>> SUSAN CHALMERS: Good morning everybody.  I would encourage everybody to join us here so you can contribute to the conversation.

So it is my distinct pleasure to welcome you to our round table discussion.  Titled "digital inclusion through a multilingual internet."

In this workshop we will address three questions.  First question is:

What are the barriers that are keeping people from using the internet in their own language?

Second, how can we surmount these barriers through technical and policy coordination?

And third, in what Kay ways does internet multilingualism support the broader goal of digital inclusion?

Today we are joined by seven expert speakers.  Two of whom will be joining us remotely.

And a subset of each of our speakers will respond to each question.  So we will punctuate these sections with discussion with the audience.  But to begin, we will hear opening remarks from our expert speakers who graciously contributed their time to the organisation of this roundtable and agreed to share their knowledge with us today.

First, I will turn to Mr. Alan Davidson.  The Assistant Secretary of Commerce for communications and information and NTIA administrator.

>> ALAN DAVIDSON: Thank you Susan.  It is wonderful to be here this morning.  Early on our day 2.  And thanks to those who are joining us online as well.  And I really wanted to say how much we appreciate our hosts here in Kyoto for having us.  To the IGF for including this session and all my fellow speakers.

Starting point, we're here today because connecting everyone to the internet is first step in building a really inclusive digital society.  And we know that being able to engage in one's own language online is key to meaningful connectivity.

Making the internet multilingual will support that connectivity.  That meaningful adoption and digital inclusion.  And that is what we're here today to explore.

At NTIA our starting point is something I think is a intuitive to a lot of us who are here.  Which is that the internet is now the essential communications medium in our modern world.  It is essential for access to economic opportunity, to job, to education., to health care online.  To the world of knowledge.

And what is amazing, of course is that here we are in 2023 and there are still millions of people.  Millions of people in the U.S., millions of people around the world who are not online.  And it may be because connectivity is too poor.  Maybe they lack the means or the tools.  Or the devices or the skills to get online.  Or they may not find content online that is compelling to them.

That has too change.  Is and I'll say domestically in the United States working to tackle this problem in a big way.  The U.S. government has dedicated $65 billion in new funding to invest in a simple and ambitious mission in the U.S.  Which is to connect everyone with high‑speed, reliable, affordable internet service.  Internationally we're trying to do our part of well.  Part of the connect initiative.  And initiatives like anyway are helping us form the partnerships and mobilise the resources to connect the unconnected around the world.

But we know that connectivity and access is just the baseline.  We must both establish connectivity and make that connectivity meaningful.  Our ultimate goal is to help people thrive online.  And that comes through our topic today, which is making sure that people have the ability to use the internet in their own language.  Everyone online deserves access to a digital sphere that is diverse and inclusive and serves their needs.

At NTIA we have some experience now with language and connectivity.  Most recently in our tribal access programme, our tribal broadband connectivity programme.  The programme dedicates $3 billion to improving access and adoption on native American tribal lands.  And an important part of this programme that's been really interesting for us is that we're funding remote educational activities.  And that includes work to preserve indigenous languages.

This is crucial.  In the U.S., there are 245 indigenous languages.  65 of them are already extinct.  And 75 of them are considered near extinction.  With only a few mostly elder speakers left.  And our awater dees in this grant proposal are connecting their tribal language and culture resources to the internet and digittizing tribal language materials.  We're funding a tribe in California that is offering online classes in native American learning and traditional skills.

We're doing something similar in Hawaii, where we're got a grant for a community that is hiring native Hawaiian language specialists to collect and translate and record public records and stories and put them online where appropriate.

In short, what we're trying to do with this tribal programme is to connect native communities to the internet in the languages that they speak.  And ensuring they can engage with the internet in those languages.

You know, a lot of times people think this is about making sure that these communities get access to the resources that the internet offers.  One thing I think we've also seen is this is about sharing their culture and their local knowledge with the rest of the world.  So this isn't just about these communities somehow being able to access.  It is also about all of us being able to understand and access their culture.

And that's been incredibly meaningful in this programme.

To conclude I'll just say we believe that every community deserves this same opportunity to meaningfully connect.  Our experience has shown that connectivity to the internet becomes much more rich and meaningful when you can connect in your own language.  And while progress has been made, as we know and we're going talk about today.  The internet is far from multilingual.

But together, we can change that.  And particularly, coordination around the promotion of universal acceptance.  The technical foundation for a multilingual internet is going to be a critical tool.  And universal acceptance is a technical standard that enables domain names and e‑mail addresses to function in non Latin scripts.  Promoting it is a foundational element of a multilingual internet.

So it is very exciting to be able to participate in this roundtable, alongside some of the subject matter experts at institutions who are focussed on driving universal acceptance forward.

Thank you all for being here.  And I'm going turn it back over to my ‑‑ I'll turn it back over to Susan and she can introduce the next speaker.  And I will say the folks at home, I apologise.  I'm going to have to leave shortly.  But it is wonderful to be here and we really appreciate this work together.

Thank you.

>> SUSAN CHALMERS: Thank you so much.  I would like to turn next to Ram Mohan is who we had scheduled.

(off microphone)

Oh.  Online.

>> Yes.

>> SUSAN CHALMERS: Okay.

(off microphone)

>> SUSAN CHALMERS: May we unmute Ram Mohan who is joining us online?  Ram is going to speak to the background and give some context to the evolution of the I subject matter of universal acceptance.

>> RAM MOHAN: Thank you, sorry for the technical troubles and Secretary Davidson.  Thank you for your comments as well.

So universal acceptance, if you look at that as a general topic.  The way it has been introduced so far, most people think of it as a technology topic.  As something that we we think of as how to make domain names interact with the domain name system.

In reality however this is really a human problem and a human topic.  How do you make systems, computers and the way we communicate, how do we make various communication systems interact with each other?

You know, I think it is quite ‑‑ I think it is quite remarkable that if you type in your e‑mail address to register for a website, that e‑mail address sometimes will work.  Many times it doesn't.

The reason why it doesn't is not because the e‑mail address is wrong or is incorrectly formatted.  It is because somebody somewhere decided that the last part of the e‑mail address, the one that ends in .jp or the one that ends in .us.  Somebody decides it is not valid and therefore you cannot actually use the systems.

So if you look at that really is universal acceptance.  How do you make all names that are valid on the internet, how do you make them interoperable and work with each other?  But there is one level above that which is the issue of inclusivity.

If you have names that don't interact with each other, if you have systems that don't work with each other, how can you genuinely say that we are building a internet and a human society that includes all digital natives?

And that, ladies and gentlemen, is really the next generation.  Everybody who is in the room has learned about the internet.  But everybody who has been born today and who is going to school today, the internet is just a native digital thing for them.  How do we make sure that those who have a digitally native Japanese address, a digitally native Arabic address, how do we make sure they all work and interact together.

The solution to that is not technical.  The solution to that is something that all of us have to work together to build policy and other governance systems that encourage the universal acceptance of names.

>> SUSAN CHALMERS: Thank you, Ram.  That was excellent.  I really appreciated how you tied all of this together.  And just to note, that Ram is it chief strategy officer at Identity Digital.  My apologies, Ram.  I should have introduced you beforehand.

I think next we may turn to our colleague Dawit Bekele who is the regional Vice President Africa Internet Society.  Dawit are you able to join us.

>> DAWIT BEKELE: Yeah.  Can you hear me?

>> SUSAN CHALMERS: We can indeed.

>> DAWIT BEKELE: I'm sorry, I'm joining from my phone.  So I apologise if the sound and picture is not very good.

Thank you Susan for inviting me for this meeting which I believe is very important.  Good afternoon, good morning everyone.

It is a great pleasure for me to talk about an issue that is very important for me.  Since I speak one of the languages that are not considered very prominent on the internet.

I believe that access to the internet is a critical need, if not a human right in today's world.  Long gone are the day where is the internet was a nice to have tool.  It is a must have tool for almost anyone around the world.  Since many things that we used to do online ‑‑ offline, I mean.  Like government services, education, work are moving online.  And unfettered and unprecedented speeds all around the world.

But too many people, more than a third of the world's population, don't have access to the internet.  Almost equally worrying, too many people amongst the so called "connected," don't have meaningful connectivity.  According to the ITU, meaningful connectivity is a level of connectivity that allows users to have a safe, satisfying enriching and productive online experience at an affordable cost.

For example, in many places in Africa, the cost of connectivity is too high that many cannot use it for important activities, such as learning online.  And of course the content you want to access, if it is not in a language you don't understand, your connectivity is meaningless.

The world has made great progress on connecting more and more people in the last details, which is encouraged.  We should all work together to connect the unconnected, including those that live in remote areas in the disadvantaged communities like indigenous people that Mr. Davidson was talking about.  Et cetera.  That are not interesting for commercial operators.  But we should also work to give everyone this planet meaningful connectivity.

That is an internet access that is affordable, safe and that she or he can use to improve her or his life.

Thank you.

>> SUSAN CHALMERS: Thank you so much, Dawit.

Now to continue our opening remarks, I would like to turn to Edmon Chung, chief executive officer of we og Asia.  And.

>> EDMON CHUNG: Thank you.  This is a topic dear to my heart for those who know me.  This is a topic I've been championing for the last 25 years.  Why is it taking so long?  Well, let's talk about that.

And building off what Secretary Davidson mentioned as I a really very important part, which is the universal acceptance really is a foundation towards digital inclusion.  And to unpack that a little bit, a foundation towards digital inclusion is a fully multilingual internet.

And universal acceptance makes a fully multilingual internet.

And today, what is still I find startling, is that almost 60% of the internet's content is still in English.  And that doesn't reflect the real world.  There are almost 7,000 languages around the world.  Many of them are here in Asia where we are.  And really a multilingual internet is essential for trying to bring the next billion to come online.  Because that is the group of people.  Those are the people where English is definitely not a first language.  And in fact they might not know English very well.

So for the next billion to really be able to meaningfully come online, internationalize domain names and the universal acceptance of them is really a foundation towards that.

One of the things that I often talk about is that this is actually building on what Ram said.  Is that we have gone beyond the technical items.  You know, in fact the last 25 years, that's why it took so long.  First 10‑12 years, we were struggling with ‑‑ not struggling.  Really working hard to build the technology that allows internationalized domain names and e‑mail addresses to work on the internet in a secure and stable manner.

Then we spent, 10, 12 years working through the policies because the policies are equally important.

Using just 26 characters plus the 10 digits for domain names, you know, have its own challenges.  But try to bring, you know, 10s of thousands of characters into that system.  And there are other linguistic issues that need to be dealt with through policies.  And that is what we did in the last 25 years.

So now is the time.  I guess the rubber hits the road.  And we need to keep driving it.

And the other thing that I always like us to think through is that we really need a kind of movement.  Because this is not ‑‑ there is what I would call a kind of a marked failure right now.  If you ask people who don't know very much about the domain name system or the internet technologies.  When you explain to them that they can actually use Japanese, they can use Indian, Hindu, other languages to navigate the internet, they are like "sign me up."

The problem is that that is a latent demand.  Because they don't even know that they can demand for it.

The other side of the story is also suppliers are also unaware of the impact they are causing.  I'm not saying that suppliers are not aware of the technology part.  That through the 20 years, I think most of the developers, suppliers of webbed hosting, registry, registrars do know.  Registries and registrars duplified I.D. and registration.  But are they ready for web hosting and international e‑mail addresses.  Universal acceptance readiness is still very low.

But the technology part, really, ask Google, Gmail is actually internationalized e‑mail address ready.  Outlook is also ready.  But does that mean the entire suite of Google services?  Does that mean the entire suite of Microsoft products are UA ready?  Unfortunately that is no.

What is interesting, adding to my rhetoric about a needing movement.  I recently came across, I think Secretary just mention ad little bit as well what is called ‑‑ well he didn't mention it.  But he touched on what is "language justice."  We've heard of climate justice and many times of justice.  There is also language justice as well.  And why we need a movement and multilingual internet is important.

And I think the technical community, some of the technical community is still resistant.  It ‑‑ they have been resistant from the very beginning.  One of the things being a lot of the technology providers are still very much English centric.  And because of that they are sometimes worried, you know.  Well if someone uses a multilingual e‑mail address, am I going to be able to see it?  Is the administration going to be problematic?  Is it going to be difficult for me?

There is a little bit of resistance.  But we need to convince them.  We need to convince them that their work, their engineering work on all of the interfaces, all the display, all the storage, all the processing has an impact on language justice.

And also, if you look at the statistics, while of course people say because nobody is using internationalized domain names, there is not much abuse.

But if you look at the real statistic, DNS abuse is a fraction in terms of internationalized domain names.  You know, it is a much lower percentage.

More important thing to convince the technical community, I believe, is to tell them that a multilingual internet actually makes for a safer internet.  You know, my dad, going online, when he looks at the URL bar, he doesn't understand it.  But if it is in Chinese, he'll be able to figure it out.  For a normal user, actually a multilingual internet makes for a much safer internet.

So I want to start you with this thought.  Many people today, it is all about the content and the domain name part and e‑mail address.  People don't even use.  But if you think about it, the DNS was introduced in 1983.  When was the web introduced?  That was 1989.  Six years ahead of time.  That is what set the English‑only naming system and paved the way for English‑dominant web.

So maybe it is a hindsight thing.  But if you have a forward‑looking perspective, I think you can understand that the foundations of the naming system does pave the way for content.  And that is why without universal acceptance of internationalized domain names, a multilingual internet cannot be realised.  And it is a matter of language justice.  And that is why we need a multilingual internet movement.

And I hope I can convince you to start here.  And this is the community to start that movement.

Thank you.

>> SUSAN CHALMERS: Thank you so much Edmon.  A number of compelling points there.  And in particular, the multilingual internet makes for a safer internet for many.

So thank you for that.

And I would like now to turn to the Senior Vice President of Global Domains and Strategy Theresa Swinehart.  Here Theresa.

>> THERESA SWINEHART: Thank you.

So first of all, it is a pleasure to be here.  And to hear the remarks.  And Edmon, your passion is fantastic.  And the part he didn't mention is he heads up also the board working group in IDEA and UA.  So we had an opportunity to work together on quite a few things.

So from an ICANN standpoint, we share the importance of the inclusivity.  And the opportunity for everybody to use any language they want in any script they want.  And this is an incredibly important aspect that we've heard in the other remarks today.

And to the points that have been made, it is not that the technology isn't available.  It is actually the ability to use it.  Around the world, for whichever languages.  And we heard some different statistics but I'll throw a little more out and then share where we're coming from as an organisation working with the community and the board on this.

So apparently, you know, there is about 5.4 billion people, from all different cultures and languages around the world.  There is about 2.6 that have never been online.  But the power to use your own language, to express yourself in your own language is incredibly important.

We all know it.  We have to experience a conversation in a language that is not our own, it is challenging it.  Takes a different kind of energy.  It takes a different form of recognition of the words and what they mean and the interpretations around it.

So if we look at the currently about 6500 languages that are around the world, but we look at the breakdown.  There is about 2,300 spoken in Asia.  2,100 spoken in Africa.  1,300 in the Pacific.  1,000 in the Americas.  And about 280 in Europe.

Plus what we heard earlier within the United States.  Even local tribes and linguistics there.

It is not just about people using their own language.  It is also the preservation of languages.  It is part of the global understanding of different cultures.

If you then look further there is about 1.3 people speaking Chinese.  The majority of which, approximately 900 million mandarin.  Many of which are outside of China.  610 billion people speak Hindi.  And another 300 million speak Arabic.  This is around the world.

These are potential users that may use the internet even more if they can express themselves in their own languages.  But to the universal acceptance standpoint, you may have an address.  You may have a domain name.  But you may not be able to use it on platforms you want to or ensure there is a receipt from the other party.

So the work around universal acceptance, the awareness and the opportunities, is an incredibly important part.  It is not just an economic aspect.  And the statistics around the economics, you know, indicate, you know, $9.8 billion growth growth opportunities.  That is from 2017.  But it is also the societial opportunities and the next users from that standpoint.

So from an ICANN standpoint, we have a few things that we've been engaged in.  One is the support for Universal Acceptance Day.  And we can talk about that later.  That had a large impact.  And looking forward to partnering with many different organisations in the future.

And then we also have what is referred to as "the next round of Top Level Domains," which affords opportunity for those that wish to apply also for a name that is in their own language and in their own script.

The application process will open up.  There will be opportunities there.  Hopefully for all the 6,500 languages in all the different parts of the world, whether they are small communities.  But the important part is that universal acceptance exists so that those names can actually resolve fully at a global level.

So I'll leave it there.  And then look forward to talking a little more about some details of things that we're doing.

>> SUSAN CHALMERS: Cheque.  I'm all set.  Thanks.

Thank you so much.  So we have just had an excellent, I think, set of introductory remarks that covered the history and background.  Giving context to universal acceptance.  The importance of meaningful connectivity.

And thank you to Dawit for defining what that means.  So it was a very useful definition.

The importance of UA to non English or non Latin speaking countries.  And also really the foundation of UA towards digital inclusion.  And some very helpful statistics.  I thought those were quite important.

So with all of that said, I might ask folks who are in the room.  And again, please feel free to come sit at the roundtable if you are in the seats.  You are most welcome.

But I'd like to just turn it over to attendees to see if there are any reflections or contributions that they would like to make.  And then we'll move into the first of our three questions.

Does anybody have anything they would like to contribute?

Please.

>> AUDIENCE: Hello.  My name is Kasike.  Universal acceptance is important.  And it is annoying that you cannot use internationalized e‑mail addresses in various applications.  So I completely agree with many of you here.  But on the other hand, we don't really use domain names for identifying information.  And identifying yourself in e‑mail.

We tend to use Facebook, twitter, or other forms of social media.  Where you do not particularly, typically use domain names or lengthy identifiers.

So my question is, is domain name or internationalized domain name still relevant in making internet more diverse and multilingual information space?

And another question is, how people ‑‑ do users prefer to use internationalized domain names as opposed to other forms of domain names like GTL Ds and CCTLEs?

I have a done a survey on user behaviour on internationalized domain names compared to other types of domain names a couple of years ago.  And I found out that they have clear, distinct ‑‑ users have different behaviours over GTLDs and internationalized domain names.  So if you talk about universal acceptance, you have to consider how people react to multilingualism on the internet.

So this is my observation.  Thank you very much.

>> SUSAN CHALMERS: Oh, I think we have a hand up online.  Perhaps to respond to this question.

From Ram.  So I heard two very interesting questions that you have posed.  The first relates to, kind of, the meaning of identity and how that is expressed on social media and other platforms versus a domain name.  And then the second is pertains to IDNs and GTLD and CCTLDs.  I'm happy to have somebody also address that question.  Because I'm not sure they are necessarily so distinct.

But let's turn to Ram.

>> RAM MOHAN: Thank you, Susan.  And thank you for a terrific question.  Two things that we should think about.

It is not only about e‑mail addresses and a domain name.  Think of writing something in a word document.  And you are typing in an e‑mail address or you are typing in a website address.  Today it automatically becomes a hyperlink.  If it is validated.  If it is supposed to be a real name.

However, if those systems do not understand the identity that you are typing.  Then your company's name, your organisation's name that is represented on the internet may simply not convert automatically into a hyperlink.  So that is part of universal acceptance.  And it to some extent is not just about "in your language."  It is about making sure that whatever it is represented.  And whether it is in your local language or whether it is in ASCII in the Latin script.

As long as it is a valid name that is accepted by the authorities on the internet, it ought to just work everywhere.  Right?

So we have to I think start thinking about these less and less as identifiers.  And more and more as ways of communication between human beings.  And whether it is a website address, whether it is a domain name address, an e‑mail address.  Or if you go to many of the social media sites right now, they give you the opportunity if you want to link to your biography, for example on a site like LinkedIn.  They give you the opportunity to type in a url.  A website address.

However, that website address is not recognised due to universal acceptance issues, you are not included into that digital ecosystem anymore.  You are forced to go to a, in many cases, inferior system to represent who you are.

>> SUSAN CHALMERS: Thank you.  Thank you, Ram.  And Edmon, please.

>> EDMON CHUNG: I just want to add a little bit.  I think those have very good questions and raised a number of times.  Whether domain names are still relevant and e‑mail addresses still relevant.

Is I'd like to say three things really.  One is that, search, if you think about search.  Typing in a local language is not a problem at all.  So typing in an IDN has no challenge for the local community.  So that is very important.

And direct navigation on the internet is still utilising domain names.  And so that is I think remains relevant.

Number two, on e‑mail addresses.  I think most, if not all, of the cybersecurity training that I've given and I've listened to startses with email address.  The first thing I teach my dad is whether spam or with the scam is to look at the e‑mail address that is being sent.  And that is why, you know, it is still relevant.  And that is why I think you know what I mentioned as a multilingual internet makes for a safer internet because that is the fundamental.  Because that is the first thing you teach an elderly using internet.  And say, stop being scammed, is to look at who sent you the e‑mail first.  So I think that's important.

And the third thing I want to mention is that in terms of the users, I'm not sure if you saw the ALAC end user survey, the large end user survey that is commissioned at ICANN.  And the findings are interesting.  I guess the findings you have is relevant.  And corroborates.  But there is an additional question that was asked that I think was really revealing which I touched on earlier.

For people who don't know that there is IDNs and don't know there is an option.  When they first heard it through the survey they were excited about it.  They want it.  But those who actually know about it says it is not so useful.  Well, that is a revealing issue, right?

The reason why, and that is why we're talking about it here is that they registered and they couldn't use it.  Was the web hosting ready?  Was the e‑mail ready?

The problem is not so much that end users doesn't want to use it.  The problem is it cannot be used very smoothly on the internet.  So it is not set up properly.  It is not, you know, cannot be properly used in a good way.  And it is sort of like a second class citizen right now.  And that's I think the bigger problem.  So that study I think is quite revealing when you look at those statistics as well.

I couldn't find it right now.  But I'm sure we can dig it up.

>> SUSAN CHALMERS: Thank you, Edmon.  Theresa, would you like to contribute?

>> THERESA SWINEHART: Yeah.  I'd be really interested to see the study.  Because it is, the points being made are really quite revealing, also from another angle.  Right?

And there is the question of demand.  Or why may there not be demand for it or awareness around it.

And then to the points Ram and Edmon had made.  It is also about the user's seamless experience.  We've all had experience with, I don't know, something online.  And then it turns that the interface to something else doesn't allow it to happen.  So you have to go back and do something else and then go back and make sure it works.  Whether it is a payment method or whether it is something else.

And so in the end it is really about also ensuring that regardless, if the unique identifier is at the front end or in a document, or a reference tool, that the experience for the user has the opportunity to be seamless as part of a global internet from that standpoint.

And, you know, there's ‑‑ I appreciate.  There's areas that are within ICANN's remit more specifically.  And then there are things that are not.  And obviously enjoy facilitating and being part of conversations and partnering with others around the awareness there.

From an overall standpoint, I think that there is also the elements of "how does one contribute to the user awareness?"

And you know, you get a new laptop or you get a new phone, and it says, you know, do you want to use English or German or whatever language it might be.  And do you want to have, I don't know, Siri, and payment online all these things.  And even the preferred language, and is it going to work with others.  Some forms of questions.  Creates the user awareness that they might have the option.  And this is an area out of my purview.  And certainly not in agreement from ICANN.  But more generally the global level, the awareness of what opportunity could I have to do something I think is valuable.

And sometimes it is ‑‑ sometimes it is hard.  But it is about what is the right thing do for a global level to enable people to use their languages of their choice.  But the study is very interesting.  And I've, sort of, gone on.  But I think in the end it is really about the seamless experience for anybody to do what they want to the right of the dot.  Or to use the languages and scripts that they would like to use.

>> SUSAN CHALMERS: Thank you Theresa.

Now I'd like for us to move into our policy questions.  The first question here is "what are the barriers that are keeping people from using the internet in their own language?"

And I'd just like to turn first to Dawit.  And then we will turn to our colleague Akinori Maemura who is the General Manager internet development of the Japan Network Information Centre.

So we'll first hear from Dawit.  And then Akinori.

>> DAWIT BEKELE: Thank you Susan.  Unfortunately many barriers contribute to the challenges people face in using the internet in their own language.  Here are some of the most important ones.  My opinion.

First, if you don't have access to the internet, or you can't afford it, you can't use the internet in any language.  And the divide is very often on linguistic lines.  For example, as you have heard at the opening, indigenous people often lack connectivity that have an impact on the use of their languages on the internet.

Some languages have limited digital content due to social, economic and political reasons.  For example, most of them more than 2,000 languages that Theresa mentioned in Africa, are not organissed as official languages in their own countries limiting support for content development in those languages.

Third, the barrier can be technology.  I'll mention only two of these technological barriers that used to be barriers for my own that uses the (?) script.  This means before that it was close to impossible have content that everyone can read easiably.

Thankfully that script and many others are part of unicode.  But this is not case of all scripts and alphabets.

Another major technical barrier is the lack of support for many languages by devices and platforms.  There are many people who are literate in their own language only for which their devices and platforms they want to use are not localized.  Can limits their use of the internet.

The fourth barrier is the lack of digital interest.  Unfortunately the poorer the community, the less it is digital literate.  Thus poorer communities tend not to have their languages on the internet.

There are of course many other barriers for which I don't have time to discuss in detail here.  Those barriers include but are not limited to limitations of ideas as said by other speakers, lack of relevant content, internet shut downs and and restrictions.  Saturation of the internet with global languages.

But in positive note, even though the barriers are many and are often huge, there are some hopes.  Technological advances are making access to content in other languages possible through automatic translation.  And we have seen it with AI recently, even interpretation.

Governments, tech companies and other service providers are taking more and more seriously the issue of localization.  There are more and more technical advances and solutions, such as community networks that can provide connectivity, even for the most remote communities.

Therefore, if governments tech company, local communities, Civil Societies, and international organisations, and many other stakeholders work together, we can create a fairer internet for all languages.

Thank you.

>> SUSAN CHALMERS: Thank you so much Dawit.

And I just want to pause to look around the room and to see if anybody has any similar kind of experiences that Dawit had just outlined and would like to share.  If ‑‑ yes.  Roberto, please.

>> AUDIENCE: Thank you.  Roberto, from European user association.

I would like to the point.  Because when we think about underserved communities in terms of languages, I think that most of us think also situations like Africa that was presented or South America.  I would like to make an example that is related to my country that is Italy.

In Italy we have several languages.  I'm not talking about dialects.  I'm talking about real languages.  That are not officially recognised.  Most of them are not.  A couple are.

But I'm talking about futlang and sardo and abresh.  And those are local communities.  And I think that there is a role that even in a developed country we're supposed to be, like Italy, there is a role that the internet can play in terms of allowing favouring local contact, favouring the distribution of information in that language.

A few years ago we had at the IGF Italy a session about this.  About what is the impact of the internet on local communities who speak a language that is not recognised.  And that was quite interesting.

Why is this a policy question?  Because without, for instance, help by the government, without also the awareness of the community, this cannot be done.  I'm talking about these languages.  I call them "endangered" languages.  Because also with the globalisation, the community that speak that language is going to shrink, unless we support this.

And I think that the richness, the diversity of culture has to be preserved.  And there is a role that the internet can play in this.

>> SUSAN CHALMERS: Thank you so much Roberto.

And now, oh, here, please, go ahead.  If you could introduce yourself that would be great.

>> AUDIENCE: Hello.  My name is Rafik, I work at Internews.  And one of the topics I work on is content moderation in non‑European languages.

And I think that there is a really ‑‑ there is a kind of an issue where you have ‑‑ when you don't have a lot of content on the open internet in a specific language.  It tends to drive people towards platforms.  You know, Facebook most obviously.

I spent a lot of years living in Myanmar, which is kind of the textbook case for Facebook becoming the kind of stand‑in internet.  And I think that that's a really important phenomenon in this wider discussion.  Because to Edmon's point about a multilingual internet being a safer internet.  In these small languages, you don't have functional content moderation, I think.  A platform like Facebook has content moderation in maybe 75 languages.

And you do have people using those platforms in, you know, a thousand different languages.  And you have very robust communities using those platforms instead of the open internet.

But, you know, effectively in unmoderated way.  But still with all the amplification and things that you see through those platforms.

So I don't know if that's kind of within the remit of this discussion.  But I think it is a really important point.

>> SUSAN CHALMERS: Thank you.  I think that your point kind of dove tails with our question earlier from our Ling wiz colleague.  So that adds a really interesting dimension to the discussion.

Yes.  Please.  You are welcome to.

>> AUDIENCE: Thank you.  I want to share a view from regular users perspective.  One of the main barriers actually are lack of demand due to I would say like linguistic fitness of input methods.

I would like to say, for example, I use a country code as example.  If I type it on the keyboard, it is only three buttons, and I solve the problem.

If I use my native language, I would use a dozen.  I would have to press a dozen buttons.

So for any regular users, if three buttons can solve the problem, why would he or she go for a dozen?

So it is a simple question of for the regular users which is the linguistic fitness to our input method which is a linguistic technical issue.

So for any language that is easier to transfer into alphabetic, that can be contained in a small keyboard, it will be easier to promote IDEA.  Or from language that is not very fit to the alphabetic, that would be difficult.  So, and consider today, I would say the most majority input method is still alphabetic keyboard.  So that would be creating the lack of demand to regular users.  Since shorter and easier would be always the choice.  It is simple concept.

And if in the future like, vocal input or even brain input were promoted, that would be much more easier.

Thank you.

>> SUSAN CHALMERS: Thank you.  And I think the question of, kind of, practicalities is a useful one.

On that note, I was wondering if Akinori might be able to turn to you?

>> AKINORI MAEMURA: Thank you very much.  Akinori Maemura from Japan ‑‑ centre.

Pick up the various aspect of the idea and universal acceptance.  The barrier for the using the internet in languages is quite good question.

And then I would like to talk about a little bit, you know, past situation.  Thirty years ago we started using the internet and started using the PC.  At the time I remember that my PC, my system itself, always some other environment of which I used is not really good to handle the Japanese.

And then there is a lot of funny characteristics in my script for example.  The black square that means that some character which the system cannot handle.  So it is very hard to use the internet by Japanese because of that kind of errors.  And then it is ‑‑ it actually overcome later.  And then we don't have ‑‑ almost virtually no problem to use my own language, which is Japanese, in the PC and IDEA systems.  So it is improved.

And actually the professor Kamiwura who mad presentation before for profession of such inclusion of the languages for IT.  So he could be ‑‑ he could make good input for that.

But in such way that the PC system, for example, the OS and then computer system gradually include lot of languages to their own use.  Now as well Japanese.  Many languages are available for in the PC.

And then for the domain names, I need to tell that (?) is already ready.  And for the JP domain name.  We can use is Japanese level the second domain level.  Jp.  But not popular.  For us, Japanese, normal Japanese people has no problem to read and write the alphabet.  They are already our own, within our own knowledge.  And then it is actually preferred to use the alphabet for the kind of identifiers.  It is simple.

Very compact set of characters.  26.  Plus alpha character to identify something.  It is actually suffices.  And that ‑‑ and simple.  And sometimes cool Japanese feel like that.  So ASCII character is preferred for that kind of use.  It is not a natural language, you know.  For example, literature.  Literature we do need the Japanese.  Lot of characters exceeding the 10,000.  Needed it for smooth and fine literature.  But that is identifier.

So it is quite a, you know, good enough with using the 26 characters.  But that's one of the points.

But I completely understand the inclusion is important and Edmon pointed out very good point of the language justice.  That's we need to do.

For example, I am now saying something in English.  But I'd like to do that in Japanese.  And then you understand collectively, including my nuance.  Then I would feel very, very, you know, included in the situation.  I try very hard to use English for this kind of demonstration.

So my point is, with that, that language inclusion is really important.  And then not everyone can use the English.  But only use their own language.

For example, I was for ICANN board another time.  And need to include everyone.  Every user for languages, and then we need to be prepared for that.  And then employing that kind of system is just left to the operators and the IT vendors.  So we try to ask them to use that.  So that is actually the effort of the universal acceptance.

So my point is, the barriers should be for example, as we, as the PC system, the computer system includes the various languages, then the people who use the computer system can use their own language.  Such kind of environment, the computer system, and then for example, the IT systems to, you know, accept the e‑commerce environment and order.  It should be able to handle such multilingual input.

They are not created and developed by the house but most that kind of system provided, the platform is provided by very big vendors.  So we need such very big vendors for the, which is why they are used, employed the universal accepts.  Then everyone in the world can be included by using their own languages.

That's my.  Thank you.

>> SUSAN CHALMERS: Thank you, Akinori.

I have some reactions to that.  But I just want to cheque around room first to see if anybody else would like to respond to Akinori?

So it seems that the ultimate objective, of course, is being able to offer everybody the option.  And to build upon a new status of connecting and using the internet anyway their own language and their own script.  But just recognizing that some communities have already, kind of, adapted to the ASCII‑only system that's been built up over several years.

And there have been some, you know, patterns and business practises that have been built around the existing system.  With that said, I don't think it is certainly ‑‑ I don't think it is a counterargument to UA.  But it is just something that might be useful to note as part of the discussion.

So thank you very much for that, Akinori.

Our next question is, "how can we surmount the barriers that we have just discussed through technical and policy coordination?"  Ram, during his opening intervention had mentioned that it's ‑‑ and Edmon I believe you mentioned this as well.  It is not so much a technical problem because the work has been done.  The as coordination problem.  And it is a policy problem.  And it is a visibility problem I think for governments.

And so my next question goes to you E, Edmon.  How can we surmount barriers through technical and policy coordination?

Thanks.

>> EDMON CHUNG: Thank you.  And yes, I guess as both Ram and I mentioned, we are, you know, at a very low level technical part, which is the protocols and those kind of things.  We are beyond that.

But the technical implementation is still, you know, a big challenge.  And we've heard from different parties.  In fact, one of the things that is important is to convince the technical community that it is an important thing.  And just to rely on market forces, I believe is not enough.  Because I think the market has spoken a little bit in saying that we tried this.  We offered this.  But nobody is using it.

But why is nobody using it?  Because they can't use it smoothly.  So we're in a chicken and egg situation right now.  And that comes to the policy part.  The fundamental policies for enabling internationalized e‑mail addresses and domain names are there.  So there is the IDN tables.  There is the different language policies for registration of these unique identifiers.  Those are there.  But what is not there are policy interventions that would motivate suppliers to flesh out the implementation for universal acceptance, ensure that everything is as easy to use for a multilingual e‑mail address than not.

Just as a good example, many registries I actually don't know in China or Japan do allow, for example, Chinese and Japanese domain name registrations.  But what if the user uses a Chinese e‑mail address to register that domain?  Because your system support that?  Maybe, maybe not.  I would guess not today.

And those are the things that, you know, once you use it, you don't ‑‑ you know, can't.  So I think the technical part is trying to convince technical people that this is important.  And, you know, on the policy side, it is about, you know, having policy intervention to motivate suppliers to actually put in place.  And the ‑‑ I think the little bit touched on by Secretary earlier.  But there is also the International Decade of Indigenous Languages that is there.  The Sustainable Development Goals that touches on information and heritable and culture.  Those are the things we need to tell people.  That it is not just about well we can also use these cool ASCII domains.  Put when we talk about choice, the other side is important as well.

Yes, we can choose these cool names.  But what if I want to choose a local name?  Right now it is difficult.

So choice?  Yeah.  That is the market side.

The other thing, one last thing that I want to touch on is in response to, I'm guessing, the speakers from Chinese.  Because they are talking about input methods.  I input method in Chinese as well.  So yeah, maybe a few more keyboard strokes.  But ask any Chinese or, you know, ask by do, do they have problems searching in Chinese?  I don't think they search in English, right?  Even for just a few more key strokes, they would search in Chinese.

And you know so navigating should not be a problem.

And the other thing about I guess in terms of ICANN and Top Level Domains, speaker mentioned that just three key strokes can type a top level domain, why would you bother?  Well, that is why single character Chinese domain names and single character Chinese Top Level Domains is so important.  Because one single character in Chinese means a word.  And that can be done in two, three strokes.

So hopefully that adds.

>> SUSAN CHALMERS: Yes, absolutely it does.  Thank you.

And I want to see if we can turn to Ram.

>> RAM MOHAN: Thank you, Susan.

So as Edmon said, the core technical challenges, and the core technical contours of this problem are already solved.  How to make these identifiers work with each other is already a known thing.  That is not something that we ought to be focusing effort on.

In the real world this is really for an organisation of any size and scope.  This is a little bug, a little thing that has to be resolved.  The reason, in my opinion, why universal acceptance, multilingualism and digital inclusion of this kind has not yet become an automatic default is a because we need a bit more policy coordination.  We need a bit more in the ‑‑ on the side of incentives.

Because if you look at it, the problem that we're talking about is so small relative to large issues that organisations and technical firms have to deal with.  That they shrug their shoulders and say this is a little thing.  And nobody is complaining.  So as long as there are no complaints, it is all right.  We can just move forward.

But imagine a world where for every procurement that every government does.  Imagine a world where the procurement says that universal acceptance and vendors who have demonstrated that their systems have universal acceptance and are ready for universal acceptance.  Those vendors will get a preference.

And then say this.  In the world that we live in, those kinds of incentives, that can be driven by policy development, that can be driven by economic incentive creation, that don't really cost a whole lot for the organisations and the governments or procuring organisations that do the procurement.  But it really provides an impetus and a prompt to those who are competing for business to say, we need to prioritise universal acceptance.

And the resolution of universal acceptance are an important priority for this organisation.  Otherwise it just falls in the priority scheme.  That is one thing.

The other thing that we ought to consider is, should we continue to look at this and talk about this as domain names, as internationalized domain names, as internationalized e‑mail addresses?  Because I don't know about you, but for me I'm a technical person.  But even for me as a technical person, when I hear IDNs or EAI or acronyms like that, eyes glaze over.

Should there not be a simplification of what we are actually trying to solve for?

We're trying to solve for not just the ability to access parts of the internet, ability to be able to communicate in your own language.  But for also solving for accessibility.  Right?

So perhaps we ought to start thinking about changing the dialogue, the terms of the dialogue from internationalisation and from these technical terms that were invented 20, 25 years ago.  I was a part of it, so I'm guilty as one of the people who have perpetuated it.

But time to start thinking about access in own language.  So about digital acceptability and digit inclusion and universal acceptance, the international domain names, that entire area fits underneath that.  And I think it is time to elevate the dialogue and focus on policy‑based initiatives that can actually drive that can actually drive that kind of digital inclusion.

>> SUSAN CHALMERS: Seems like there is definitely a space for some marketing wizards as we move to promote this whole issue set.

Ram, I just ‑‑ I did want to note that and emphasise Ram's point about the role of governments and procurement policies.  I think this has been consistently would have the recommendations that has been made by the community that is focussed on universal acceptance.

Oh we have a hand.  Please.

>> AUDIENCE: Hello, thanks.  My name is Neev, Brock.  I think it is very interesting debate.  My question goes a bit, is this about acceptance of multilingualism?  Or is it also about the construction and co‑creation of multilingual internet?

So my point would be, are we aiming to make more languages available on platforms?  Or are we also looking on how can we build a more decentralized internet from bottom up, to so say.  And the experience I I want to share is working on network and vision to create proper infrastructure.  For instance, why not host a local server.  And local e‑mail server.

And that problem I think is twice‑fold.  I would say there is an intersectionality between language, but also big tech and alternative technologies.  And I agree it is like policy issue.  But if you would like run a local server, even before and with a local web hotsing.  Maybe even before to touch the language base, it would be just blocked because there is a protocol that is not updated.  Or it goes into spam because automatically for some reasons configurations of Gmail or whatever.

This is not shown as other e‑mails would be that come from, like, big provider.  And I think this is a very important intersection.  So my question would be, I agree, it is not only about the domain names.  It goes a lot deeper maybe of this level.  And yeah, well, question, if you see this like in a similar way?  Or how do you also think if we think of a diverse internet also in terms of technology and policies that should, yeah, provide for that.

Thank you.

>> SUSAN CHALMERS: Thank you.  I'd like to turn to one of our speakers, expert speakers to see if they might be able to respond to that question.

Or if anybody else.  Dawit has his hand up.  So Dawit.  And then Edmon.

>> DAWIT BEKELE: Thank you.  This is an excellent question.  Or comment.

I completely agree.  One of the frustrations of many communities is that there are less and less, you know, empowered to decide on what they can ‑‑ how to commute their languages.  Especially for example platforms might decide which languages should be, you know, allowed on their platforms.  And you have to wait for someone else to decide when your language will be available.

So I completely agree that we need to empower communities to decide what is important for the languages, what should be supported.  And not someone else.

Unfortunately this is easily said than done.  Very often, you know, this platforms and other tech companies are becoming more and more powerful and they decide on our lives.  So in a way, we had more possibilities to decide for ourselves before than we do today.

So I don't know how this will happen.  But I completely agree that it is important that, you know, the communities have a say on what is available online.

Thank you.

>> EDMON CHUNG: Yeah.  I think it is a great idea.  I don't know whether you intended it as an idea.  But bringing up the community networks, I think it is a really good model to learn from and work with.

Because that would be a group that is just building up the infrastructure for themselves.  And obviously, local language makes a lot of sense in that case as well.

But back to your question part.  Is it acceptance or co‑creation?  I think it is both.  It is co‑creation for a multilingual internet which is to work with local communities to build up a multilingual internet for them.  And domain names and e‑mail addresses being the fundamental parts of that multilingual internet.

And the acceptance part, as you really pointed out.  The acceptance part is the big tech and the other systems around the internet.  So that it remains one internet.  And for them to accept as well.

Yeah, of course you can build some complete separate infrastructure for multilingual spots of internet.  But we want one interoperable internet.  And that is the acceptance part.  So I think that is a very good and interesting way to think about it.  And definitely, is one of ‑‑ I don't know why I've never thought about it.  But the community networks group initiatives are really something we can learn from and work with.

>> SUSAN CHALMERS: Thank you so much.  Just in the interest of time, I'd like for us to move onto the last question.  Which is, "in what ways does internet multilingualism support the broader goal of digital inclusion?"

I feel like we have canvassed a lot of this already.  But I think it is a very crucial question that we should examine.  So I'd like to turn to Theresa Swinehart.

>> THERESA SWINEHART: Sure.  So I think we actually covered quite a few areas of what this entails.  And I really like the conversations about the local community networks and some of the challenges also that have been experienced at different points of using the internet or using multilingual content around that.

I think it also demonstrates no one entity can face this alone.  But the broader awareness of the dependencies amongst each other of the different systems, whether it is the underlying technology, the domain name system, internationalized domain names or the ability to access content and how that works.

So I think, you know, the reiterate the opportunity here around more and more awareness to this.  And so as we look at some ways to help support this, you know, there had been, last year we had the opportunity around the Universal Acceptance Day.

And really quite a remarkable level of participation by the community, by the global community around that.  And there will be another one this coming year.  But listening to this conversation, there is actually I think an opportunity to look at it.  Not just as a Universal Acceptance Day and awareness.  But actually awareness about all the different elements that contribute to it.

The local community networks.  The ability to have certain content online.  The partnerships with different entities and different organisations around that.

So that is certainly one area that we'll be looking at.  And I think we'll certainly help support the broader aspect about the inclusivity.  It is all very nice to talk about the inclusivity.  But there are different roles and responsibilities around it.  And those roles and responsibilities in many ways are driven either by economics or demand or passion or whatever it might be.

And I think the examples of procurement, government procurement options, we've seen that in IPD6 and other things.  To be successful.  And I think we can see it in other areas as well.

So in addition to the universal acceptance awareness building and really creating that awareness.  I think the other area ‑‑ and e touched upon this briefly.  We have the opening of the next round.  We're still developing some of the policies around that.  Some of the policies actually do go towards the internationalized domain name elements or applications that might be coming in in different forms around that.

So that work is still under way.  And the guidelines for the actual taking of applications is still under way.  And when it opens up, hopefully that affords another tool in the kitchen cabinet, if I could put it that way.  And another resource and opportunity to contribute towards the broader elements of a global internet.

And I think the terminology of acceptance and co‑creation in contributing towards that.  So I'll leave it with that.  But really this has been an invaluable conversation.

>> SUSAN CHALMERS: Thank you so much.

Len.

>> AUDIENCE: Very fascinating conversation.  And I might toss in a couple of other thoughts.  One is, and I ‑‑ I learned this I guess just coming to Japan and trying to figure out how the remote control works or how to get ATM, money out of the ATM.  Or even, you know, most other things in terms of navigating.

But one is, I think we should ‑‑ and I think it was mentioned earlier.  We shouldn't discount the role of iconography in this whole discussion.  I think there's been a lot of ‑‑ especially as an entry point to get either access to the internet or get started somewhere.  To me it is like, you know, driving a car.  I think about small things, like knowing which side of the car the gas tank is on.  Just little things like that.  I think that have been built up as convention over the years.

So I think convention and iconography can be a valuable tool I think here.

The other one is, is that as I watched ‑‑ especially since certainly the beginning of the pandemic, of the use of, you know, just not typing anything.  But just pointing your phone at something.  Or even speaking something that at least gets you through some sort of artificial intelligence that e says I recognize this language, I'm going to convert everything.

So I guess my main point is here, is, and I know it is the topic of the conversation.  But I think, you know, relying on the domain name system, if you will, and domain names to solve the context problem of language is probably not very scalable in the long run.  Because there is going to be so many other things that we're going to have to do with the internet that are going to rely on that context.  So I would make a case for saying, let's look at the efforts going on in identity management and some of those contexts to be able to help.  Because we'll have a lot of technology to throw at this.  We'll have a lot of stuff table to say within one or two phrases I know what you are language you are speaking no matter where you are in the world.  And then be able to adapt that.

And then the content and everything else gets kind of set.

Or I arrive even as an e‑mail address.  Maybe, you know, even the e‑mail address system relice more on identity management.  I gave a talk long time ago back when I worked for a telephone company that, you know, we should just forget about using phone numbers at all.  Because my daughter knew none of her friends' phone numbers.  You know, she relied on some other con context for that.  Through whatever platform she was using.

And remember, people are going to have more than one presence on the internet.  Based on social network or business network or something else like that.

So anyway it is just a thought.  But I would say that we should look for other ways to take care of this problem other than the domain name system.  And I'm not discounting making it a good working domain name system in the context of language and characters.  But still I think we shouldn't get overreliant on that as a solution.

Thank you.

>> AUDIENCE: Hi.  Pat cam with various line.  We talked this morning about preservation of cultures.  Preservation of languages and drifted into domain names and DNS.  And I think the example with put with the two coming together the last 20 years is dot cat.  For catalogue.  So.cat.  They did TLD and one policy.  If your cat linked to a website.  You had to have a catalon content on it.  Catalon content grew in that time frame.  26 characters on the alphabet.  They have the sedea on the keyboard so they can do other language.  But that is really the opportunity here.

When talk about creating a movement for universal awareness.  We're talk about a spending lot of money to solve a problem at the edge of the internet that is thwarted every day by the core of the internet and that is the ASCII problem that DNS handles that.

What I don't think we've done recently the taken a look at the DNS infrastructure itself in terms of resolution software.  What happens in recursive servers what happens in authoritative servers?  I do know years ago we did use a wild card at the authoritative server to interpret UTF8 and UTF16 so that we didn't have to do just ASCII.

It worked clunky.  But it was initial implementation and then a wild card prohibition put in place, for a lot of reasons.

But have we thought about solving it at the core so that we don't have these limitations that we have at the edge today?  Or should we?

>> SUSAN CHALMERS: Thank you.  We'll just, recognizing the time I'd like to make a few points to conclude.  And if any of our speakers would like to consider maybe a very brief intervention or a winding‑up point, please do.

But I would say this has been a tremendously vibrant and dynamic conversation.  I know we aren't many in this room, but that hasn't limited the kind of diversity of thoughts and perspectives that have been contributed to this conversation.

We had intended to focus on the collaboration between different institutions moving forward to promote universal acceptance.  But we've run out of time.

I would just say that in the upcoming council working group on internet at the ITU, there have been multiple submissions from different countries on this very subject.  On universal acceptance.

So those submissions will be discussed on October 18th.  So I think that this discussion was quite timely.  And I'll wrap it up there.  But I see Akinori, you would like to say something?  And I don't know if anybody else would like too too.  But let me just say thank you so much to our expert speakers for their time.  And thank you also to everybody who's joined us today.

But Akinori, over to you.

>> AKINORI MAEMURA: Thank you very much.  I'm really impressed with this discussion.  And I am thinking about it is quite multistakeholder process to make this inclusion.

And then, you know, in the part of the set up identifiers policy, I can still a job with the community.  Great variety of the community effort.  And then many, many language community have been involved in setting the rules.  And then ICANN still doing the policies for the new TGAD programme.  So that is quite big effort.

And as I said, it is really integral part is is the platform vendor to cover the market angle.  And IBM and universal acceptance ready and that is crucial part.  Because the common platform manufacturer, if they are the UA ready, then almost all the system UA ready.  Then that's very crucial part.

And then the public policy will encourage a lot of end user system move forward for the universal acceptance.  That's very important.  So this is quite everyone need to do their own job for universal inclusion which is really like multistakeholder approach of the internet.

Thank you.

>> SUSAN CHALMERS: Thank you.  And just one last important point.  So the organisers of the session pore over the transcript and contents and create a synthesis of the discussion and develop a report.

And you can probably anticipate that coming out in maybe December.  Because we want to take some great care with how we treat the content of the session.  So please look for that on the IGF website.

All right?  Thanks everybody and have a great day.

(Applause)