IGF 2017 - Day 2 - Room XXVII - WS145 The Internet of Things & Accessibility for People with Disability

 

The following are the outputs of the real-time captioning taken during the Twelfth Annual Meeting of the Internet Governance Forum (IGF) in Geneva, Switzerland, from 17 to 21 December 2017. 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. 

***

 

>> GUNELA ASTBRINK: I'd like to start by talking a bit about what this session will be about. Related to the Internet of Things as well as reducing barriers. This may include laws, heating, entertainment, security systems. This has been expensive in the past. Mainstreaming such systems.  However, interoperability with existing systems and user interface need to be taken into account. So new barriers are not created. This brings together experts for the private sector civil society organizations to discuss our policy standards and innovative design can help ensure the internet of things is inclusive of the parts of community. We have a great group of panelists.

 We have Paul Timmers with the commission on digital privacy security, et cetera. Maarten Botterman and Satish Babu, from India, the school of internet governance, et cetera. Julia was not able to join us for lack of travel funding and also Gerry Ellis has connectivity problems with the web conferencing system. But we have a great group of panelists and we want this session to be very interactive.

 So I'll be asking each panelist three or four minutes max of introduction comments and then we have questions, they're being discussed with a panel. But in each question, I invite people in this room and remotely to discuss. And sitting next to me ‑‑ sitting next to me is Andrea Saks from the dynamic coalition on accessibility and disability. So we have a great group of people here. We have Ivan Ing, the repertoire of this session.

 All right. Well, please Vint, would you mind going first?

>> VINT CERF: Well, thank you all for inviting me to participate today. There's a lot more to be said about IoT and disabilities and accommodations. Let me discuss a couple of important features, one of them is reliability.

 If things are not reliable, especially if you depend on them because you're disabled. They are not useful. I think I put a very high value on reliability. Similarly, safety none of us want a device that is safe that we're depending on regardless of whether we're disabled or not. One of the things I try to explain to my engineers when they're doing design is to keep in mind that disability's covering an extremely broad spectrum. Visual impairments range everywhere from moderate amounts of visual difficulty to complete blindness.

 The same arguments can be made about deafness, I'm hard of hearing. My wife is completely deaf, has two cochlear implants. People who are born deaf who might prefer ASL or other sign language. First of all, don't imagine you already know. What a disabled person needs in relation to your interfaces.

 Short messages from the mobile, they need to accommodate all possible means of communication. Now, that puts a big burden on the design of any one piece of equipment, especially low‑cost device. That leads to at least two other considerations. One of them is the notion that these devices do not operate. They may operate with near nearby devices that may operate in tandem. Although, I will tell you one of my highest priorities is internet accessible. And if that's not true, I'm not interested in your advice thank you very much.

 And the other thing ‑‑ even with an attempt at universal design, there may be situations that a person with a disability has, which is carefully triggered to assist them with whatever disabilities they may have. That with IoTs. That means you have constant global standards (fading in and out.) Appreciating the nuances and range of disabilities and how they manifest.

 At Google, I've been promoting the idea that our devices should get exposure to people who have disabilities, live with assistive devices every day to tell us, how did we do? How well did we get this interface? And are we really helping? And the last point I would make is that these devices are going to be filled with software.

 And as I have said repeatedly in the last couple of days as a person who is to make his living writing software, no one seems to know how to write software that does not have bugs. If there are devices we're going to depend on, whether disabled, or not, and if we anticipate those devices, we'll have mistakes in the code, we had better make sure there's a way of getting new code into those devices.

 And while we're at it, let's make sure it's the code coming from the right place and not the piece of (?). Articulate them in properties (audio fading in and out).

>> (Audio fading out).

>> Right. I hope I'm audible now. I would say that in countries like India, a large (fading in and out). That mobile phones (?). Multiple devices, (audio fading in and out).

>> It's still not working. Could I ask that you keep the microphone close to you?

>> Try the other microphone.

>> (?). It's a machine. Okay. (Audio fading in and out).

>> You have to get your mouth right on top of there.

>> (?).

>> Thank you, (audio fading in and out).

>> Okay. Well, when we continue. For now, I'll pass it over to Maarten Botterman to make remarks.

>> MAARTEN BOTTERMAN: Okay. On behalf of the ICANN board, I'm speaking here for (?) ‑‑ we just had a meeting. And (?) At the local level. Locally. And I think (audio fading out).

>> Thank you very much. (Audio cutting out).

>> You have to talk about large scope. And touch upon inclusion and accessibility, but only touch upon so there are many issues that many places where, perhaps, accessibility can get better into the picture in combination with IoT. The potential is there, but it does not happen. Yet, in terms over policy, there are quite of number of folks that are the same story. But it's not as anchored around accessibility. So Europe, for example, critical around IoT, is what are you going to do with the data that are coming from these devices from IoT environments? So there is an initiative around free flow of data. And that's really, actually, more about accessing views of the data. That's a perfect opportunity to move into where demands and understanding around accessibility and even timewise, that policy still needs to be fully implemented into the course of next year.

 If it doesn't happen, the policy is going to be there, and the moment in time has passed. Same story about what we talked about this morning. Certification of security certification. The accessibility can enter, and the moment is still there.

 You have to be very realistic, on the other hand. I was at web accessibility director in 2010, 2011. We started negotiating in 2012. Ultimately, we got a watered down web accessibility directive accepted into Europe in 2016. Four years and there is no IoT there. It still needs to be really applied to it.

 There's a relevance for IoT, but it still really needs to be interpreted, and some of that work to get it properly interpretive. My kind of conclusions, to be short here, is I would say TLR plenty of initiatives and in policy and in emerging policy and also, in research and innovation.

 But even if these are there, at the moment, they are really soft looks. And you have to be taking into account if it is about political decisionmaking on accessibility, it's taking a lot of time. It's going to go slow.

 And the digital debate is much faster. It's driven by innovation and it's driven by money. And capacity debate. The time is now, it's not trivial to make use of it. And actually, to debate around policy, digital policy is rather uninformed around accessibility. There's an opportunity to provide guidance for existing policy, how to into that, and guidance into policy negotiations, like around European accessibility. And then, finally, I find that there is not a really strong, informed ‑‑ strong organized, and strong resourced voice around IoT accessibility. That's a pity, but you'll have to work with limited resources, have to be smart.

 Honestly, I think some of the disability organizations could actually do more, be closer involved in innovation. Because innovation is actually also in a certain sense to realize rights or policy tool.

>> GUNELA ASTBRINK: Thank you very much, Paul. We have a lot of big issues there to consider and I think, again, we hear that this is the time. This is the time, really, to work on this. And to find a meeting of IoT. Experts to work together. Next, we have disability expert Shadi.

>> SHADI ABOU‑ZAHRA: Hello. I'm the technology specialist at the worldwide web of consortium WTC. So the topic of IoT is very important to WTC. Because we see the web as a platform you could see it as an application, living on the internet. But it has ultimately become the predominant interface to the internet. I think this could provide a huge opportunity. I think we've mentioned several times, IoT in itself is a huge enabler of ‑‑ I remember already years ago, people using smart glasses and using them to recognize faces which, you know, has all sorts of privacy issues.

 The idea was actually just to recognize the emotions, or the face expressions and it was a blind researcher giving a keynote presentation and saying, why are you frowning? And of course, it was a very ‑‑ strange situation when a blind person is asking somebody why they have a certain face expression. And the glass enabled that. By using the connectivity to the internet and all of the big data and stuff in the background that goes along with IoT.

 Huge enablers and I think, really, it's a potential here for very different kinds of assistive technologies and you see a whole new generation of the assistive technologies that could enable things very differently.

 One of the big challenges that I see is the interoperability. I would put that higher than reliability. The experience of playing around with the one of the home assistants and put it in my living room and try to just ‑‑ and it could speak with my television very well.

 But it couldn't turn on the captions. On the television. It couldn't know how to ‑‑ even though the television had the captioning capability and the device understood what I wanted to do, but it didn't know how to turn on that command. So this is not only the user interface level. This is the interfacing between the devices. Between the connected objects where we right now have, I think, a situation where more because of the business models, closed environments or semi closed environments. You have to buy the right lightbulbs to go along with the right device. And if you have the wrong type of phone, you may not be able to access your fridge or, you know, this lack is a huge barrier to assistive technologies. In a lot of these current IoT services is sometimes offer accessibility features but they'll have predefined ‑‑ you're deaf or you're blind or certain kind of categories. Deaf, blind or ‑‑ blind and having a disability or these combined disabilities that come often into play where only customized solutions and assistive technologies that we have currently on the traditional web, may I say, enable these custom assistive technologies, these custom access, accessing the data and the interfaces rendering the way the user needs the full access from the user interface level through all the cascades of different intermediaries on the way. Which we currently don't have on the internet of things.

 And this is where, again, WTC is looking at the so‑called web of things. This is not ‑‑ it's not in any way a rephrasing of the internet of things or ‑‑ but it's the idea of, like, the traditional web with the traditional internet. Could the web of things be the interface to the internet of things and be this interoperable platform where different solutions and systems could actually really thrive like they've done on the traditional internet.

 

>> GUNELA ASTBRINK: Thank you, I'll pass it over to Andrea to provide some remarks.

>> ANDREA SAKS: I thought, as I was listening to everybody, to share what happened, when you don't have certain things. I was a banker at the age of 2, also an expert on pharmaceuticals because mom and dad who were both deaf who Vincent knows, or knew, would take this 2‑year‑old, give it something to eat and say, I want you to listen and tell me what they say. And I would do this without voice.

 And they would lip read me. You know about this. Anyway. So I go back to, how people survive when they don't have these things. And there was a big drawback to that when I turned 14 years old. I was really not ‑‑ yes, I didn't want to do it anymore. But I had an allowance and that got cut if I was disrespectful, and that was quite often.

 So a lot of things, children, are often the people that create accessibility for persons with disabilities. And though, we ourselves don't think that's a big deal, there are drawbacks to that on many levels. I mean, bank's not going to ‑‑ when I could talk more, I knew the banker, so the banker would allow me to speak from my mother who didn't really want to get into that. But a lot of times, there's certain barriers that can't be crossed.

 You want to look at the human factor. One of the things that really excited me, a lot of people don't like the ITU because it's a weird place and I will agree. I went in there because they didn't know what they were doing. And I ‑‑ they just started a study group on the internet of things. And lo and behold, the two Korean guys who were absolutely delightful found me and I began to help them. And all we had to do was visualize what it would be like in a house ‑‑ what is it ‑‑ you have one of though talking things couldn't, you know, communicate with the television.

 Well, I have a cat that is the most audible thing in the world. I might come home and find there's a delivery of 4 pints of milk outside the door because the cat ordered it. It's not ‑‑ and it has a running conversation with me when I'm on the phone.

 And if I'm away, I say, put the cat on and the cat will sing because it knows it's me. I want people to think about accessibility is not something that people in high places decide upon or make policy, I want everyone to think about what it would be like for yourself and anybody you can influence, whether you know somebody who works in an industry. Because it's people who change what goes on. And a lot of times, people who make things have a direct connection with a person with a disability that either have family members or friends or something.

 And if those people can communicate to the people in industry who do not, that helps. The other problem is the misconception that it's not financially viable to do this. But I don't know. I get the idea, if people really got, it would be great to sit in your house, might get fat and do everything, including your housework without getting up.

 There are wonderful visions that could be done if people use their imagination. And that's how things change. I don't have any solutions, but I have a good imagination. Come up with your own thoughts and ideas and communication. And that's the way things change is through people.

>> GUNELA ASTBRINK: I think using imagination and that creates innovation and so, we have the opportunity in this room to talk about some of the potential benefits and applications for people with disability. So I'll open this up. If Vince would like to start with some of those ideas.

>> VINT CERF: Thank you very much, madame chairman. I wanted to react to some of the things so far. What is it about cats and the internet? They come up all the time. One of the things that drives innovation is discontent with the status quo. I look for people who aren't satisfied. I don't want them winding all over the place. But people who aren't satisfied with the status quo are the ones that make change.

 We should take advantage of the fact that everybody at one time or another is likely to be somewhat disabled whether you broke your leg and you're in a wheelchair or, you know, you have a cold and you can't hear. There are a variety of conditions that take people who are otherwise completely healthy and put them in a temporarily disabled category. It's amazing how that changes people's attitudes when they are suddenly confronted with, oh, my god, I can't get down this stairway and I'm in a wheelchair. Temporary disability is a part of the sales pitch. Keep that in mind.

 Another thing, which we have to worry about with regard to IoT is fire and forget kind of attitude. I built the device, sell it to you, I'm not going to have anything to do with it anymore because I got my money out of it. And too bad if there are problems.

 We can't allow that to happen. We have to have guidelines and rules that say, you know what, that's not acceptable. There is, however, this other wonderful positive potential. These devices are typically programmable. Which means, they can change. And my favorite example of this is my wife's Tesla.

 Because she gets into the car and in the morning, it says, guess what I can do now that I couldn't do yesterday because I've downloaded some new software? Now, honestly, it doesn't say that, but the screen pops up and says, here are the new things the car can do. This is a big opportunity. Software is creating flexibility that we didn't have in the past. And so, this was a good thing. It leads to customization.

 And someone mentioned 3D printing and things like that. That's an example of a technology that can lead to some customization of physical devices that are designed to fit your need. And so, we are edging into a time when we can do more customization than we could before. Either 3D printing, reprogramming and the like. There was mention, also, of the data that is accumulated by these devices or detected by these devices. And I want you to keep in mind that privacy is an issue here. A web cam which is not under control can transmit its images where the device is told to send them to. We don't want that to happen. You don't want your assistive devices to reveal private information that shouldn't be assessable to any you've authorized. So that's another one of those listed desirable properties that we need to keep in mind.

 Final point I wanted to make has to do with generality. Even if a device has been designed with a particular assisted need in mind, sometimes, it turns out that that also satisfies some other requirements. And I think of captioning on the television, which many of us fought very long and hard for. Turning out to be another wonderful tool for teaching of English as a second language, for example.

 And so, anytime you can figure out a way to take a special purpose device and turn it into something that has other applications, you may build a bigger market for it, which means, you can drive costs out and make it affordable, accessible in the economic sense to a larger community. Those are all things and instantaneously come to mind as this conversation comes unfolded.

>> GUNELA ASTBRINK: That's quite true when it comes to particular innovations that come out of assistive technology. We know the scanner came from the reading machine. People who couldn't use their arms to type on a keyboard. There's huge potential for innovation in IoT for people of disability that can have this very much wider application ‑‑

>> I think you're right. A little bit towards the future where all of the technology will become less visible for everybody, whether you have sharp eyes, or not. And indeed, the language interface will become more natural and developed for people with less may be useful for those of us ‑‑ I can see that it'll be easier and easier to operate the consumer machines because everybody has a problem with it ‑‑ I see some of the problems today with accessibility will almost naturally disappear, but let's keep our eyes on the ball and make sure that we also continue to help. It's not good enough.

>> VINT CERF: Our ability to understand speech to recognize speech and to translate it has gotten better and better. It's a hearing aid in an interesting sense because it hears in languages you don't speak. That's an example of what happens when you get very creative about powerful software techniques.

 

>> SATISH BABU: Thank you. Two aspects of what is mentioned by Vint, one is that sensibility in the era of mass customization. Makes sense to have customized kind ‑‑ we have a ‑‑ to your specific ‑‑ (?) We have all of these cheap ‑‑ like others, and so on. We can customize a lot at the grass roots level itself. And going on from there, the fact that manufacturers come and go, without records to maintenance and all of that can be communities that are trying to do this work themselves. We need to develop our skills within our own communities also. Thank you.

>> I'm from Bangladesh. So one thing, you know, our government have launched some innovation fund that's called (?) Innovation fund, it's around $30,000, anybody can apply for that. Any persons with disabilities. Innovation came through these funds. One is the WFS2017, that is the books ‑‑ accessible dictionary. We have developed a dictionary that is available online. Anybody can act success it. Primarily targeted to the blind people. 1 million people already use this dictionary. We target it to the visually impaired people primarily. And when, we make something accessible for all, that could be benefitted to everyone. To run a campaign to make an inclusive development. Product development and then everyone can enjoy the rights and we don't need any special accessibility. Thank you.

>> GUNELA ASTBRINK: Thank you very much, inclusive and universal design is the way we would like to go as much as possible. Andrea?

>> ANDREA SAKS: I was just sitting here listening to one concept. Do you remember when the television manufacturers fought tooth and nail not to put captioning in their television sets and now everybody does it in their television sets and it wouldn't be ‑‑ it's just not reasonable not to do it. If it is mainstream into regular society, we can look at the uses that would be applicable to us all, then we have a chance. They don't want to do niche things. Wouldn't you agree? And you know a lot about that.

 

>> I was one of the guys on the battlefield, particularly fighting a peculiar situation where a simpler design was being essentially denigrated by a more complex one, which had much more, you know, beautiful symbols and colors and all this other stuff. And in some of the television, not just manufacturers but also television producers who were resisting the use of the simpler technology.

 Essentially, trying to put off into the future a particular implementation. And we basically fought them down and I was very pleased about that.

>> GUNELA ASTBRINK: Yes, question, please, from Pakistan.

>> Thank you for the record, this is Mohammed Sabid. I have a couple of points. I think one of our earlier speakers mentioned that we actually need to know what we want to achieve here. Had a competition, mobile had evolved. Interesting thing about that was related to person with disabilities coming up with ideas that were developing applications that would address the needs, any needs of any particular person with disability group.

 Good, but they'll have to be addressed by someone. We launched a product. So we launched a campaign, a country by campaign. We organized workshops where we would ‑‑ we would bring together the developers and the person with disabilities and developers would discuss their ideas with person with disabilities. And see that if there is a possibility to cut the long story short, at the end, the application that was first and given the top price of the president of Pakistan was an idea based on IoT. And it was an idea that which would use hardware as well as software and it will be ‑‑ it was a smart house idea that you have a mobile add on your mobile. Person with disability who cannot use or cannot ‑‑ he can use that simple add using mobile, he can he or she can open doors on or off. Needs to stick together, come together.

 In my country, I have been a number of times called to test a device that whether this would be useful for person with disabilities, or not. After the device has been made. So I ask the first question to the developer that where your research came from.

 So I think it was very ‑‑ I'm sorry I'm covering the name. That we need developers and that person with disabilities need to stick together to formulate the way forward. I may have to use it. Is the awareness of what technology can do for person with disabilities and what person with disabilities can themselves do with the help of the technology? Three years back, before December 2014, bankers would not allow a visually impaired person in Pakistan to carry credit card to use internet banking. Now, see here, it is the money by the own person. And bank ere is requiring visually impaired person to come as a witness to draw his or her own money.

 So we came to the state bank. And to add insult to the injury, they would say that they would claim it's less for your protection. We are not doing it for ourselves, we want to protect you. So at the end of the day, we were able to convince them and in December 2014, state bank of Pakistan issued the guidelines where a person with disabilities or visual impairment specifically can carry credit cards and ATMs.

 Even now, when people with vision impairment go to the banks and demand Visa, which would say, no, this will not ‑‑ this will put you in more trouble. I would not be in trouble more than any other carrying this credit card. It's my risk if I want to take this, who are you as a banker, you have a right to advise me. But you do not have the right to stop me from getting a service. Thank you.

>> I had an experience like this. We were trying to do something at Barclay's bank. We were trying to do it remotely. And we were having trouble hearing the person at the bank with this weird British accent. If I offended anyone, it was on purpose. I was there with her. And I said, may I repeat the questions for her in case she is having trouble understanding you? They said, no. They refused. And I said, I'm her husband for one thing. Well, we don't know that. They said you had to use a teletype, which they threw those away 20 years ago, they were too slow. We didn't make any progress at all. They refused to deal with us.

 I took some action after that. But it's an example of people who think they know better than you do about what risks you are willing to take or what steps you're willing to take in order to work in an imperfect world.

 

>> GUNELA ASTBRINK: It accounts back to the independence.

>> Thank you. Francesca from the global initiative for inclusive technologies. I just wanted to make a little comment on what was said earlier. (?) This is the seventh year. And we actually discussed since the inception key topics.

 And we are going to include the IoT, we had approached that in the past. The IoT and consumers technology product, sufficient intelligence, robotics and so on.

 We actually worked with different partners. But essentially, also, with the consumer technology association. With the organizer of CES in Las Vegas. And obviously, many other partners. And very glad to actually glad to say that the previous announcement was ‑‑ this year we'll host the keynote speaker. So I just wanted to say that, yeah, there are initiatives. And hopefully, this could be of interest.

>> Lack of interoperability, a lot more can be said about usability. I think it's a particularly good topic to mobilize policy and nonpolicy tools.

 So kind of user supply corporation. Also open innovation, which can be stimulated by governments. You can actually get money for that. That's part of research and innovations. Only to these guys to these private/public corporation, there is to some extent a legal requirement for that.

 If you take data protection, at least in the European scene, there's a rise to data portability. And you cannot realize data portability if you don't think of interoperability. A means to put pressure on the system to realize interoperability as a legal obligation. Business does have to some degree an interest in it. The frustrations in the assistive technologies field is all of these companies are small and local. They don't actually go into the larger market. And the larger market is often fragmented.

 Well, in Europe, it's not committed to fragment the large amount. This is illegal. So interoperability is a legal requirement. Actually, all of the accessibility legislation. It's not actually in the legislation. It's in what we call an internal market source. An economic legislation. And then, I think the real interesting thing, again, to come back to that is, actually, what are you going to use that data for? Ultimately for service and maybe multiple industries of that. To some degree, some people are realizing, digital is going into the sectors and boundaries between the sectors are starting to blur between, say, energy and health is going to blur between energy and mobility going to blur. And so we will need to think about how can data flow from one sector to the other sector and bring back very much to interoperability issue. Happening, actually N the electricity sector people realize that if you want to do a good job in terms of sustainable energy productions or also consumers as producers of electricity, you will work with smart leaders.

 But hey, the smart leaders can also be used to transmit data that have to do with daily living. The mobility and the house with mobility outside the house with vital signs. And so, actually, the other is a real pressure on the system to say we might be able to use the same data in the system as in the health system or the independent living system.

 And that, again, brings it back to very basic interoperability that have good economic sound and interest in it.

>> I do think this is another very important piece of the puzzle. That is a key enabler. We had the situation in Austria where the national broadcaster was required to provide certain amount of captioning and they did. When they put those materials on the video on demand so online, they did not provide the captions there because they did not consider that as being broadcasting, until there was the legal interpretation also applies to streaming and that's also a form of broadcasting. It's what I'm saying is I think a lot of the accessibility features that we're seeing are not because of the mainstreaming aspects. I think this sometimes comes after when people are seeing the benefits.

 But there is a lot of policy directing that happens here, as well, that needs to happen, I think, also, in future technologies. The other thing I wanted to reflect on a little bit is the timing. I think we're really key point in time here where we have a unique opportunity accessibility, things get developed or technologies get developed, get deployed. And they are inaccessible and have to retroactively make them accessible. I think a lot of the questions that we're trying to figure out in how to make IoT accessible are challenges that are across the bench. A security, privacy and so on. These are not exclusively accessibility challenges.

 At the same time, accessibility, I think, can provide really interesting use cases here. For instance, regarding ‑‑ regarding privacy aspects. The fridge that automatically tells you what you need to shop, which I think is really handy, really good because I always forget. I always come back from shopping, having forgotten half the things I should buy.

 At the same time, it can disclose so much information to your shopping service. Just alone by your dietary habits, maybe even diseases that you have, disabilities and things like this. And this is something that maybe the mainstream may not immediately think of, what you're actually disclosing here. But by thinking of accessibility cases, starting, you always say a design if you actually have your target population, people with disabilities, you're going to actually get all of the rest for free anyway.

 So when you're making usability studies, do it with people with disabilities, you're going to get the extreme cases. Like when you're testing, the extreme things to try to break your code to debug it.

 And the rest comes for free. To the policy thing, a very interesting on the applicability. And I think this even though the CRCP is several years ago, but it's still the spirit of it still applies to the future technologies of IoT. And this is something that the community should be looking at more closely.

>> GUNELA ASTBRINK: Now, we have Gerry Ellis speaking remotely. Hopefully, we'll hear Gerry Ellis from Ireland.

>> GERRY ELLIS: Hello, Gunela, can you hear me?

>> GUNELA ASTBRINK: Really well. Thank you, Gerry.

>> GERRY ELLIS: Thank you, thank you.

>> GUNELA ASTBRINK: Great to have you.

>> GERRY ELLIS: We've been having problems with the WebEx. Just a couple of points, to extend out what Shadi was saying about interoperability. Making products financially viable. If you've got a country like India, which has many different languages, or if you look at the European Union for the 28 countries, soon to be 27, of course. You have to have harmonized standards across country borders and across languages and so on.

 And the way you do that is you work with international standards bodies. So it's very, very important that we access international standards. Like ITU, that's a very important task for us. The second thing I have going to talk about is IoT usually does not work by itself, it works with the likes of artificial intelligence and machine learning and stuff like that. We need to make sure that the prejudices that exist in the physical world are not perpetuated and not continued on in those areas in machine learning and whatever. And if their interfaces are accessible. The way to look at two key concepts of fairness and bias. And we need to look at those.

 And my final point is really who can gain out of this. Temporary disabilities. We go beyond that and we look at permanent, and we look at temporary, we look at episodic and situational disabilities. When I got GPS. I never missed by bus stop. I was just as blind before and after, I was less disabled in that situation. And another example, I was in a bar in Washington about 6 months ago where there were about 50 screens in a sports bar. But there was only commentary for one of those screens, the rest were all subtitled. Nobody in that bar was ‑‑ had a hearing impairment. Nobody in there was deaf, but every one of them used that access service. They were gaining benefit. I'll give it back to the floor.

>> GUNELA ASTBRINK: Thank you so much, Gerry for coming in remotely and making those comments. I'm so pleased we managed to include you. And, yes. That's great. Okay. We have another 20 minutes to go. And we have ranged across different areas. I'm wondering, we are talking about IoT community, we're talking about the disability community and there needs to be, maybe, closer links.

 I'm wondering, we have a lot of people here in the room. Are there ways that we can find to collaborate and, for example, we have Google here, we have a range of people from civil society, as well. And about extreme testing. I'm wondering, can we discuss any way that's possibly where you'll get that collaboration?

>> This is just an idea. Speaking specifically to the web type interfaces. It has occurred to some of us that it would be very interesting if every IoT web page. And the web page is only to the party that is supposed to be having control over that device. You can imagine some cryptographic methods generated to acquiring the device and getting a key created which you have access to. I bring this up ‑‑ it's an interesting possibility that the web page associated with the device is the way in which you configure it.

 There were regularity about the way in which you manage these devices. This is just not well thought out notion you sort of transfer authority to the web page who whoever requires it or ‑‑ it's little ideas like that that create commonality that can actually go quite a long ways toward making the manufacturers of these things create uniformity for you. Even though the controls may be different. We're seeing something similar right now. What they're supposed to do. Skills, Amazon and called actions approval in my company. There's a place called Schema.org where there's cooperation going on to create these sets of identifiers associated with functionality. Once again, a tool for creating commonality among the diversity of the companies and devices that are being created.

 I have one other anecdote to tell you if you can tolerate it. It has to do with the, I would say the proliferation of voice‑based interfaces. Everyone who has gotten utterly fascinated by talking to Alexa or Google home or Siri or something else. And I consider this to be interesting but not satisfactory. How do you refer to the lights you want to turn off and on? Do you have to give them names like "George" and" Frank" and "Eddie"? How do you introduce your guests to the house? How does the house know which guests are allowed to do which things.

 I don't want to have a lengthy debate with my house about which light I'm trying to turn on or which toilet I'm trying to flush. And so we literally need to think our way through scaling properties of a number of devices. So many engineers fascinated with mobile internet app device, end of story. And if you have a lot of devices at home that are internet enabled, you don't want to go flipping through 800 of them on your mobile to figure out how to open the refrigerator. I'm afraid not enough people are actually thinking their way through these deployment scenarios. There's parents and children, and sometimes they have the same responsibilities the parents do. Except, this day in age, kids know how to get around better than the parents do. It just gets more complicated now. When somebody comes to visit your home who happens to have a disability, how does the system adapt to that?

 And, actually, that's a scenario that I had not, frankly, thought about until just. That's a whole new dimension now. Where your house is all set. It works great. You know how to get the lights off and on and the heater to go and the doors to unlock. And suddenly, a friend shows up who needs assistance and needs in your house.

 And you didn't think of that until they arrived at the front door. Now what?

>> GUNELA ASTBRINK: Now, this is a wonderful opportunity to work on that mainstreaming, including ‑‑

>> Indeed, there's activities on web things. And what you're saying there, the idea only the web page ‑‑ and I know you know very well it is not a human readable page. A series of data flows. And you have the description for the objects. And it would have it ‑‑ would disclose its properties and what you could do. And that's the idea, trying to build foundations for other things to plug in and say, I have these lightbulbs here and I will abstract that. And instead of Freddy and all of these other names you mentioned, it'll be called, I don't know, evening mood and, I don't know, daylight. Something. You know.

 And simplify that on an abstract. And you could have these cascades. Mohammed, please.

 

>> Yeah, the comment about talking to the house is very interesting. I'm not going to comment on that. I will take you a bit back where someone ‑‑ issues related to the policy process and the IoT and technology development. Yes, policies sometimes, very big problems with the policies. That’s why I mentioned when I was there, I wrote ‑‑ I write in my ‑‑ this line that accessibility should not be the stand alone or an independent venture. It has to be what we see in the house is mainstreamed or it has to be incorporated in policy, planning and design. It has to be everywhere. It cannot be a stand-alone venture and then, again, we can claim that we have made our difficult things or whatever accessible.

 It would be much more difficult, much more costly. And time taking, as well. This is it. Thank you.

>> GUNELA ASTBRINK: Thank you. Maarten, please.

>> MAARTEN BOTTERMAN: No, I was thinking we ended our discussion into internet of things, safety, security and privacy. And what comes up for me, it's very important, also not to hinder ourselves to develop applications to help people that really need them.  I remember having seeing ads For an American firm where they had amazing homes that observes your parents 24/ 7 and has all kinds of tools to make sure everything goes right. And my first reaction was, shoot, I want to be in an environment like that.

 Now, my mother's suffering from ‑‑ what's it called? Alzheimer's? My father is struggling with keeping her helped. I can see how this is not a pleasant solution. This allows them to stay at home in a different way. If we legally exclude this is possible because whatever reasons, we also shoot OURGSs in the foot. So let's be sensitive for that. That's my call. On that.

>> The bridging between the IoT and visibility accessibility inclusion community, as I said in the beginning also, I think it's quite important that we have these cases that illustrate was actually the issue at stake. And I think here are several very interesting well illustrated case have actually been mentioned. And that needs to be communicated to the IoT community. And the key performance indicators for next year, talking about this topic. Well, I think I would like to see that it's not only in G3ICT, but other places where accessibility and IoT is really taken up. And why not, as a matter of fact, request these conferences that are going on to include this as a topic that could be a possibility together is a few of these cases that illustrate likely unrecognizable way personally what it is about because you have the personal appeal.

 And I must say, I think it's worthwhile to think a little bit about the words that we use. In another areas like privacy, privacy by design, people start to ‑‑ people get a picture of what that might mean, security by design. Here, for a long time, we have had the design for all. And I'm not sure whether that as a word sufficiently with the IoT community. Continuing to debate and sometimes, you also need to find the right word. Something used as a common term and you'll find it.

>> ANDREA SAKS: I have written to the ITU dictionary on accessibility terms. And every time, and actually, you have to ‑‑ we come to something, somebody writes a check to it. There's a situation that Gerry pointed out that is the term is universal design and not designed for all. Terminology is very important.

 We've gotten rid of things like deaf and dumb, we've gotten rid of that one. But there are plenty of other ones that we haven't gotten rid of and I have to keep reminding myself that what you say is so significant. You can do it in one language. We did it in English. Other languages, for instance, in French, it's handicape, if you say that in English, you're a dead person. I think we have to coexist with what we've got.

 The main thing I think is very important situation is to remind people who do not think they have a disability or do not have a disability that when they get nice and old and I freak out realizing how old I am. Because I'm 70 this year. How did I live to go on? And I find that parts are falling off. So I have a cane now and one replaced knee. And I got my glasses changed six months ago. I need another pair.

 The emphasis for young people who are in control of us today, they have to remember that they're going to get to have gray hair, bad knees and everything else. Accessibility affects everybody at some stages.

>> GUNELA ASTBRINK: Okay. Look, I think we've had a fantastic conversation here about IoT and accessibility. Thank you very much for all your input. We've heard about involvement of people with disability in design, working with developers the importance of reliability, global standards coding, involvement of accessibility in IoT conferences. And the free flow ‑‑ how timely is this to really incorporate accessibility in all of the IoT developments that are happening.

 We have the same issues in IoT discussions about privacy and security, ethical issues: They need to be addressed in accessibility for people with disability, as well. Really, various devices that should be talking to each other in various ways are really effectively designed to do this. And we've talked about terminology, the control of the devices. And functionality, of course, is very important.

 So there are so many issues here that we can go away with. And work with in our own organizations and give our own networks. Let's try and continue this discussion online so that we can mesh this general IoT development with accessibility and the dynamic coalition is a great space to do that.

 I'm sure I've missed things, but did Vint want to make any closing remarks? Okay. Any other last, last, last minute comments before we close? Fantastic.

>> Thank you, everyone.

>> GUNELA ASTBRINK: Okay. Thank you very much, everyone.

 

[ Applause ]