Universal acceptance – is the Internet reaching the people it needs to? – WS 06 2018

From EuroDIG Wiki
Jump to: navigation, search

6 June 2018 | 11:00-12:30 | MASTER ROOM | YouTube video
Consolidated programme 2018 overview

Session teaser

There are limitations to the character strings that will be accepted by some software. Is this limitation an obstacle to the universality of the Internet?

Keywords

  • Universal Acceptance
  • IDN
  • Script
  • Language
  • Inclusion
  • Search engines
  • EAI
  • gTLD
  • ccTLD

Session description

A significantly large number of people cannot use the Internet conveniently and effectively because of barriers associated with characters, writing systems, language. The goal of this session is to explore what the barriers are and how we can reduce them.

Format

Workshop format. A small panel of 3 to 5 people, representative of the different points of view, will introduce briefly relevant aspects, problems or experiences. Then we open the floor for questions from the audience to the panelists as well as contributions on other ideas. Depending on time, room will be provided for final remarks by each panelist.

Further reading

Some material referenced in the email discussion:

People

Please provide name and institution for all people you list here.

Focal Point

  • Roberto Gaetano, Public Interest Registry (PIR)

Organising Team (Org Team)

  • André Schappo, Universal Acceptance Steering Group (UASG)
  • हरीश चौधरी (Harish Chowdhary), National Internet Exchange of India (NIXI)
  • John Klensin, Internet Hall of Fame Inductee
  • Maarten Botterman, ICANN Board
  • მზია გოგილაშვილი (Mzia Gogilashvili), Georgian National Communications Commission
  • Наталья Филина (Natalia Filina), EURALO Individual Users Association
  • სოფო კიზირია (Sophie Kiziria), Caucasus Online LLC

Key Participants

  • Душан Стојичевић (Dušan Stojičević), Serbian National Internet Domain Registry (RNIDS)
  • Maarten Botterman, ICANN Board
  • Patrik Fältström, Security and Stability Advisory Committee (SSAC)

Moderator

  • Prof. რამაზ ქვათაძე (Ramaz Kvatadze), Executive Director, Georgian Research and educational Networking Association (GRENA)

Prof. Ramaz Kvatadze graduated from the Moscow State University. In 1982 he obtained Ph.D. in Physics and Mathematics at the Moscow State University and in 1996 Doctor of Science degree in Physics and Mathematics at the Tbilisi State University. He was working at the High Energy Physics Institute of Tbilisi State University at different positions starting from Research Scientist to Deputy Director responsible for research. R. Kvatadze conducted his research in high energy physics at several world leading scientific centers: Joint Institute for Nuclear Research, Dubna, Russia; Niels Bohr Institute, Copenhagen, Denmark; European Laboratory for Particle Physics, Geneva, Switzerland; University of Manchester, Manchester, UK; Institute of Nuclear Physics, Lyon, France. Since 1999 R. Kvatadze is Executive Director of the Georgian Research and Educational Networking Association GRENA. From 2012 he is Professor of the University of Georgia. R. Kvatadze participated in more than 50 research and development projects of the Ministry of Education and Science of Georgia, European Commission, NATO Science for Peace and Security Programme, International Science & Technology Center, etc. and is co-author of 116 publications.

Remote Moderator

The Remote Moderator is in charge of facilitating participation via digital channels such as WebEx and social medial (Twitter, facebook). Remote Moderators monitor and moderate the social media channels and the participants via WebEX and forward questions to the session moderator. Please contact the EuroDIG secretariat if you need help to find a Remote Moderator.

Reporter

  • Adriana Minovic

Current discussion, conference calls, schedules and minutes

The debate has been done entirely by email, using the WS6 mailing list. The team agreed on a title and scope that was matching the different points of view. The relevant discussion was whether we should limit the discussion to IDN or extend the scope to other Universal Acceptance issues. It has been agreed to take a wider approach to provide contributions also on items that are less often discussed. The final list includes the following:

  • gTLD IDN
  • ccTLD IDN
  • UA issues with ASCII TLD
  • Search engines
  • EAI - Email Address Internationalisation
  • Are Domain Names the main issue?

Messages

  • A significantly large number of people are unable to use the Internet conveniently and effectively because of the barriers associated with characters, writing systems, and language. Universal acceptance is not only a question of internationalised domain names (IDNs), it also includes new generic top-level domains (gTLDs) in Latin script and their use in search engines, e-mail address internationalisation, etc.
  • These barriers are affecting people’s ability to send and receive information, or to create audio, text or video. Therefore, the ultimate goal should be the possibility for people to create and change content in their local language and script, regardless of whether it is text or audio.
  • Usage of IDNs and related standards is low because there are insufficient financial resources available for registrars as well as registries. IDNs need to exist because they are a cultural thing and something each country needs, but they are not popular. This is supported by the fact that their usage is still very low in many countries.
  • The problem is in our habits and it is visible in the very example of using a keyboard with our local alphabet, which is quite rare to find nowadays. So here is the first problem, followed by problems with content, and finally IDNs.
  • The implementation of standards rather than the standards themselves is the issue. So what we need is to make sure that the standards are implemented.
  • We going around in a circle on this topic because there is not that much deployment in practice. If people used IDNs more, they could become more common among manufacturers and others.
  • If we want to keep the Internet open to everyone, we need to try to preserve its diversity, especially through mechanism such as IDNs

Find an independent report of the session from the Geneva Internet Platform Digital Watch Observatory at https://dig.watch/resources/universal-acceptance-%E2%80%93-internet-reaching-people-it-needs

Video record

https://youtu.be/DnqZKsij3G4

Transcript

Provided by: Caption First, Inc. P.O Box 3066. Monument, CO 80132, Phone: +001-877-825-5234, +001-719-481-9835, www.captionfirst.com


This text is based on live transcription. Communication Access Realtime Translation (CART), captioning, and/or live transcription are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings. This text is not to be distributed or used in any way that may violate copyright law.


>> Okay. Maybe we can start. Thank you all for coming. This is the session about Universal Acceptance. And the big question is: Is the internet reaching all the people it needs to reach and what are the issues and problems that prevent us from doing so? I would like to first do a quick presentation of the speakers. And then leave the floor to the Professor Ramaz who will be called in attending the contributions, speeches. I want Roberto Gaetano. We are involved in IDNs and we have some IDNs in my portfolio. That's my main interest. And other speakers are Patrick Foldstrom who will be joining us remotely among others. He is a member of the security and stability advisory committee. And then we are going to have Maarten Botterman, member of the ICANN board. And Dušan who is from the Serbian registry who also has experiences with IDNs and a lot of technical competence.

Then I think I'll leave the floor to Professor Ramaz Kvatadze who is the executive director of Georgian Search and Networking Association. And Gadinna will be the moderator for this session. Thank you.

>> Good morning. I think our meeting will be successful. I don't see too many participants which is -- well, a little bit strange because we are going to discuss today very interesting and very, very useful things, at least I hope so.

So it's my pleasure to present first speaker Robert Gaetano. He introduced himself. So please, Roberto, the floor is yours.

>> Roberto Gaetano: Yeah. I am not going to really have a speech, but I would like to start this session by making a short of summary of the topics that we have been discussing on the mailing list in this last couple of months that will give also ideas about where the debate is and maybe solicit some questions from the audience after the speakers have spoken.

So we have touched several issues. We have started by saying that Universal Acceptance is not just IDNs. Although, we have problems in Universal Acceptance of TLDs in non-ASCII Scripts, but we have the same problem also in all the well known problem. We had this even after the first round of new TLDs when we had TLDs that were simply longer than the three characters that were considered wrongly so by some developers and some implementers as the standard size of a TLD. We have touched on the fact that Universal Acceptance has an impact not only on domain names but also on search engines as more and more people will use to search engines to get to their preferred domain. We have -- we have also put the accent on the issues that are related to e-mail and e-mail address internationalization. And then towards the end, we had the sort of over arching question that will be addressed by our first speaker which is the question that on whether domain names is really the main issue or whether we have other things that we are failing to consider that are equally important. That's it from my part. And so I would like to go back to the moderator.

>> Moderator: Any questions please? You are welcome.

>> (speaking)

>> Moderator: Okay. So are we ready? So then next speaker is Patrik. Please, you're welcome.

>> PATRIK: Do you hear me?

>> Hi, Patrik.

>> PATRIK: I am trying to share my screen. I don't see that works -- you seem to see my video. There. There's my screen. Okay. Cool.

So anyways, what I will do is give an overall description of where I think we are where, of course, internationalized domain names fit. I think it's important to try to understand and take a step back and understand what the actual problem is that we're trying to solve. We -- we have DNS services for the time and frequency in Sweden. We do tons of things. We also run internet exchange point and stuff. But what we have -- what you have been doing the last two years, myself and John and Roberto, many people, other people in the room, we have been trying to understand what the actual problem is. If we take a step back and think about also the goal of EuroDIG, we talk about bringing more people online. To me, bringing people online is bringing the people the ability to communicate F. I break this down, I find various issues regarding communication it has to do with the ability to create, text audio and picture which means both creating and receiving such content and also in receiving it, they would be able to present it in a way. It is digesting it and understanding what is happening. And when you have this information that you have been able to create, you also must be able to send it somewhere. And to be able to send it somewhere, you will be able to search for the recipient address. After finding whatever you would like to do, you need to look up the address, sometimes in your address book as in a local cache, you need to be able to use the address and when you receive information that is internationalized, you need to also be able to reply. And the question is: Where do we have the nationalization use in all of these steps? Creation of content, digest content, look up search for things, look up things, use addresses and reply. So there had been a couple of issues mentioned already. For example, internationalized domain names where we have issues and we diversed policy and we have contract or compliance issues. We have non-contracted parties where we cannot use the non-contracted compliance. We have (inaudible) that live under different umbrellas. We are both registries and registrars where they try to do the right thing when interacting with multiple registries. You have a diverse policy which makes the life of the registrars not easy. We have the e mail address internationalization. We have two manufacturers of e-mail clients, Microsoft and Apple and to be blunt, if they just implemented whatever we wanted, we would be done. Regarding the web itself, we have a search bar with a content window where they content and display today. Those things are managed in W3C with the help of various decisions and agreements there. And, of course, software developers in a few web browsers that we have today.

We, of course, have the actual implementations in JavaScript by programmers that make mistakes saying that the number of characters is only three and not more than that. We also have other applications, of course. The 80 to save content. And the ability to have file names in non-ASCII which might be problematic if you use different file systems. Apple had their normalization. Store and manage the information itself, which, of course, the librarian community I'm looking at for quite some time, but all of those applications or most of them are using related to IDM. Two libraries the lib curl and the question is: How much emphasizes and work has been done in those communities regarding proper IDMing management.

So from my perspective, what really should have impact is to make sure that people have the 80 to enter information in their own Language and this requires good Script and character support in all applications to be able to pass the information on and receive and interpret it by others. I think it is also important to be able to help participation by having the ability for people to translate and transform the information with the help of text to speech or speech to text and have that to be able to be used in multiple Languages. This would also help people with disabilities because many of the tools that disabled people are using require functionality in these areas. We have user interfaces where you need to be able to have correct interaction, which is quite complicated in certain Scripts. If we just look at character interaction, character based interaction, we have global devices automobile over the world for the next billion users which might have memory in old devices that must be able to support new Scripts and must be able to support text to speech and text to speech and translation of audio. We -- when I talk about text to speech, I also talk about automatic sub-text mechanism in, for example, the application YouTube and what others use to participate. Regarding uni code and if we limit ourselves to text, we have transformed text to pictures which have to do with some other normalization issues. We have problems select copy and paste text which is not so easy by direction environments. It is also where we use text for example in the search bar in browsers. We also have problems to be able to enable people to search, which, of course, includes a calculation of matching algorithms. So they might be known or unknown and there might be various negotiation of context before the matching algorithm is chosen. It is used there. And this goes all the way down to individual character is used, but for the Uman, it is much more interesting to do searches.

One of the subs I am working on at the moment is just look at very, very limited scope to be able to search using text with instructed data and global context. This requires all different kind of things from schema management if you -- if many people in the room might know about the doubling core initiative in the 1990s and doing that in the global context is kind of complicated. But just agree on attributes. This touches translation, but do always include as it seems negotiation of content because all boundaries and when I talk about contextual boundaries, I don't always talk about changes in the Script and Language environment, but also norms and also other kind of context. If you talk about physics or astronomy or you have different content where matching algorithms are different.

We also see an increase of environments where it is easier to implement searches within a closed environment within Facebook, but more and more on the communication in the world is within these closed silos. That is a good thing for the end user, but it might be the case we get increased divergence that with end users moving between the silos, they will get surprised because the actual content and how to handle context in a negotiation varies between the silos. So the over arching question is: Where in all of this is IDM and electronic e-mail address and where is uni code there? They are there as pieces of the puzzle, for example, in internationalized domain names, but I think it is really, really important to remember the over arching is here which has to do with the ability for people to create content and change content in their local Languages and Script regardless of whether it is text or audio. When we have this discussion, just encourage everyone when we try to say oh, we need to solve the IDM problem. We need to not forget the big picture and see what the ultimate goal is. Thank you very much.

>> Moderator: Thank you very much for interesting presentation. I would like to ask for questions now and please present yourself before asking questions. Thank you.

>> Hello. Good morning. I'm sorry I came a little bit late. My name is Tody. I am also a youth participant. My question is: How can you prove internet access internet to the underprivileged refugees within Europe? I know for a fact that refugees don't usually have access to internet and because it takes a long time for them to integrate in society, it takes about 3 to 6 years. So they have gone a long time without having internet. How do you improve access to internet in regards to underprivileged refugees? Thank you.

>> Moderator: Thank you.

>> Patrik, did you get the question?

>> PATRIK: Yes. But let me reflect on. This regarding refugees, a lot of refugees that are moving across country boundaries and culture boundaries are moving to an area where computers and others are not installed with what every Script and Languages they're using. This is one of the globalization issues that I was talking about where I think it's really important that every computer that is -- and every distribution ever operating system, every browser need to have support for all the Scripts and Languages in the world so that people can both display, listen, talk and otherwise use whatever Language they are using. So I think this has not only to do with refugees. It is actually for all of us because all of us humans are moving all over the planet and all over the place and to believe that we are staying within the cultural norms and Languages and are the context where we have been staying, that is long gone. We are moving all over the planet and we need to be able to support that. So I think the important thing there is to be able to -- as I said, to insure that distribution of operating systems and software is not culture dependent in two different countries. We only get Swedish versions. We need all support for all Languages all over the place. Thank you.

>> Moderator: Thank you very much. Other questions? Please.

>> Hello. My name is Katie. And my question is during your presentation, you presented slide where we have resources and there was very briefly presented resources. Can you say for Europe what is more critical issues and how you think it can be handled? Thank you.

>> PATRIK: Thank you very much. Let me Connect to the previous question they got. I think the most crucial thing is that when Europe need to understand that there are many individuals, many humans within Europe that use Scripts that are much more complicated than what we are using in our official Languages. In the official Languages in Europe, we use Greeks, Sulfuric and Latin Script, but there are tons and tons of people living in Europe and visiting Europe that are also using Arabic Script for various Languages. They are also using Chinese and such things. So we need to be much better within Europe to support and to understand other kind of Scripts, for example, that we are using. Thank you.

>> Moderator: Thank you.

>> First of all, thank you very much for the presentation. And also reading the critical issues and maybe connected in terms of resources. What do think makes these things sustainable? Really often when we talk about inclusion and making the internet more accessible and putting in place different resources that can be put there, really is the issues and ability that it seems really expensive and not really achievable. Do you think it's really as such or it's something that can be actually achieved but there is not really dangerous to do so because of mainly reason that can be political or others?

>> PATRIK: A lot of the discussion as (inaudible) can tell you about the discussions in preparation for this meeting, I think there are two mistakes that we have made. We work with translation. The first one is we are focused too much about internationalized domain names and we have allowed people to not follow the standard. We need to be much, much more strict as a protocol police and hit down very, very hard on people that do not follow the standard as defined by the IETF. And also, of course, the follow-on standards regarding input and content creations in that is done very often in W3C. The section thing we have not been doing apart from the standards compliance is to work with generic issues regarding Script and text input and output. So what I'm saying is that we have been talking about the addressing and internationalization of domain names, but we have forgotten too much to talk about that. As many of you know, we have talked about internationalized domain names for years and years and maybe I have made the wrong thing. We should talk more about content. The ability for people in sweeten to be able to enter, read, and type Arabic text in one of the Languages. We don't talk enough about that. So standard compliance and normal information managements. Thank you.

>> Moderator: Thank you. Do we have no questions? I don't see. Thank you very much again for interesting presentation, Patrik.

So our next speaker is Dušan, right? Dušan, please. The floor is yours.

>> DUŠAN STOJICEVIC: Thank you. I am coming from Serbia and I'm working for registrar international registrar. I like to highlight something for new faces here regarding IDNs.

So when we speak about IDN domain names, we are speaking about ASCII. ASCIIs standard allotting Script with some of Script, we present every other Script. This is the way how IDNs work. So apparently everything in IDN world is about a visualization. So why am I speaking about this? More technical people will say all IDNs are (inaudible) something. It's represented to the user via browser in their Script which is let's say (inaudible) Arabic, whatever Script. It is only represented to the user. It is not that Arabic script or any other Script will go through the network. This is about IDN domain names and this is the first thing that you need to know. Second thing and where I have some doubts and I shared my doubts even with Patrik is EI or e-mail address internet serialization. What is the problem there? There you have on the left side on the right side of the e-mail, usual e-mail address, you have ASCII, so-called puny code. On the right side, you have unit code. By mixing those two types of -- we probably made some mistake, but we have the standard and I can prove you that it works. So it can go through the servers, through the network and such e-mail addresses. I have one because I was at ICANN on a (inaudible) places regarding IDNs. For example, both (inaudible) generation panel and I'm using Cyrillic in my country. And I have let's say one e-mail address which is Dušan@(inaudible) all in Cyrillic. So whatever is below in the standard, you will see these and you can send me e-mail on this. I will respond to you and it works. Patrik is right on one thing. Mostly adoption of these standards are problematic. So we have for now -- I was testing this solution and I can say that we're doing fine with IDNs. Secondly, when we consider Apple with default settings, there is not much support in this RFC, which was done by ITF and Patrik also.

So, when we are talking about IDNs, we need to know those basic things. Second -- what we need to know about IDNs is how many participants do we have here? It's not that much. Whenever we speak about IDNs, we have low number of participants. What it means? It's the same in business. In Serbian registry, where I was participating at the time of establishing the Cyrillic domain name in Serbia being coordinator or member of various working groups for this TLD, we hoped that the number of the domain names that people sell would be huge. In fact, when you see around the world, maybe two or three IDNs are respectful with the figures of the domain names. Everything else is statistical error to the number of -- to the number of existing mostly existing ASCII domain name in your country.

So the usage of IDNs and all other standards is low because there is no money for us as registrars, for us as registries. There is no money in that and I'm considering IDN as operating music. It's probably a younger generation doesn't hear it. It's not popular, but it needs to exist because it's a cultural thing and it's a thing that you need to have in your country. You need to have ability to use your own Script. When you see -- when you put the thing differently around the world, you can exclude up to six countries. USA, Australia, New Zealand, maybe a couple of them more who are using only Latin Script. Everything else including Sweden, including Europe, France, Italy, et cetera, they are using IDN Script, something which is in uni code, something which is not ASCII. Latin world is most exciting for me because they are cutting their Script to use ASCII Script, to use Latin Script. So for example, Italians are cutting or French are cutting accent graph or circumflex. They're cutting and they're adopting their Language to Latin Script. Most importantly, by the techniques let's say mobile phones, if you want to send short a message SMS, if you want to send an SMS, then if you are using your own Script, you can use less and maybe in some countries it will be bigger prize for this SMS if you are using at least you can use half of the characters that you can use in ASCII. So in all countries, they developed let's say some variant of Latin version of their Language. So Georgian, we are here. They have very nice, very old Script, but apparently when they are lazy, when they want to reply fast on Facebook or on any other place, they are using Latin. Pure Latin, pure ASCII. So this is about the content. This is about the usage. Technology is not giving equal opportunity. For me, for example, I'm asking Georgian guys, do you have keyboard on Georgian Script? Let's say, can you go to the shop to buy laptop with Georgian Script? No. You cannot. Probably you can buy Latin or maybe in some places, some Cyrillic and this is Russian version of Cyrillic. Chinese, they have some keyboards, but the problem starts here not with the content. The problem starts with my habits, my habits to use the Script. If you are cutting the keyboard on the laptop to use my Script, all I need to go to shift to my code page and to use it, this is not fast. So give me keyboard Cyrillic keyboard, Chinese keyboard. This is -- I need to. So this is one of the -- one of the problem which comes before the content. And then on the content, we have IDN domain names.

So I can agree with Patrik in some ways and I can agree with him that we need to be standard police to make people use the standards but the problem is much broader than the standard. I will compare IDNs and also all Universal Acceptance problems like new Gs coming with four or more characters. All those problems like idealistics adoption. You need to change software on your e-mail server to get UTF8 compliance. You need to change the software on your laptop, et cetera, et cetera. So this is about adoption. This is not about whether IDNs are good, RFCs or standards are good or the usage is there any usage. Right now as I said, there is no money in IDNs. That's why adoption is slow. That's why we are too few listening this story.

>> Moderator: Thank you very much, Dušan. Questions, please. Please, Robert.

>> Roberto Gaetano: While you were talking, we had a consideration that we need money and we don't need to be lazy was letting me think about other problems that we have that is, for instance, about security where we should implement DNS and that has been a slow implementation just because of this. We have several other cases and migration from IPV4 to IPV6 and so on. So there is an over arching question about where to put money on the strategic objectives and so the question is: Is Universal Acceptance the strategic object that we need to pursue and so a request that we have more investments on this. The second comment was earlier. And I would like to bring this to people's attention because maybe we don't always get the overall complexity of the issue. When you were talking about the e-mail, you said oh, this on the right side of the e-mail address, we have a code and then on the left we have a uni code. While you were talking, I was thinking. If the e-mail is in Arabic, your left side is the right side. And the right side is the left side. So if you have something with a mixed page with a mixed Script, you have an Arabic page and then all of a sudden, there's your e-mail address and accepted you dent put e-mail addresses in work pages. Then all of a sudden, the context changes and I was wondering if either you or PATRIK could spend a couple of words to introduce and at least touch on this higher complexity about having different Scripts and maybe even going in different directions in the same context.

>> I will explain this comment like this. I'm not English speaker. So I can explain to you in Serbian. I cannot explain on English. Joking, of course, but this is about IDN. I was using left and right just to refer that we are here. If I was talking the same thing in Arabic world, let's say middle east DNS forum, I would use something differently to explain. Left and right side, but it's not the problem about the standard. The standard is there. Okay. I can complain and I can exchange with PATRIK my concerns about using mixing Scripts, but standard is there. It's working. Test it. By me, by a lot of people, it's working. It's there from 2013. It's let's say 5 years. IDNs are there for 10 years or so. Everything is not about -- it's not about the standard itself. It's about the adoption of standards. We don't have adoption by April, by let's say -- let me give you good example. I'm booking the hotel in Georgian. So can I use Dušan Stojicevic in Cyrillic in the form for booking? No. It says it's not valid e-mail address. Sorry, guys. It's valid. It's according to the standards. This is adoption. I need to call this guy to explain. Please validation of e-mail is not correct and I need to do that at each hotel, at each form on the internet. This means adoption. And if you don't have in five years adoption that I can book a hotel here in Georgian or anywhere in the world, then people will say this standard sorry. It's something. We will stick to ASCII. So use ASCII, please.

>> Moderator: Thank you. More questions?

>> Hello. My name is David. I work for the Georgian research. We asked for the IDN, which we have since 2016. And I think the problem, the most problem is that IDNs are not very much used by the people. I don't know the case with Serbia, but in Georgian, I can say that, for example, the registrations of IDNs are let's say 2, 3% comparing national domain. It is very low number. And the reason why people are registering Georgian IDN, is just addition of the G domain. It's not translation. The register IDNs to have Georgian domain name, which is similar to which they have (inaudible). And the question they have no question in technical aspects. They want to register similar domain name, not use it. Maybe they're directing when you are typing in Latin to be shown in the local Georgian Language. It's not common issue. So people are not using it. Despite our big efforts to make it popular, still the registrations are very low. What's the case, for example, for Serbia? Is it popular?

>> DUŠAN STOJICEVIC: As I said, this looks like statistical error when you are considering the number of domain names in your second ASCII domain names. First, when did you -- from where there is (inaudible), what is the year that you started dot G on ASCII?

>> (speaking) (low voice).

>> DUŠAN STOJICEVIC: And 1992 -- from 1992 till now, you are using dot G in ASCII. And you're cutting your Script to ASCII and you are using only ASCII for this. So it became like a postal address like this, some sort of postal address. So right now, how many years? 18 years or something you are opening or more. You're opening on your Script and you expect that it's going to be popular. No? It will not be the case. It's not the case everywhere in the world as I said except ration and some Chinese IDNs. It's similar. It's a statistical error because they came late. People are all over the adopted ASCII. Also, the problems that are to be PATRIK and me we were mentioning during our speeches like keyboards, like content, et cetera, et cetera. So same thing in Serbia if you are asking me statistical error.

>> My name is Alex and I work for center. Thank you so much, Dušan and Patrik for the great presentation. I want to mention this is a topic very relevant for country code, TLDs. I have a bit of information sheet I thought might be interesting with this group as well about which TLD has adopted IDNs at second level and which haven't. The only thing I would like to highlight is I find it interesting that some TLDs have adopted IDNs as second level, but only for the top level IDN, but not for the ASCII top level. So it's a bit confusing and it's work in progress. So I think that's really something that's interesting to see that even at a country code level, there is still work on this. I'll just leave this here.

>> Moderator: Thank you. I think Patrik had some comment, right?

>> PATRIK: Yes. That is correct. (inaudible) (low voice) so my point is exactly what she just said. The main ability for people to actually use their own characters and Script and whatever they're using, for example, in messages and in messages and in checking what is the local messages is by far the most important thing. For the actual standards, they also agree with you and my point was there that we don't have to work on the standards. We have to make sure they're adopted and used. If that's the case they're not used, then maybe they're not needed. For the EAI, the question there, of course, is if any e-mail address really is needed and if it creates, if it really helps or it creates more problems. For myself for example, that use of Swedish characters which are very, very close to Latin, for me it is easy to say that, well, my name I have always had my name as a comment and then the e mail address itself is ASCII only. Maybe that is good enough for people in the world. I don't know. Today the younger people they don't know what e-mail is anyways. They are using Facebook or Snapchat or whatever. Maybe it is important for them to use their own namings and own local Script as identifiers in those applications. I have no idea. But this is once again why I wanted us to take a step back and exchange information with content within each of the various applications that we use because the standards as which I agree with Dušan, the standards are already that we need to make sure they're implemented and adopted. Thank you.

>> Moderator: Thank you.

Now I would like to ask Maarten.

>> Thank you. It's been wonderful to hear. I think people are deeply involved with the two most important stream of making this happen. The universal access steering group. And Asac in which Patrik is very important. It is in recognizing the importance of linguistic diversity and recognizing that we need to prepare for that future for the first steps taken. ICANN has such a limited role in that. It is really on making the DNS work except IDNs and in a secure and stable way, which is not a (inaudible). If you realize that some letters in different Scripts and (inaudible) in another Script and one of the dangers of the internet is that it's been used by malicious people to confuse people and lure them into activities they don't want to be a part of. That is a very important aspect to do. So on top level domain, on the ability to reach others in Scripts, it's crucial you also take that aspect into account. In addition, in some scripts, there's different expressions of that Script like in Chinese. So again, you see the complexity, which is not a problem locally if you use it because then people use the same characters in their mill and whatever. But in their e-mail, it becomes bigger problem.

So from that perspective, I can assure you that also there I can provide support, but it's not leading the activities, but they help make the Universal Acceptance to the group acceptance and they felt like it is part of ICANN, but also the board is very much appreciated and listens very carefully to the input that is received from that. So the latest state of affairs is 151 audience have been delegated. So in principle, they're reachable. As do (inaudible), it doesn't mean there's a high usage of them. 95 of them are gTLDs. And I feel and we feel that this is an important first step. I think you mentioned that parallel to IP version 6 in some way its is and IP6 is a necessity and IP version 4 is just too small. In IDN though, that's part of to market that there will never be a market unless we offer IDNs to make it possible. And that relates to facilitating communities. And they're a point that was yesterday also coming forward in several sessions. Communities are importantly driving our future, also our future use of the internet. And in that, you see that for some people, the IDN will probably bury their prime community that is correspondence of people for the regions using electronic means with the same Script and really are (inaudible) in how using many other Scripts are very happy to use that local Script. In India, we see some regions where local trade together provide sometimes products for global trait. But the communication in that region will be very much facilitated by that, of course the mobile (inaudible) needs to support it too. So that's one thing. The other does come together with the IP version. It's been involved in the readiness for IP version 6. We have been able to in researching that even 10 years ago. We sold the readiness, the preparedness to be able to deal with IP version 6. From below, well below 50% and 20% and now most ISPs can deal with it and this exceptional for those who can't. At the same time, the use state away for a long time and it is now becoming real. So IP version 6 becomes real now. I think you will see the same. First we need to be really ready before the uptake will grow because unlike the earlier adapters, the techies like -- I'm looking at this about others who are trying and doing everything they can to make it work. Most of us won't. Most of us will just say it doesn't work. Don't use it or I'll use the English. So the other thing we saw in IP version 6 there is those who were the early users unless there were people like PATRIK or others, they stopped using it in some cases. They were using it and they stopped using it because nobody else did. So I think we'll see those disappointments as well in here. I do believe strongly and that's my personal opinion, but it also has the ICANN board support that investing in making this possible is something we need to continue to do on the long stretch. We need to be aware that it may make to take more time. But the way to make it happen and to make the internet truly usable is by continuing to support this and continuing activities like the S.

>> Moderator: You have very much. Questions, please?

>> My name is Al. I come in Serbia and I'm a site member. I just think that we consider this issue in very short period of time. When I was (inaudible), it was 1987. I use PC the XT machine for typing my graduate exam. Because I could use (inaudible) and somebody remembered, I could use our Latin Script in a Serbian (inaudible) and doing some graphic squares and routes and indexes and everything to use PC in that time, I would use electronic type machine. So if we could use Latin Script 30 years ago MS Dos time, it is a time when we use all the script in the internet world nowadays. So we must be patient and we must continue to work very hard on that. Second thing, I can say example for Serbian, we have a (inaudible) in Latin national domain name, but we don't have a Latin IDN in our domain name. So I register Swedish domain name with the IDN with our (inaudible). So I have a domain name (inaudible) in that. So registries and ICANN is responsible to enable us to use all letters and what is important. This is when we use all software with English. It was a problem for many accounting people to use applications. They prefer to use a notebook and pencils and everything and then when we change the (inaudible) with local Language, they accept it and now every company uses a PC like standard tool. So definitely we want to reach 1 billion more internet user. We must adopt everything to do anything that we need to enable to use their own Language, their own lectures and their own tools and they will use PCs and everything will get modernized and their likes and they will have them use by technology and everything. That's my (inaudible).

>> Moderator: Thank you.

>> After one IDN session somewhere in eastern part of Europe, one I will not say the name but one guy approached me and asked me. I can solve IDN problem very easy. All the money that ICANN is putting in USG, whatever let's say group that are dealing with IDNs, switch those to Russia and Chinese to learn English and we will solve the IDN. So this is a joke, of course.

>> I disagree.

>> DUŠAN STOJICEVIC: Yes. I disagree with that, but when -- he said about ICANN putting a lot of money in USG and are the groups, I suddenly remember that there is very simple solution for IDNs to raise IDNs, of course.

>> One thing more. I'm sorry. I think that people (inaudible) that even all the Latin lectors used in the world not available in the internet. So there is not a working group that worked on Latin IDN to enable to use in the internet domain names. So obviously it is a lot of work to do and a lot of effort to enable the last people, the last person to use internet in the world. It is a question only on time.

>> Moderator: Thank you. Please.

>> Adam Peek. You mentioned keyboards and the unavailability of them. That really goes to PATRIK's point about the ability to create. That's really strange. I lived in Japan some time ago. In 1990s, my colleagues had keyboards and getting online in the early '80s. MSDOS allowed them to be displayed and had drivers Princed. It's not real a new issue, but keyboards are not as widely used. In you go to a Japanese store, there are hundreds of keyboards and entering the text isn't hard. Is a phonetic Language. You try KI on my laptop and RE and up comes the cangue. So they type as quickly as any western typist. But the keyboard issue is an interesting one. So keyboards are not -- no. The other thing is emoticons have that emoji. They're the original IDN. People used to develop characters on radio pages. It was very difficult to type Japanese characters into a radio page. You develop weird looking things that are now developed into emoticons because that was an easy way to create different types of smiles. You have hundreds of different types of smiles. So late 1980s, 1990s, people were using early versions of emoticons and it's in early IDN I suppose.

>> DUŠAN STOJICEVIC: We have here Armenians are using very old and beautiful Script. Do you have to buy in shops laptop with Armenian Script on the keyboard?

>> No.

>> DUŠAN STOJICEVIC: Same thing --

>> Stickers.

>> DUŠAN STOJICEVIC: Yes.

>> (inaudible) (low voice).

>> DUŠAN STOJICEVIC: Even these stickers that is shown is putting a lot of effort to change. For ordinary people, I don't mean on me or anybody in the room. I mean for some people who are not speaking English or didn't use Latin in their lives, it's hard to buy a laptop on a Latin Script than to learn how to type on domestic Script. This is --

>> hi. My name is John Pedro. I will try to put a young perspective on the table. I heard some type machines and things like that. But limiting the conversation to laptops where in the internet nowadays, we're using our mobile devices and especially with digital touch screens. I think there's an opportunity here to change the discussion from the hardware itself to the software because we can find the way of including digital keyboards that you can actually touch, there will -- that would be such a discussion right now because it is up to the programmers and to the demands to provide with the solutions necessary to confide those.

Another thing about a comment from the remote participant about a young usage and the behavior itself. Nowadays, the world is --d communication is world wide. We have friendships around countries. So we try as much to keep the communication as simple. So, I mean, it is relatively simple to understand why you were approached with such a solution to everybody learning English. I'm not saying it is right or wrong. It's like you said. There should be an alternative way of doing it. My opinion and picking up on your parallel between IPV4 and IPV6, again, both protocols are working right now. So I would -- I would propose that as those protocols, they should work in parallel as well. It's not -- it's not as if the transition should be from one day to another. It should be as integrative as the protocol as IPV4 and IPV6 were. Thank you.

>> Moderator: Thank you. Yeah. Please.

>> Just realizing we're approaching the end of the session. I would love to very much recognize the excellent presentations given so far and also for your leadership. We wouldn't want to leave this presentation without recognizing Roberto Gaetano and bringing this together and being open towards different stream of thoughts that originally was (inaudible) and allowing this to make it happen and also for celebrating his birthday today. So happy birthday, Roberto.

[APPLAUSE]

>> Roberto Gaetano: Yeah. It's a wonderful way to celebrate my birthday. About changing, I would like to respond to this. I think we have sort of a cultural issue. I think that IDN the reason I have always been open to IDNs when I was on the ICANN board for pushing the fast track and so on, is because we need to be open. We need to be open to new things coming and to enlarging the offer of the things that we provide to reach more people. If there is a sort of decepting view that comes, I think it is absolutely important and it is important for whoever coordinates the sessions to be open and to give space to the discepting view. Otherwise those meetings will be completely useless because it will be the same old people who are talking about themselves. But there was one thing that -- one comment that was prompted to me by hearing the different comments. I'm sort of wondering whether we are facing a chicken and egg kind of issue. So there is not much -- not enough that is done in order to develop new things, new keyboards and so on. On the other hand, there is not enough adoption of alternative ways like the local Scripts because -- because -- maybe because we don't have a full deployment of the tools. So this is what I -- I'm an old-timer. So I will use all the paradigms. One that I was using 30 years ago when I was in the Tellco, was the fax problem. Fax is useful if you have enough people that have faxes. Otherwise we have this nice thing that serves no practical purpose. So again, we need to do something in order to incentivize the market to use these tools and the producers, the manufacturers to produce these tools. So there is an effort that we have to have. The answer that I have for the comment that you had is simple solutions are not always the best. It might have been simpler, but then the next step is: Why don't all the people in the world speak one Language? We have seen the experiment still something that comes from way back in the past. It didn't work out. We have -- we have wonderful diversity in this world. We need to go against the push that we have in this globalized society that pushes towards uniformity and the reduction of the communication richness. And we have to invert this tendency and the internet is giving us those tools and this possibility. So the internet if we are really serious about the internet is for everyone, then we need to endorse and push the diversity that will come with Universal Acceptance with IDNs with different Languages, different Scripts and so on including accepting dissenting opinions in the workshop. So thank you. Sorry for this long speech. Non-intended. Since I'm now having my birthday, I felt that I could have this extra minute. Thank you.

>> There are some main points that were raised and, of course, in the end you can agree or disagree and express your opinion.

As the main topic is introduced by the fact that there is a significant large number of people cannot use internet with the barriers and associated with characters and writing system Language, the barriers go beyond the systems. And that is including TDLs, search engines, e mail address and authorization, et cetera. Hence, Universal Acceptance is not only the question of IDNs. This barriers that are affecting people's ability to create audio text and video, therefore, the ultimate goal should be the ability for people to create content and change the content in their local Languages and Scripts regardless of whether it is text or audio. Usage and all other standards is low because there is no money for registrars or registries. Hence, it is something that need to exist because it is a sculpture thing and it is the thing you need have your country, but it is not popular. This is supported by the fact that their usage is very low in many countries. The problem is in our habits and that is visible in the very example of using keyboard in our local (inaudible), but it is quite rare to find nowadays. So here is the first problem to start and then follow by the problems about the content and finally about the IDNs. Standards are not the main issue, but rather the implementation of standards. We need to make sure the standards are implemented really in practice. Also, are we going around with this topic because there is not that much deployment in the practice? If people are using these tools, then they would maybe become more common of manufacturers and others. If you want to keep the internet open to everyone, we need to preserve the diversity on it especially through the mechanism of IDNs. That is pretty much it.

[APPLAUSE]

>> Moderator: Finally, I would like to thank first of all the speakers for interesting view. And for all participants for active participation. I think this was rather interesting and useful. Finally, I would like to thank technical (inaudible) who organized this meeting for us. Thank you very much. Thank you.

[APPLAUSE]


This text is based on live transcription. Communication Access Realtime Translation (CART), captioning, and/or live transcription are provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings. This text is not to be distributed or used in any way that may violate copyright law.