies were very expensive. Now we see Google, Amazon, Apple, and many other operations operate the systems for the general public at low costs. Which is really great. But, now people with disability may count new barriers due to the lack of interoperability between devices and inaccessible use of interfaces.

      So, we need to ensure and we've heard already from speakers about these types of issues. We need to ensure that main stream software is interoperable with various devices. And also with a person's assistive technology.

      So, for example, I think it's an example that Shadi used previously is a home assistive device can turn on or off a TV but not able to turn on captions on the TV. So, the way interface is communicate with each other may be limited due to business decisions to operate in the closed or semi closed environment. In other words, with particular operating systems and hardware and software is part of it.

      This is obviously detrimental for consumers and it may lead to negative business outcomes in the medium to long term. I remember many, many years ago, we first had SMS text messaging, and companies said okay, you can only send texts within our own company. And they realized they would make a lot more money if they allowed the text messaging to go across various networks. And that, of course, has matured. Now we need to see the maturing of some of these IRT systems. I just also wanted to mention that in Europe, there is in train legislation related to the internal market which may include interoperability. This is something to watch carefully.

      So, when using interfaces are based on Universal design principles, devices meet the needs of many more people in the community, which we've already heard, and we can't emphasize this enough, really. So, if we have intuitive and accessible interface design, it means that people of all ages and mental abilities can easily use a home assistive device.

      Markets with aging populations especially benefit. User testing and advised by people with disability in the early stages of the design process will lead to more user-friendly and accessible products.

      So, while product safety and ref liability are important for everyone, it's vital for people with disabilities to be able to function independently. IRT issues such as privacy and security are potentially harmful for people with disability and older people. For example, we had data used in a home care environment. So, that is the internet of things, IRT. IOT, and accessibility should be addressed in policy, research, and technical settings along with other IRT issues. And this should be included in any policy development. And, of course, that policy needs to be translated to action. So, the time to do that is now. Thank you very much.

     >> GERRY ELLIS: Thank you, Gunela. And just to make a point, with one person there from Australia and two others from varying parts of the world, and if they weren't able to access remote participation, they couldn't give us their words of wisdom here today. That in itself is the demonstration of the importance of accessibility and we can't begin to tell you how difficult it was to get Shabbir onboard because of the inaccessibility of Webex. In itself is a great achievement. Thank you to all three remote speakers.

      Now we're going start to hear from speakers in the room. First on my right, professor

     >> DERRICK COGBURN:. So, derrick, looking forward to hearing what you have to say. Thank you.

     >> DERRICK COGBURN: Thank you very much, Jerry, and to our remote speakers. Thank you to the DCAD dynamic coalition on accessibility and disability for inviting us to participate here and to continue this important work with IGF.

      It's also good to see so many people here for an accessibility workshop. Starting this morning at 9:00 with the morning session and continuing on, and this is part of a number of workshops here at IGF related to accessibility and disability, including on Sunday where we had the discotheque on accessibility and development. So, really excited to participate in this process.

      And I want to share a little bit about disability inclusive development. As we think about how do all of the various policy frameworks that we're talking about really access rate disability inclusive development around the world? 

      So, I want to do four things this morning. I would like to say a little bit about big data and text mining. I should have given you my full title that I moderated. Understanding disability inclusive development through text mining, IGF, SDGs, and the CRPD.

      So, I will start this morning by saying a little bit about what we mean by big data and text mining, some of the kinds of data that's available for this work. I'll give you a methodology to use. Some of them may be interested in pursuing some of these technique s. And I'll give you some key approaches that we used within that methodology to answer questions and problems from big data text mining sources.      And then I'll try to do with whatever time I have left three briefcase studies looking at text mining around the internet governance forum, the sustainable development goals, and the CRPD.

      So, let me say first about big data, analytics, and text mining. Internet government research has been around since the mid to the late 1990s as an interdisciplinary research domain. If you look on the screen, I have a slide showing Google searches for internet governance. Around 2005, 2006, it got a big boost on WSIS. It really accelerated how many people think about internet government research and moved a lot of people in to that process.

      Some of you that may be here on Thursday, the GGANET, the Global Governance Academic Network will hold the 13th annual symposium associated with IGF on Thursday.

      So big data is focused on what some people call the three V's of big data, volume, velocity, and variety. There's a lot of it. It's coming fast, and there are many types of data we can use for big data analytics. In particular, text is one of the largest sources that tends to go unutilized the most. So, for yourself right now, think about the sources of text that you can imagine that are available for project works.

      So, archives of international organizations, meeting transcripts such as the IGF meeting transcripts, e-mail listservs, reports, speeches, social meet the yeah, legal documents, all of these are available to us with a growing amount of computational tools. Some of the tools are commercial tools and some of what I'm going to show you today are the commercial tools that we used. It's called a Provalis ProSuite. But there are some that I teach in my classes such as R, an open source data management software.

      The methodology we use is Crisp-DM for cross industry standard process for data mining. I like to approach this because many people don't know how to approach data mining projects. This is a very straightforward way of thinking about a data mining project. It starts by clearly identifying what's the purpose, what's the problem that you're trying to address through the data? So, you think of all of the text that you have, what kind of problems might you want to address with that text? What's the availability of that data? Then you have to prepare the data. You have to develop and assess models that allow you to analyze that data. Evaluate your findings, and then deploy the results. Rinse and repeat as necessary. So, this is a very standard process for data mining.

      So quickly, I want to talk about some of the approaches that we take through data and then I'll turn to our case studies. So, the first are what we call inductive approaches. You have a big corpus and you're trying to understand what's in that corpus without coming to that corpus without any specific questions. You want to know what are some of the key trends and themes and patterns in that data? How does that change over time? How does it change by key variables such as maybe date or organizational type or person. And you can take an inductive approach. You can do descriptive analysis of key words, key phrases. You can use a technique called entity extraction where you can do topic modelling and you can pull out named entities in a data set.

      You can also take a deductive approach, that means we're going to the data set asking specific questions. Are the things that I'm looking for there or not? To what extent are these concepts available? And there's a technique that I'll show you in a moment called categorization model development or dictionary development which allows us to build a model, a semantic model around a concept and we can see if that concept is in the data or not. We can use machine learning and we will can use classification models, both supervised and unsupervised machine learning approaches.

      With that, I would like to turn to three case studies. To look at some of these techniques. The first will be the IGF. And if I have time, I'll turn to the SGD's and the CRPD. We're in what I consider to be an important time, a convergence of a number of disability inclusive policy frameworks globally. So, we have the main one that my colleague, Gerry, talked about earlier, the CRPD, the convention of the rights of persons with disabilities which he introduced as a human rights treaty of the 21st century. It's a development instrument. A treaty that's a human rights instrument and a development instrument. It's comprehensive, addressing almost every area of life from awareness raising, recreation, education, culture, almost independent -- independent living, access to justice. So, the treaty is very, very comprehensive.

      But we also have the sustainable development goals and the high-level political forum, the WSIS + 10. Habitat 3 in the new urban agenda which focuses on inclusive cities I know my colleague Anthony focuses on, as well as the global platform for sustainable disability and inclusive disaster risk reduction in the framework. All of these frameworks are related to persons with disabilities.

      So, we have an opportunity now to advance disability inclusive development faster than we might have been otherwise. What we want to do is look at how each of those or many of those frameworks are actually advancing the cause of disability inclusive development. So, we start with the IGF. We look at the IGF. We know that we've now had -- this is our 13th IGF. And our study around the IGF is looking at what has the IGF talked about over its 12 years? What have been the key focus areas, key concepts? Key themes within the IGF and in particular, to what degree has the IGF talked about disability and some of the other specific frameworks?

      So, we have an opportunity of 12 years of transcripts from the IGF driven in large part by the disability and accessibility coalition, dynamic coalition because of the focus on making transcripts available for persons who are deaf and hard of hearing, it's also a Universal design benefit to everybody who's in the room to be able to draw from the transcripts.

      And there's an ancillary benefit of now being able to have 12 years of transcripts to be able to mine. So, you think about everything that was said in the formal sessions and the workshops at an IGF, think about all of the sessions that are going on right now where you've chosen to be here with us but there are parallel sessions going on right now. This technique allows us to be able to mine those transcripts to see what we actually talked about.

      So, what were the key themes at the IGF over 12 years? So, if you look across all 12 years of the IGF, from 2006 to 2017, the most dominant phrase that was talked about that we have access to was human rights. So IGF has been a human rights-driven vehicle. It talked about human rights, freedom of expression was next. We talked about developing countries, young people, stake holder, data protection, but if you -- if you look across the IGF, you also see using another technique called entity obstruction, you see women, cyber security, internet of things, blockchain, and so forth.

      This is another technique called hierarchal cluster analysis. This looks at how words were clustered together and looks at their proximity and it looks at themes and concepts based on this clustering of words. And this is just a short snapshot of the visual clustering of these words and I'm showing this -- this pattern on screen with different colors for these clusters. So, look at this first cluster and you can see the words "abuse," "sexual" "child," images, gender, women, men, violence, hate, and speech. You can adjust how tight or loose you want the clusters to be. This is obviously a cluster that was talked about at IGF related to child sexual abuse and violence against women.

      You can look at on the far right-hand side, smart city devices and internet of things, another cluster. So, this is all coming from the text itself. Perhaps somebody who didn't even attend any of the IGFs is now able to get a really strong sense of what's here within the IGF. We can also look at top phrases over the different periods of time. You can see digital inaccessibility were active concepts in the IGF. If you look at topic modelling over the years, we see accessibility in 2009, 2011, 2014, and so over the years, we've seen disability and accessibility being a key component of the IGF.

      Now, I want to look at which quickly the bonus in my remaining two minutes here. So first, I would like to say a little bit about the SDGs and the CRPD. Let me jump ahead to the CRPD. I think that's mostly important. When we look at the CRPD, every country is required to submit a state report on how much progress it's making in implementing the CRPD. So, Article 33 of the CRPD talks about state implementation. So, we had a book that came out two years ago looking at southeast Asia and the implementation of the CRPD in southeast Asia. So, in order to do this, it was an extensive project. We had to have teams of researchers on the ground in each of the countries in southeast Asia and we were able to make a lot of progress looking at what they focused on in southeast Asia.

      But we wanted to look at the rest of the world, we couldn't do that. We don't have teams on the ground in the rest of the world. So, what we did was we looked at all of the shadow reports, we created a corpus, a data set, a text-based data set of all of the state reports that have been submitted as well as all of the shadow reports. So, we have 88 state reports and 103 shadow reports. And we built what's called a categorization model, which let us look at the article 33 implementation components, which says there should be a focal point and coordination mechanism. There should be an independent human rights evaluation process based on the Paris principles, and the third is there should be civil society and DPO involvement, there should be multi- stake holder. So, we drilled down to each of the three categories, that's why it's called the categorization model to develop sub categories and develop specific words and phrases and rules that would let us tease out what do these state reports focus on how much progress is made in each of the three areas. We can see quickly here that almost all of the attention in the state reports has been focused on setting up the focal point, the intergovernmental focal point and the mechanism to be able to move forward, which is a good place to start.

      Less is focused on the independent mechanism. No work has been to the independent mechanism and the Paris principles and so forth. Over time, I will stop there. And I look forward to any questions or comments I might have later. Thank you, everyone.

     >> GERRY ELLIS: Thank you, that was fascinating. It just shows, though, what we see and what maybe society sees as just something for people who are deaf or whatever can be taken and used back -- taken and used by people who are interested in gender or any other issue and use in that way. It's not just for people with disabilities. As Gunela said earlier, accessibility is not just about people with disabilities.

      My last speaker but by far not our least speaker is Anthony. And I'm not going to try and announce his second game again. Anthony, 13 minutes? Go.

     >> ANTHONY GIANNOUMIS: Thank you, Gerry. I will briefly be discussing today a bit ant life-long learning and inclusive education in Norway and how we can use ICT to promote these activities. For those who don't know, inclusive -- lifelong learning is basically learning in the purest form, it's self-directed, voluntary, ongoing throughout a person's life course.

      And it could be for personal reasons, pursuing knowledge for the sake of pursuing knowledge. It could be for professional reasons to upscale or become more competitive on the job market. But fundamentally, lifelong learning helps to promote social inclusion and active citizenship which is a huge component of just participation in society.

      To start off by saying Norway, we look at disability as a component of the aging process. It's a component of life. Statistically, we're all very likely to experience disability in some shape or form, particularly if we live long enough. Norway's focus on disability and inclusion is really rooted in the concept of Universal design. I have a short video I'd like to play you from the Norwegian governments.

     >> In the course of our lives, we all face obstacles sooner or later along the way. Stairs are no problem, until one day, you are standing there with the crutches. That's when you are pleased someone has thought about you and you can keep on going. The same when you're on-line. Videos with subtitles are good with someone in hearing loss or a commuter on a train. Easy to read text is helpful for someone with impaired vision or if you left your reading glasses at home. Key board navigation helps if you have a physical disability or if you're suffering. Society's demands do not meet our needs, a gap arises, we need to mind this gap. In Norway, we want a society where everyone can join in. Therefore, we're working to reduce this gap. We're doing it in two ways -- by strengthening the ability of individuals to join in. For example, when someone uses glasses when their vision is impaired or when someone uses a wheelchair, we're now unable to walk. By changing society's demands, by adjusting them to a level that does not shut people out, by providing good technological solutions for everyone, only then can we achieve Universal design and a better life necessary for some and good for everyone.

     >> ANTHONY GIANNOUMIS: I think it's the last phrase, it says it's necessary for some but good for everyone. That's the spirit of Universal design in many respects.

      So, I believe that Universal design can change the world. I believe it has the power to transform our attitudes and behaviors really where the pedal hits the road -- or the metal hits the road, whatever the saying is. It's particular when Universal design is applied to education and learning.

      Now, I have the privilege to be able to stand -- to be able to sit here in 2018 and discuss these ideas. These are ideas that in 1950, 1980, or 2000 are unheard of, idealistic, or just considered absurd. Education and especially higher education has historically and to some extent still is accessible  only to a small segment of the population, not only in Norway, but everywhere.

      This one is counter to our human rights. According to the United Nations convention on the rights of persons with disabilities, states' parties have an obligation to ensure an inclusive education system at all levels and persons with disabilities are able to access general tertiary education, vocational training, adult education, lifelong learning without discrimination and on an equal basis with others.

      That last clause is particularly provocative. That persons with disabilities have a right to access education on an equal basis with others. What does this mean for someone who is dyslexic to have access on the basis with others. What does it mean for someone with down's syndrome or someone on the autism spectrum. And since we're talking about life-long learning, what does it mean for someone who has dementia or Alzheimer's to have access to education on an equal basis with others.

      Now I'm a researcher and I can unequivocally say we have no idea. We have some knowledge about how to make some things work for some people some of the time, but we have no clue how to approach a system, an institution like the learning and education institutions we all have in our countries in a way that makes them truly universally designed.

      But fortunately, we do have this set -- this guiding light, the set of principles of Universal design that the U.S. has given to us as a potential solution for ensuring that everyone can realize their right to education. So, the United Nations defines Universal design as a design of products, environments, programs, and services to be useable by all people to the greatest extent possible without the need for adaption or specialized design.

      Universal design is made up of three clauses, one, that learning and education products and environments, programs and services should be usable. Second, that usability should extend to everyone and in every possible way. And third, that use cannot require any kind of modifications or ad hoc solutions. That's a very simple, but uncompromising view.

      When we apply Universal design to learning and education, I don't want to be misleading. It does not mean one generic vanilla solution that tries to meet everyone's needs but in reality, meets the needs of no one. It means creating a solution that is one size fits one. Not one size fits all.

      This means rejecting the idea that there's some kind of magical quote/unquote average, typical, or quote/unquote normal student. There are no average, normal, or typical students. All students lie on what's called the neural diversity spectrum. And this is how we approached learning in Norway, by trying to take a student-centered approach to learning and trying to customize the learning experience for each student. It's not perfect. It's a not a perfect system. It's something that allows the student to participate on an equal basis.

      This means providing multiple ways for the students to engage in the learning process. Providing multiple ways for students to act and express themselves both in and out of the classroom and providing multiple ways for representing information for the students. This means if all you're doing is lecturing and giving an exam at the end of the semester, you are not universally designing your learning experiences.

      Now, when we talk about accessing information from the Universal design perspective, we're focusing on removing barriers that emerge when an individual engages in a specific activity or a specific educational environment. Me, accessing learning content at home sitting on my couch using my I pad is a very different experience than me sitting in the classroom watching the same presentation in a much more formalized setting.

      The barriers that emerge in one environment may not be the same barriers that emerge in another environment. We have to be conscience that learning in a particular life-long learning is carried out in multiple different settings.

      One of the surest ways of ensuring that your educational experience, your learning experiences, can be universally designed is by involving students as co-creators in learning processes. At my university, Oslo metropolitan university, we directly involve students in a development of the courses that we offer. This means substantive involvement, not just come in, have a cookie, tell us what you think, now go home. It's also including students with disabilities and this is whenever we create new courses or programs, this is whenever we're designing learning outcomes for specific courses. This means determining what kind of formats are useful and accessible for students learning -- for learning outcomes.

      And we also work with students in the design and development of the new educational technologies including things like e-learning systems, multimedia learning materials, and technologies, I mentioned design and development. I'll add to that list, procure. Because if the university or education institution is buying a piece of technology it should never purchase anything that's inaccessible or unusable for anyone at that school.

      By involving students in the design of learning experiences, it ensures that the courses and programs we offer at Oslo metropolitan university are accessible and usable for everyone. I would be grateful if we could continue this conversation on-line.

     >> GERRY ELLIS: Thank you, Anthony, that's excellent. Unusual for IGF. Some people didn't use their full allotment of time. So, we have plenty of time for questions and answers. First, can I ask the remote moderator, can remote participants hear what's being said in the room?  Shadi, Gunela? Shabbir?

     >> Yes.

     >> GERRY ELLIS: Excellent. So, you can join in in the conversation.

      On my left will be my spotter seeing that I'm blind. So Shabbir, or Gail, do we have any questions? Tell us who you are or who you represent?

     >> I work for the European commission in a unit dealing with the web accessibility directive. And my question is to Anthony asking whether similar framework exists for school and secondary and primary school level education in Norway?

     >> ANTHONY GIANNOUMIS: The oh b ligations are across all education levels. The implementation of those requirements are never the same. In some ways, it shouldn't be. In other ways, there's a gap in the implementation process. So, short answer, yes.

     >> GERRY ELLIS: Can I throw it back at you and ask you a question, please? You're involved with the web accessibility directive. Can you give us some time scales as to when that -- I know the date when it came into force, but maybe some people in the audience didn’t? So maybe you can talk very briefly about that?

     >> So, the deadline for transposing the web accessibility directive into national law was the 23rd of September. This year. And it would mean that the old websites or existing websites, they should be accessible by September, 2019. And new websites by that -- by 2020. And for mobile applications it's as of 2021. So, I mean it has to be transposed to law and then implemented as you say. So the policy is there and re-enforcements are implementing it and starting to monitor how that's working. That's for public sector and mobile applications of public sector bodies.

     >> GERRY ELLIS: Thank you. We almost got a bonus presentation for free.

     >> Bonus information.

     >> GERRY ELLIS: Always helps. More questions, please?

     >> I'm from autistic international. I'm autistic. So, my concern primarily is obviously cognitive access and it's more a comment than a question. Perhaps even more touched on in the talk this morning. It's about the use of guidelines and standards. There seems to be a lot of faith being put on the ability of guidelines and standards to translate to outcomes for end users. I would have to say from my point of view as an autistic person, I would question that. Because they supposedly meet guidelines where I can't get in past the log-in page. So perhaps there could be some -- I don't know if anyone has any comments. Lots of other things I'll not talk about. It's all very interesting. But that in particular. And I know that someone -- that perhaps could involve in the future like Chaney knight who works for the BBC that takes care of -- the biggest thing is accessibility. He's autistic as well. He said that the BBC do not use guidelines and -- they don't work, or they're not applicable.

     >> GERRY ELLIS: Shadi, maybe, is our standard bearer. Would you like to comment? 

    >> SHADI ABOU-ZAHRA: Yeah, with pleasure. Yeah, am I audible?

     >> GERRY ELLIS: We lost Shadi?

    >> SHADI ABOU-ZAHRA: Am I audible? Yeah. Okay. Sorry. I'm not -- never sure when I'm muted here. So, absolutely I -- this -- the area of -- the broad area of people with cognitive and learning disabilities, including mental and behavioral disorders as well is, indeed, an area where accessibility guidelines and standards need far more improvement and I'm saying this is a self-critique as well. This is an area where we have oh done some improvements in 2.1, but there is a way to go.

     There's a research challenge in many cases, many situations, the actual requirements, the user needs are not well understood. There's a lot of progress right now on so-called personalization semantics, work that would allow parts of webpages to be described in a way that the machine could understand it, the computer, and could adapt the content in more personalized fashion to individual needs of individual users.   Because we know that in particular in that space, one size doesn't fit all, and it needs a high degree of adaptability and personalization. So, we're progressing on a lot of work in this area.

     But another thing to always keep in mind is that standards are tools and they're a means to achieve accessibility and inclusion, but they're not, you know, the end in themselves. They are helpers. If you don't have the right mindset, if you don't have the right approach to Universal design, to understanding their end users and designing the end users rather than for any particular tool or standard, then you will -- you will usually not be as effective as if you're trying to understand your end users and using the standards and guidelines as a way to -- to -- as a way to achieve these. And then the summary of a broad set of user requirements. But not all user requirements. It is something that's an ongoing development as we learn more, as research provides us with more input, we can define better user requirements.

     >> GERRY ELLIS: Thank you. Any other speaker like to comment about standards? Maybe I'll take the prerogative as a chairperson to throw my own two cents' worth. International standards are voluntary, not mandatory. What makes it mandatory is the legislation that underpins them. And one of the things that has been happening very well in the last ten years or so is the Harmonization of standards across the world.

      There's the European standard called EN 301 549 which is about the procurement of accessible ICT harmonizing with the American section 508 and which is adopted verbatim by Australia. So, as standards become more conditionally used across the world, it's -- you don't have different requirements in France and Germany and Ireland and Australia and America? So, it's easier for them to become main stream and it's easier for the products to result for them to be cheaper and having the products -- what does the internet say about standards is we heard a lot today about how accessibility is not just for the people with disability. It's for all of the people and for various things. So, if we can get that awareness out there that this benefits everyone -- that awareness out there that this benefits everyone, the will get adopted. Do we have another question here?

     >> Thank you, Gerry. I come from Kenya. Listening to Shabbir making his presentation, I could easily say he's presenting on behalf of Kenya. Because it seems like it's the same across the board. I will just wonder if there are any best practices concerning the things he's talking about, especially the many policies that we have that are not being implemented. What do we do about that? And then for derrick, I would like just to ask apart from the discussions that we've had are the IGF, probably this is something that you would discuss at the -- meeting. What are the accessibility issues we talked about are going to be implemented or not. But I would be more interested to know if the data mining, if there's information on the persons with disabilities, whether it's increased over the years at the IGF or it's decrease canned, thank you.

     >> GERRY ELLIS: Derrick? Shabbir, did you want to comment on that? We'll give the remote participation first?

     >> MUHAMMAD SHABBIR: No, go ahead.

     >> GERRY ELLIS: So, so that -- the data set that we're looking at is the transcripts from the session. So the main sessions -- and if you go back -- if you can switch over. But if you go back, back, back, back. To three -- three forward. Right there. So, we had 1,020 transcripts. If you see it started slowly with 11 transcripts from the first year. I think that's important. So even the first year, we had transcripts from the -- from the sessions. Going up to 114 transcripts in 2010 in Vilnius, Istanbul with 138 transcripts all the way up to Geneva and Guadalajara, 205 and 215. So, the transcripts is increasing. So, the amount of coverage for what that data set looks like is growing. If you look at the transcripts, that didn't cover the whole three or four days we were in Athens. It was only in certain sessions and so forth. But now we're getting to the point where I would say pretty much everything that is said in the main session or any of the workshops that has captioning is covered.

      So, we can see that -- we can look for trends and patterns in what is being said in terms of the content. Something else that we could mine would be the user list, you know, sort of participant list, where people are coming from, all of this text becomes possible data for us. So that's one of the reasons I like the kind of work we're doing so much. And we need more people who want to do this kind of work. Because there's an enormous amount that one could mine from these kinds of meetings.

      You can mine all of the submissions that came in for workshops. You could mine all of the regional IGFs as well. So, anything that's text could be used as data. So, there's a number of questions like that and others that we could explore. So, in terms of the people with disabilities that are participating, I don't think that's indicated anywhere on the participant list or anything like that. But in terms of the content across IGF that's related to disability, we could mine a whole range of those kinds of issues.

      Like I didn't get to show -- I was going to show you cyber security and how that as a concept unfolds, if there's a particular topic that you might be interested in, such as Universal design for example, or accessibility standards. So, you can build a categorization model simple or complex around that concept and see how that concept has unfolded over the course of the 12 years. I would be happy to work with you or anybody, Judy, to look for anything that you might be interested in looking for in this data set to see if it's there or not.

     >> GERRY ELLIS: Thank you, derrick, thank you, Judy, for the question. Time for one last one? Going, going -- gone.

      Okay, well, that leaves me -- everybody to say, really, this has been great. I really interested in the attention from the audience. I would like to thank all of our speakers. I would like to thank the IGF, the amount of work they did to ensure remote participation. I would like to note all of the accessibility services involved. It's amazing all of the work to go into this. ITU here beside me who helps with all of the hard work. Andrea, who's not here. She's always present, even if she's not here. Amazing amount of people involved just to make this happen.

      I want to leave you with one thought. We're talking about accessibility and how it relates to the main stream of society. And we find out when we're talking about accessibility and we say the technology where we are, we're talking more now to UX people, user experience people. Rather than just Universal design people. We're finding that more, to get people involved in the main treatment, Gunela was talking about it. Diversity is the norm in society now. It's okay to be different, because everyone is different. Now leave you with one last thought. Here in IGF, here in the room, or in your place of education, or in your place of work, leisure, your place of worship, whenever you come together with a common purpose, unity is our strength, but diversity is our wealth. Thank you all very much.