Transcript: Opening Ceremony


Due to the difficulties capturing a live speaker's words, it is possible this transcript may contain errors and mistranslations. APNIC accepts no liability for any event or action resulting from the transcripts.

Monday, 21 February 2011 at 14:00

Speaker: Ladies and gentlemen, a very warm welcome to the first ever joint event of APRICOT and APAN held right here in Hong Kong.

Before we begin, let's watch a short video to show what Hong Kong people think are the top 10 defining moments and technical milestones of the internet's development in Hong Kong over the last 20 years.


Desiree Ho: Good afternoon, ladies and gentlemen, and a very warm welcome, once again, to the first ever joint event of APRICOT and APAN 2011.

I am Desiree, one of the NetMission ambassadors.

Lily Kung: And I am Lily, also one of the NetMission ambassadors.

We are very excited to welcome you to the biggest internet event held here in Asia. It is the second APRICOT in Hong Kong after 14 years and APAN's first in Hong Kong.

APRICOT -- short for the Asia Pacific Regional Internet Conference on Operational Technologies -- is a prestigious annual regional ICT event that aims to provide a unique educational forum for internet builders to learn from peers and leaders and the internet community from all around the world.

Desiree Ho: Yes, and APAN -- the Asia Pacific Advanced Network -- on the other hand, provides an advanced networking environment for the research and education community in the Asia Pacific region and also promotes global collaboration.

Lily Kung: Yes. In fact, did you know that just in this event today, we have over a thousand participants coming from 65 countries all over the world right here in this hall?

Desiree Ho: Yes. So to welcome all of you to this meaningful occasion, let us first invite Mr Charles Mok, the Chairman of the Internet Society Hong Kong and local host of APRICOT-APAN 2011, to say a few words.

Charles, please.


Charles Mok: Dr Cerf, Dr Zhang, Permanent Secretary Chair, our honoured guests from all around Asia Pacific and, indeed, the world and locally from Hong Kong, of course, welcome to the opening session of APRICOT-APAN 2011 in Hong Kong.

This year, APRICOT is returning to Hong Kong for the first time since 1997 and for the first time, APAN is held concurrently with APRICOT, bringing together what I would call the plumbers and the engineers behind the net, with the heaviest and most innovative users from the industry, as well as the R&D community.

APRICOT-APAN is, I think, very important for Hong Kong. As I said, this is the second time that APRICOT was held in Hong Kong and our internet industry and community has come a long way since 1997, when it was first in Hong Kong.

This event -- APRICOT-APAN -- I hope, will emphatically reaffirm Hong Kong's role as an internet hub for Asia and, indeed, the world.

And, in fact, in an especially unique and important role for Hong Kong as a gateway for China, the country with the largest internet population today.

APRICOT-APAN is a two-week event and we are already into the second week actually today.

Last week, we held a series of very successful and well attended technical workshops at Hong Kong Cyberport, which is actually our workshop sponsor.

We thank Cyberport for their support.

This morning, we already had a very special section on DNSSEC. And now we are reaching yet another climatic point of APRICOT-APAN with the upcoming keynote addresses from Dr Vint Cerf, Chief Internet Evangelist of Google and, of course, one of the fathers of the internet; and Dr Zhang Ya-Qin, who is the Corporate Vice-President of Microsoft.

Now, for a conference about innovation and ground breaking use of the internet, how can I not tell you about the network infrastructure that we have built for this event? For the first time, for a conference of this kind and nature and scale in this venue, we have set up a network running at 100 gigabits per second, connecting you to the outside world over the internet. This enabled us to do live webcast, as well as remote participation through our apricot-apan.asia website to all parts of the world.

And we have set up 70 WiFi access points all around this conference meeting room and 12 of them are serving you right in this room. Possibly for the first time in this facility, we have both IPv6 and IPv4 running on this network. engineers that are working last night and actually through the last couple of days, that we should leave this network behind in this facility and it will be a great showcase for this facility. We will simply send the bill to the Convention Centre later on.

LAUGHTER But, seriously, if you have a chance to stop by the exhibition outside, please take a look at that little glass room that you will see that we have set up racks of equipment with blinking lights and what is actually running this network that you are using now.

Our IT infrastructure team made up of volunteers have been working day and night throughout this past weekend to put it up. And many kudos to them and thanks for their great effort.

A few words about ourselves, Internet Society Hong Kong. At ISOC HK, we believe that the internet is for everyone and we can all make a difference.

And that is why we like to say: don't ask what the internet can do for you, do what you can for the internet.

With this principle in mind, we have have been and will continue to focus our attention on promoting the adoption of new technologies and standards, such as IPv6 and DNSSEC, and the protection of a healthy and free environment for the internet to grow and expand, including participating in and even driving locally and regionally, the policy issues that affect how users use the internet every day. We welcome all of you in Hong Kong to join us.

Last but not least, I want to thank all our sponsors and supporting organisations, although really there are too many for me to name one by one, but you can see them on the board.

In particular, I would like to thank our co-host, Hong Kong Internet Exchange; Hong Kong Internet Service Providers Association and Joint University Computing Centre; our host fibre broadband service provider, PCCW; our diamond sponsor, Cisco; and our platinum sponsor, APNIC, and all the other sponsors and supporting organisations, without whom we would not be able to put this event together. Thank you very much.


Desiree Ho: Thank you, Charles. I'm sure the Convention Centre will be very happy to get the bill.

Next, may we now invite Ms Elizabeth Tse, Permanent Secretary for Commerce and Economic Development of the HKSAR Government, please.


Elizabeth Tse: Dr Smith, Prof Wu, Dr Cerf, Dr Zhang, Mr Mok, distinguished guests, ladies and gentlemen, good afternoon.

It is my great pleasure to join you here at the Joint Opening Ceremony of APRICOT-APAN 2011, which marks the biggest internet conference in Asia and the first ever joint event of APRICOT and APAN.

I would like to congratulate the Internet Society of Hong Kong and the DotAsia Organisation on the successful launch of the event. I also welcome all participants, particularly those from overseas.

As our great philosopher Confucius said over 2,000 years ago: how happy we are to meet friends from afar.


Elizabeth Tse: I'm delighted that Hong Kong has been chosen as the host city for this remarkable event, which aptly reflects our position as the leading digital city in the world. You made the right choice.

Hong Kong is a place where the east and the west blend together to form Asia's world city, a free, open, progressive and vibrant society. I will elaborate on these attributes.

Free and open. Hong Kong possesses world-class infrastructure for free flow of capital, people, goods and information.

We have no censorship on content transmission, which is crucial to the sustainable development of both the information and communications technology industry and the creative industry, as well as other sectors, such as financial, trade and logistics.

We have been rated the world's most free economy by the Heritage Foundation since 1995. We are rated the second in the world by the World Economic Forum in terms of openness to international trade.

Progressive. Hong Kong has achieved advances on many fronts, including the technological front.

Thanks to the liberalisation of the telecommunications market, Hong Kong offers one of the world's most affordable internet and mobile phone services. Over 81 per cent of our households are connected to broadband internet. Our mobile subscriber penetration rate exceeds 180 per cent, with costs amongst the lowest in the world.

The government now provides free WiFi access at public WiFi hotspots in Hong Kong exceeds 9,000. We will be expanding the program soon.

With easy and affordable internet access, the monthly mobile data usage in Hong Kong has surged to over 1,400 terabytes in 2010, more than double the figure in 2009.

These advances are underpinned by our advanced telecommunications network, with over 3,200 gigabits per second of activated capacity connecting to the mainland and foreign countries, as well as by the strong research and development support offered by such world-class facilities as Cyberport, the Science Park and the Applied Science and Technology Research Institute.

This is evident by Hong Kong being ranked the third worldwide in the Global Innovation Index Report 2009/2010 by INSEAD, the business school for the world.

Improving on e-readiness is important, but leveraging on Hong Kong's e-readiness to improve our economy and the quality of living for the society is equally, if not more, important.

For this reason, Hong Kong will spare no efforts to bridge the digital divide, improve web accessibility, improving on the wise use of the internet, et cetera.

Another attribute of Hong Kong worth highlighting is vibrancy. Our ICT industry is bubbling with energy and great ideas. Having a close partnership with the mainland of China, we are developing into a hub for technological cooperation, trade and innovation in the Asia Pacific region and even worldwide. Seeing domain names as an important piece of internet resource and infrastructure, Hong Kong will be introducing .hk (Chinese spoken) Chinese domain names later this year.

This will enable hundreds of millionsof Chinese-speaking people to access websites of Hong Kong companies and organisations easily by using web addresses wholly written in their mother language. This will enable e-business development for the Chinese-speaking internet community, strengthen Hong Kong's position as a technology hub and promote the digital economy of Hong Kong. All these will certainly make Hong Kong even more vibrant than ever before.

I'll now touch on a couple of issues of common interest. Recent years have seen the rapid growth of mobile connectivity. The number of mobile subscribers in the region already exceeds 2.6 billion, while globally this number has surpassed the 5 billion mark in July 2010. We have also seen the rapid convergence between telecommunications, fixed and mobile, broadcasting and IT services.

These prompt us to reflect on how we could best use connectivity and convergence to improve the quality of life of the people in the region. We witnessed the explosion of user-created content and the popularity of social networking on the web.

Their impact on individuals, lifestyle and the society is enormous. Yet, they also raise concerns over information security and infringement of intellectual property rights.

The exhaustion of IPv4 addresses is also one of the issues facing the internet community. To lead by example, the Hong Kong Government Network now supports Internet Protocol version 6. Community wise, our internet infrastructure also supports IPv6, including the Hong Kong Internet Exchange on Routing of Internet Traffic, HKIRC on Domain Name Registration and Resolution Services, Hong Kong Cyberport on provision of IPv6 services to its tenants. There are also commercial IPv6 services already available in the market.

If industry predictions about the transition of internet-based products and services to the cloud computing model are anything to go by, we are going to see many more changes, challenges and opportunities on the way. We encourage the industry to seize the opportunities and help the economy prepare for the new cloud computing advances.

All these call for the collaboration of the ICT industries, the ICT professionals and the governments in the region to seize the development opportunities and to cope with the challenges.

Regional organisations like APRICOT and APAN are no doubt instrumental in the process.

In the past week, through the workshops and in the coming week, through the conferences, these issues have been and will be covered in depth.

Albert Einstein has often been quoted as saying: the wireless telegraph is not difficult to understand. The ordinary telegraph is like a very long cat. You pull the tail in New York and it meows in Los Angeles. The wireless is the same, only without the cat.

Although the internet and related issues are much more complicated than wireless telegraph, through that the issues will be intelligently articulated and become more easily tackled as Einstein's cat-less telegraph.

In closing, I would like the wish the APRICOT-APAN 2011 great success.

To our overseas participants, I also wish you a joyful stay in Hong Kong. There's plenty for you to explore in this vibrant city, not to mention our charming skylines and famous Chinese and international cuisines. You can enjoy the world's largest permanent light and sound show, A Symphony of Lights, to be staged on both sides of the Victoria Harbour every night, to experience the attraction of the pearl of the orient.

We, the people of Hong Kong and the ICT industry, wholeheartedly welcome your visit. Thank you.


Lily Kung: Thank you, Elizabeth.

Next, let's invite Dr Philip Smith, Chairman of the APRA Board, please. Thank you.


Philip Smith: Good afternoon, everybody. I'm going to try and keep this as short as possible, because I'm sure we're all waiting to hear our keynote speakers speak.

I don't think I've ever stood at this podium at an APRICOT and looked out at the huge sea of faces that I'm seeing here. This is absolutely amazing, impressive attendance. I'm almost speechless, I would say.

I hope you enjoy the event that's coming up over the next five days. Some of you here have already had five days of workshops, as has been mentioned.

So I look forward to it. I hope you have fun.

Please go to all the sessions that we have put up for you, both APRICOT and APAN. APRICOT folks, please go and have a look at APAN sessions. They're doing interesting stuff. APAN folks, please come and have a look at APRICOT. We do interesting things too that's probably slightly different.

The whole idea is to have the two communities -- the commercial internet and the research and education internet -- know what each other is doing, be more involved with each other, make friends, make good business relationships and we can help build a better internet in the Asia Pacific region.

With that, thank you very much.


Lily Kung: Thank you, Philip.

Next, let us invite Prof Wu Jianping, Chairman of APAN.


Jianping Wu: Good afternoon. All our guests, our friends, ladies and gentlemen, on the opening of the APRICOT-APAN 2011 Hong Kong, on behalf of APAN, I would like to extend congratulations to this great joint event, as well as convey my warm greeting to all of the experts and guests, and a special greeting to those who have travelled thousands of miles to participate in this event.

Since APAN was established in 1997, this is the first chance for APAN and APRICOT to be gathering together in partnership with APNIC 31. And it will be a challenge for the organisation, the team, for a big conference, but it will provide a very good platform for the scientists and the experts and networking in Asia Pacific region to exchange their views and ideas ...

Hong Kong is a very good place for this big conference, that most experts and guests have travelled easily and conveniently for the meeting together here. Here, I would like to take this opportunity to express my thank you to the organisation team, including the program committee, API, APNIC and the sponsors -- so many industry sponsors here -- our host team, the APAN Secretariat and so on.

It is your great effort and impress us with this remarkable joint event in Hong Kong and I think this is a very good opportunity for the networking conference in Asia Pacific.

Finally, I hope this conference is successful.

I hope everybody enjoys this event, shopping and tour. Thank you very much.


Lily Kung: Thank you, Prof Wu.

May we invite Mr Paul Wilson, the Director General of APNIC, please.


Paul Wilson: Thank you very much. It really is an honour to be here with this cast of VIPs and dignitaries, with quite a number of good friends and a huge crowd of participants from here in Hong Kong and all over the region. This looks like being a fantastic APRICOT. I really am honoured to be here.

You might know that APNIC's role in this region is as the Regional Internet Registry. We distribute IPv4 and IPv6 addresses to those who need them to You might also have heard just earlier, if not before, that we're approaching a very interesting time with the exhaustion of the available IPv4 address space.

That's an exciting time and it's also a really critical time for this region looking forward and we do have a lot of work to do.

If we take a moment just to look back on the work that we have done to date, I think this community has a lot to be proud of in the efforts that we've made and the grassroots bottom-up efforts that we've made to extend the v4 pool, for as long as we have, in the face of this incredible wave of growth of the internet. And it really is our great success so far.

But I call on the community, on all of you, to continue the work at this important time. Please join us in helping to manage the ongoing audit rundown of the available IPv4 space. Please also use this opportunity to set your sights on IPv6, which is the next generation, as I hope you all know. Work together here to share your knowledge and experience, also your questions and your challenges about IPv6 deployment.

We have policy sessions about address management.

We have plenary sessions about life after IPv4. We have v6 training and workshops. So there's a lot to take in and I hope it's of great value to everyone, because the question now is not whether we're deploying IPv6, but it really is when and how we're deploying IPv6. So please, if you haven't already, then start this conversation.

The internet, as we all know, is a realm of incredible competition which has brought benefits to everyone. Within this community, though, it's also a realm of incredible collaboration and that's what APRICOT, the APNIC meeting, is all about. So please take the time to learn, to talk, to argue, and get the most out of this week, for a most, I hope, productive and enjoyable time for everyone. Thank you very much.


Lily Kung: Thank you, Paul.

Lastly, let's invite Mr Edmon Chung, Chief Executive Officer of DotAsia Organisation and local host of APRICOT-APAN 2011.

Edmon, please.


Edmon Chung: Thank you, everyone.

From DotAsia, we are really very honoured to be hosting the APRICOT-APAN event, especially because DotAsia was actually born from this community, from the APRICOT community.

In fact, just over 10 years ago -- actually, more than 10 years ago, now that I think about it, the first idea for having a region-wide domain name happened here at the APRICOT community. It was, I think, first brought up by -- I hope you're among the crowd right now -- Prof Qui Nam Chon. So we really are excited about this. Since then, we have gone through a long process, over many years, to bring DotAsia into an actual reality and we are really excited that today, we can start to give back to the community.

In fact, DotAsia, as a not-for-profit organisation, that is exactly what we're doing. And our mandate, to promote internet development and adoption around Asia, is driving why we're committing into this community. Also, I think besides hosting this event this year, in the future, I think one of the things that I want to let everyone know, besides welcoming everyone to Hong Kong, is that DotAsia is here to help the internet community around Asia. Please do not hesitate to call us up. We're more than happy to provide support.

As Philip earlier mentioned, it is this collaborative effort that made DotAsia possible and it is this collaborative spirit around Asia, from all around, from engineers, managers and businesses around Asia, no matter from mainland China, from Japan, from Hong Kong, from Taiwan, from everywhere around Asia, from India, from all around Asia, the collaborative spirit is what makes this conference happen, is also what drives DotAsia. That's why we think that this bringing together of APRICOT and APAN is really an historical event, for really bringing the people together, because it is the people that will make the internet better.

I hope that this time is the first time that we were able to bring APRICOT and APAN together and I hope it is not the last and I think this is definitely something that especially exemplifies the collaborative spirit around Asia.

One thing, besides thanking all the sponsors who made this happen, there's one person that I really want us to give a big round of applause to and to thank, because it is him who had the vision to bring together APRICOT and APAN and bring it back to


Edmon Chung: Also, I was sitting there and I was chatting with Paul just now and I was thinking that I remember that at ICANN, we have this little pin that has attended more than 30 meetings.

You are probably one, besides Philip, who has attended all of the 15 APRICOTs, so maybe we should introduce a 15-APRICOT pin. Che-Hoo, you would definitely be one of them to first get it.


Edmon Chung: With that, again, I welcome everyone to Hong Kong. And please, as Philip said, not only explore the APRICOT program, but also the APAN program, but also, most importantly, go shop, go eat, that's what Hong Kong provides you and Hong Kong is a very dynamic city and the restaurants are brilliant. Thank you.


Desiree Ho: Thank you, Edmon. Please remain on stage.

Now we would like to invite our local hosts, co-hosts, keynote speakers and diamond sponsor to come onto the stage for the official kick-off ceremony of APRICOT and APAN 2011 Hong Kong.

APRICOT and APAN have traditionally been two separate events, with APRICOT representing operational technologies and APAN representing advanced networks.

APRICOT and APAN will be officially joining together this year to make this the biggest internet conference in Asia ever.

This collaboration represents a significant moment in the history of the Asian internet community.

In this kick-off ceremony, the two glass panels on the podium represent APRICOT and APAN respectively.

Previously separate, they will now be joined by our guests on stage.

Lily Kung: Let us now invite Mr Che-Hoo Cheng, Chairman of the Local Organising Committee, APRICOT-APAN 2011; Mr Gerrit Wbahiman, Director of the Joint Universities Computer Centre; Mr Philip Leung, Managing Director of Hong Kong Internet Exchange; Prof Wu Jianping, Chairman of APAN; Ms Elizabeth Tse, Permanent Secretary for Commerce and Economic Development, Government of HKSAR; Dr Zhang Ya-Qin, Corporate Vice-President of Microsoft Corporation and Chairman of Microsoft Asia Pacific R&D Group; Dr Vint Cerf, President and Chief Internet Evangelist Google; Mr Charles Mok, Chairman of Internet Society Hong Kong; Dr Philip Smith, Chairman of APIA; Mr Paul Wilson, Director General of APNIC; Mr Edmon Chung, CEO of DotAsia Organisation, Mr Lento Yip, Vice-Chairman of Hong Kong Internet Service Providers Association; Ms Cho, General Manager of Cisco Hong Kong and Macau.

Desiree Ho: Thank you for your patience.

Now we would like to invite all our honourable guests to please place your hands onto the glass panel on your respective side of the podium and -- not yet, please wait -- together, slide the panel to the middle when we count to three. This will represent the official joining of APRICOT and APAN.


1, 2,3.


Lily Kung: Could all the guests please remain on stage to take a group photo, please.


Desiree Ho: Thank you, guests. You may now be seated.


Edmon Chung: Thank you again for everyone joining the event.

It is really my great, great honour to introduce our first keynote speaker today. And I have had the honour also to get to know Vint for more than 10 years now, back when he was Chairman at ICANN, between 2000 and 2007.

I was and still is pushing very hard for what is known as the introduction of the internationalised domain names, that's multilingual domain names, like domain names in Chinese and all those different languages.

But perhaps what is more important for today's occasion actually is that this APRICOT-APAN event would not have happened here without Vint. Because Vint actually played critical roles in the starting up of both hosts ISOC and DotAsia.

For ISOC Hong Kong, Vint was the Founding President of the Internet Society between 1992 and 1995. And for DotAsia, Vint, in fact, signed the contract for the creation of DotAsia into the root of the internet. That happened in 2006.

Not only that, I remember still that there was a long, drawn out process for DotAsia's creation and we had our share of controversies back then.

During the long process, there was a point where the ICANN board actually moved for a vote to reject it was Vint's one vote and in that critical hour that saved the day for DotAsia.

DotAsia -- and kept DotAsia alive and DotAsia, of course, later on, we actually received unanimous approval from the ICANN board to be established. So DotAsia is greatly indebted to your support, Vint, and that is why I'm saying that without Vint, we wouldn't have this event here in Hong Kong, at least it wouldn't look like what it is today.

As I was preparing for the introduction earlier today, I was extremely overwhelmed by the many, many, many, many awards and accolades that Vint has had. Please do look into the program booklet. It's very amazing. It's not only that about the volume, but also the different areas of expertise and different types of awards that Vint has got, that is really impressive and overwhelming.

I think everyone knows that Vint was the inventor of the internet's most important protocol, the TCP/IP, while he was at ARPA and it was also this work that won him what is called the Nobel Prize of the Computer Science, the Turing Award in 2004.

He also received a US National Medal of Technology from President Clinton in 1997 and the Presidential Medal of Freedom from George W Bush in 2008.

One thing that really stood out for me very interesting is in 1994, People Magazine identified Vint as one of that year's 25 most intriguing people in the world.

Among the many titles and positions on various boards, Vint serves on the Advisory Committee of NASA's Jet Propulsion Laboratory, where he's working on the design of an interplanetary internet. So when we're worring about running out of IPv4 addresses here on earth and the introduction of internationalised domain names, which are both such problems of the 1990s for Vint or so last century for Vint, Vint's vision is now extending well into our future and the future of the internet.

So without further ado, please join me in a big round of applause for the visionary, the father of the internet, Dr Vint Cerf.


Vinton Cerf: Thank you very, very much. That was an exciting introduction.

I feel as if I shouldn't take too much credit for anything to be quite honest with you, because if you take too much credit, eventually when they don't like whatever is going on, you have to take a lot of blame.

As a parent of two sons, I learned that important lesson a long time ago.

It's really a joy and an honour to return to Hong Kong and particularly in this joint session of APRICOT and APAN.

I have prepared remarks, but before I go into that, I was listening very closely to the various introductory speeches and I wanted to comment on some of them, because I resonated so strongly with them.

First of all, Charles, you mentioned the internet is for everyone. I am so strongly a believer in that. My title at Google is Chief Internet Evangelist, but as I look at the statistics, I realise that only 30 per cent of the world is on line and the evangelist has 70 per cent of the world's population to convert, so I ask your help in this room and your friends and everyone else to help build more internet, so that everyone can in fact be on.

Elizabeth, your speech was most stunning. It was just extraordinary with superlative statistics and I thought you outlined in very vivid terms, not only the benefits, but the challenges that we face in making the internet a constructive force.

I thought about this first opportunity for APAN and APRICOT to jointly sponsor a meeting of this type and I thought maybe you needed a kind of joint not to and I suggest for your consideration: APAN APRICOT, where ideas become engineering reality.

It's my hope that the academic community and the engineering and operational, commercial community will blend their strengths together in meetings like this to make internet a continually evolving system.

One thing which was mentioned several times in these opening remarks is the term collaboration.

I think that it is most strongly exhibited in this gigantic list of participating organisations to say nothing of each of you sitting in this room together.

Internet is indeed a gigantic collaboration.

There is no top down control. There is no requirement that anyone be part of it. There is no requirement that companies build pieces of it.

There is no requirement that they interconnect.

They do so because to interconnect is to increase the value of the system for everyone.

So collaboration is really a key element in all that we do. friends and I hope I have a chance to greet you personally, but if I cannot do so in the next couple of days, I want to tell you collectively, it's a joy to see my old friends here and I look forward to making many new ones.

Finally, Paul Wilson talked about the importance of moving to IPv6. I want to thank you for pushing that so hard. It's very important. The internet won't be able to grow if it runs out of addresses.

It's sort of like trying to sell a telephone without a telephone number. It doesn't work very well.

Let me begin now with some prepared remarks.

This is to go back some 42 years. The predecessor to the internet was called the ARPANet and four nodes were installed at UCLA in 1969. This is where the beginnings of a packet switching experiment began. I was a programmer at UCLA writing the software to connect the Sigma 7 computer to the packet switch of the ARPA Net.

The Sigma 7 is in a museum now and some people think I should be there, too.

LAUGHTER If you skip forward to the invention of the internet, which, by the way is the creation that Robert Kahn as well as my work, to say nothing of literally hundreds of thousands, if not millions of people, subsequently. Once again, I remind you, it's a collaboration.

This picture helps to emphasise the point. Each of the colours represents a different internet service provider or a different autonomous system.

This imagine is generated by looking at the global routing table to see all the different autonomous systems and what their connectivity is. So again I emphasise the collarative nature of this internet environment.

If you look at the number of machines that are visible publicly on the internet, you see a growing number. This does not represent all of the computers or devices on the internet. Many of them are hiding behind firewalls, many of them are only episodically connected, like laptops and mobiles and the like. So this picture is only the lower bound on the number systems that one can find on the network.

The number of users on the net exceeded 2 billion in 2010. I don't have the statistics for January 2011, but I would guess that we will see well in excess of 2 billion people on the net.

As has been mentioned earlier, 5 billion mobiles, of which probably 15 to 20 per cent are internet capable.

That fraction will grow over time.

Here is where it is estimated the users are and all of you in this room part of the Asia environment will recognise that you are now the most populous internet community, in addition to being simply the most populous group on the planet.

This is very important for all of us who are trying to do work on a global scale. We absolutely must pay attention to our Asian colleagues, your interests, your cultures, your languages, your strengths, because you are going to be reflected in this global internet increasingly. Here we see only 21 per cent population penetration of the internet in Asia, when you reach the 70 to 80 per cent that you see here in Hong Kong, the absolute numbers in the Asian populations will of course be dramatically bigger.

There have been mentioned already some important changes in the internet. This is probably the most momentous year in internet's history, in terms of functional changes that are under way, IPv6 of course being the most significant one because it's changing a fundamental protocol in the internet from the 32 bit format to the 128 bit format.

Second, the internationalised domain names which have been frankly a long time in coming are finally happening in the net.

Third, the security of a domain name system, which has been weak for many, many years, is becoming improved with domain name system security extensions, the root zone was signed in July of last year and now I see and look forward to penetration of DNSSEC in all top level, second level and subsequent lower level domains over time.

The routing system is also vulnerable to people simply announcing parts of the internet, even if they don't have the authority to use those address bases.

The regional internet registries are altogether looking at a mechanism for digitally signing the assignments of IP addresses to the various autonomous system operators. That will help us minimise what we call the poaching or hijacking of internet domain space or internet space.

There are in addition to that massive increases in the number of censor networks on the internet.

There are new devices that are becoming part of the Grid program, with devices that are -- appliances that use electricity are now able to report their use and can accept guidance as to when they should operate, so as to minimise and attempt to use electricity during peak level, that will reduce the cost of implementing power generation, so that you don't have to spend money on capital for very, very infrequently used generation resources and finally, of course, mobiles are becoming a major part of the net.

This may not be terribly visibly to folks in the back, it's simply to illustrate that the non Latin character sets are being part of the top level domain space. Here are some additional ones that either have been or will be approved for use on the net.

I would point out, if you see in the lower left, one of the scripts looks like it's a set of little square empty boxes. This is because my computer didn't know how to show the Sinhala alphabet.

I'm afraid that we are all going to experience things like this as the top level domains and other levels have these non-Latin characters in them and we are going to have to work hard to make sure everyone can display them.

There are a number of security problems on the net. This list is not complete by any means. I put it up only to remind you that there's still a great deal of work to be done to make the internet a more secure place for people to carry out business.

The point I want to make here is that unless people feel confident and safe in their use of the internet, they will tend not to want to use it for commerce. What better place than Hong Kong to emphasise the importance and value of using the internet as a tool for electronic transactions.

But in order to do that, we need to cope with many of the problems that I listed up there.

I'm going to go a little further here. Some of the reasons that we have these security weaknesses in the internet are consequence of operating systems that can be readily penetrated or browsers which download files from home pages and then become infected because of the execution of Java script or other high-level languages that install viruses or worms or Trojan horses or other kinds of malware on hosting operator system.

We also don't do very good job of access control.

We don't do a good job of picking passwords. We don't use two factor authentication. That's changing. But we as users are going to have to accept that our choices of passwords are not very acceptable. Some people use the word "password" as the password, because it's easy to remember.

Unfortunately, everyone else might know that too, so it doesn't give very good access control.

We need to change people's attitudes about the inconvenience of ensuring better security, but in fact we need it or we won't be able to protect them against themselves.

Another major problem is configuration.

Configuring things wrongly is really easy.

Understanding that you have misconfigured something that might have major consequences is hard.

It's especially hard if you want a piece of software to tell you that you've made a mistake in configuring this system, a router, an operating system or your own computer.

This is a subject of deep research. We don't have good solutions for automatically detecting bad configurations. We might be able to detect the consequences of a bad configuration and it's actually embarrassing that some of the worst misconfiguration events caused the largest problems in the internet, like misrouting and black holeing and things of that sort.

We also run internet the problem that if we are not good about protecting our machines, they become invaded, they become hosts for worms and viruses or Trojan horses and then they become part of what we call botnets, robot networks, that are used as resources by other people to generate spam, to generate denial of service attacks. It's actually a line of business for people.

For a while, I was afraid that the botnets would be used to simply destroy the internet. But now it's clear that there's a business model behind the botnet herders. They don't want to destroy the internet. It's a way of generating revenue. So they would like to protect it.

In fact, they even go so far as to protect their captured computers, so some machines that have become part of a botnet have code in them to defend against the other botnet generals that are trying to get access to the machines.

So it occurs to me that if some smart people among us could invent a good botnet, it would protect all of us from all the bad botnets out there.

So we have to know which botnet to joint in order To make matters worse, the process of abuse has extended now from simple hacking, possibly just to show that you could do it, to organised crime and other kinds of organised abuses of the internet.

There are huge numbers of policy implications that are international in scope about trying to defend against these various bad practices. I don't have time to review those today, but I want to emphasise that the solutions to these problems are not entirely technical and the solutions will require international cooperation; once again, that word cooperation and collaboration being a key part of internet success.

Speaking of collaboration, there's another phenomenon in 2011 which has been brewing for several years and that's cloud computing. It's a buses word. It's not a well defined term. There are many different clouds that are offered by many companies. Google uses cloud for its work, Amazon and Microsoft and IBM and others are implementing cloud based executing systems.

The reason I put this slide up is to emphasise that it is the interaction between clouds that is going to become very important.

We are in 2011, where the internet was in 1973.

There weren't computer networks in 1973. They were proprietary in their scope. IBM had systems network architecture. Digital Equipment Corporation had DECNet. Hewlett Packard had something called DS.

The proprietary machines could be linked together very effectively. Digital equipment machines on DECNet, IBM machines on SNA and Hewlett Packard machines on DS. They just didn't talk to each other. The networks didn't even know that any other networks existed.

Into that vacuum comes the TCP/IP work to create a language that would allow those different networks and the computers on them, to interoperate.

Here we are in 2011. We have very disparate cloud-based systems. They don't know how to talk to each other. They can talk to clients on laptops and desktop and notebook machines, but they don't talk to each other.

To give you an example, most of the clouds don't know that any other clouds exist. They can't refer to another cloud. There's no language for it. If I have information in cloud A and I want to move it to cloud B, I don't have any way to tell cloud A and cloud B to cooperate to move data. I don't know in what way they should talk to each other to move the data from one place to another and if I have access control in cloud A, to protect my information, that me to data needs to move to cloud B in order to perform the same protection. I don't have a vocabulary for saying how to move the data and how to interpret the data. I'm sure you can generate your own long list of things that cloud should be able to do together. But that's the task of this 21st century second decade is to figure out how to achieve that interconnection.

There are other challenges in this digital age that we're in. I don't have time to go through all of them. I do want to draw attention, as if enough hadn't already been drawn, to the treatment of intellectual property in a digital environment.

We all understand that digital information is easy to replicate and it's easy to distribute.

It's very clear that in an environment where the replication of physical objects is hard and therefore let's say somewhat easier to protect against; in the digital world, it's very difficult problem. In fact, the business model of controlling copies may turn out to be very hard to implement in this digital world. We have mechanisms for digital rights management, we have cryptography in order to protect information that needs to be accessible, those with authorised access. But I think that it would not hurt for the legal community and the engineering community to sit down together, maybe this is something that APAN and APRICOT could attempt to foster, to talk about the technology of the internet and worldwide web and how one could adapt the kinds of rules for intellectual property that serve the old environment, how to adapt that to this new world.

It may be that we are simply trying to fit an old model into this very dramatically new, very different economyically structured system.

I want to finish this slide by mentioning that each piece of software that you use, each application that you use, often generates a complex digital file. This digital file might be a spreadsheet, it might be a presentation, a PowerPoint kind of presentation, it might be some other very complex object used for playing digitial games.

That complex file is not meaningful unless you have the application software to interpret it.

Here is the long-term concern I have. Let's done a Google search and you have turned up a 1997 PowerPoint file. So the question -- imagine you are running Windows 3000, so now the question is: does the Windows 3000 system know how to interpret a 1,000 year old PowerPoint file?

Before my friends at Microsoft think that this is a gratuitous dig, it is not. Even if we were using open source software, the idea that a piece of open source software can still interpret a 1,000 year old file is still a big challenge.

What I'm worried about is that as time goes on and as software evolves, that we may not be able to interpret the bits of the files that we created with some of these complex programs.

It could be that a company that makes the programs says I can't support this program any more and I'm going to move on two something else and it's not backwards compatible.

Or they might say this application used to work on this version of the operating system, but it doesn't work on the new version of the operating system.

You can make up your own scenarios for some of these problems.

There are intellectual property questions. What if the application has to be preserved even if it's not supported any more, it has to run on a particular version of the operating system and maybe that operating system has to run on a certain class of hardware which might not exist any more.

So what to do about preserving the bits that we create over time? We may have two emulate the hardware, get permission to keep running an older version of the operating system and the application that ran on top of it in a cloud environment in order to preserve the meaning of the bits that we're producing.

If we don't find a way to do that, I am afraid that in the year 2100, people will wonder about the beginnings of the 21st century, all of our information will be digital, but it won't be interpretable, because the applications won't work any more.

If you don't think this is a problem, I guarantee you it's already a problem. Some image formats, for example, are not properly treated with newer software because backward compatibility was just too painful to maintain.

This is a serious technical problem. It may also require some rethinking about who should have access to software after it's not supported any more. You could ask can clouds be helpful in maintaining access to older software? So I leave that to your imagination.

I want to mention also that this, in terms of internet evolution, more and more devices are showing up on the net. In Europe, it's sometimes called the internet of things. There are an increasing number of internet enabled things on the net, things like refrigerators, picture frames, things that look like telephones, but they are internet enabled surfboard. I guess he was sitting on the water on his surf board thinking, if I had a laptop in my surf board, I could be surfing the internet while I'm waiting for the next wave.

So he put a laptop in the surf board and a WiFi service on the rescue shack and now he sells this as a product.

Earlier, I mentioned sensor networks. I wanted to mentioned one that I have in the house. This is a commercial product. It was made by a company called Arch Rock, which has been acquired recently by Cisco. They run six LOPAN IPv6 radio mesh network, so each of the sensors is, in fact, a small store-and-forward node. It's maintaining connectivity by measuring radio linkages among all the devices in the house. Each of the rooms of the house, the 12 principal rooms of the house, has a sensor that is measuring temperature, humidity and light levels every five minutes.

That information is recorded in a server in the basement and I get reports every -- times a day on what the status of the house network.

One of the rooms in the house is very important to me. It's the wine cellular. It has a couple of thousand bottles of wine in it so that room has been alarmed. If the temperature goes above 60 degrees Fahrenheit, an alarm goes off and I get an SMS on my mobile. A few months back, I was visiting the Oregon National Laboratory in the state of Illinois in the United States and as I was walking in the door, the mobile went off, it was the wine cellular calling and it says you have just broken through the 60 degree Fahrenheit barrier. Unfortunately, my wife was away on holiday, so she wasn't there to reset the cooling system. For three days, every five minutes, I got a note saying your wine is warming up.

So I called the Oregon guys and said you made remote actuators and they said yes and I said OK, as actuator to put in the cooling system we so the 15 year old next door doesn't reprogram my house while I'm away.

Then I got thinking, I can tell if someone has gone into the wine cellar because I can but I don't know what they did in there. So I thought why don't I put an RFID chip on each bottle and then I can do an instantaneous inventory of all the bottles in the wine cellular and that would tell me if any bottle has left the wine cellular without my permission.

Then one of my engineer friends was de-bugging the design and he said: you know, could go into the wine cellar and drink the wine and leave the bottle.


So I had some debugging to do to fix this design.

In the end, this turns out to be a very practical kind of system to have, though, because now I'm going to have to put sensors in the cork of each bottle. As long as you're going to that much trouble, you might as well detect the state of the ester in the wine that give it its various flavours.

So before you open the bottle, you interrogate the cork. So if this is the bottle that got up to 90 degrees Fahrenheit because the cooling system failed, that's the bottle you give to somebody who doesn't know the difference.


So this is a really practical kind of system to build and there will be more of them in the future, I'm sure.

I'm going to finish by giving you a quick status report on the interplanetary internet. Everyone who hears this for the first time thinks I have gone crazy and I'm waiting for the Martians to come and communicate.

This project was started because my colleagues at the jet propulsion lab and I believed that we needed richer net working for space exploration than simply point to point radio links. We imagined that there would be multiple sensors on the surface of the planet that needed to be locally networked, that there would be multiple orbters and mobile devices as there are today on Mars and we wanted a rich networking environment. We started out thinking that TCP/IP would work. This is a picture of Mars in Google Mars, in case you haven't gone there, this is the kind of picture that you can see in Google Mars as.

The store and forward networking on Mars has gotten fairly elaborate. The rovers in the lower left were originally designed to communicate back to earth with a point to point raid you link at 28 kilobits per second. When they were turned on, the radios overheated and they had to reduce the duty cycle to keep the radio from harming itself. The scientists were upset about that because 28 kilobits was already pretty slow and now they weren't going to get continuous transmission from the surface.

The programmers at JPL recognised that they had another radio which couldn't go all the way back to earth but it ran at 128 kilobits per second and it could reach up to the satellites that were in orbit.

The radios were compatible so they reprogrammed the orbter and the rovers so the rover could transmit data up to the orbiting satellite. The orbiting satellite would hang onto that data unill it got to the right place to transmit back to at least what packet switching is. That is how the internet works.

The information that is coming back from Mars is coming back with a storage forward system. This works so well that when the phoenix lander landed at the North Pole of Mars, they designed it around this store forward system. So we got quite excited about that demonstration.

I just want to emphasise that when we started, we thought we could use TCP/IP to make all this work and it would work fine inside the spacecraft. It will work on the surface of Mars just as well as it works on the surface of earth, but it doesn't work between the planets.

The problem is that the speed of light is too slow. If you look at the distance between earth and Mars, when we are closest together, it's 3.5 minutes one way of propogation delay. When we're farthest apart, it's 20 minutes propogation delay and 40 minutes round trip.

Can you imagine trying to run a web page where you hit the mouse button and 40 minutes later, the first bit shows up? I know some of you have experienced networks like that here on earth -- LAUGHTER.

-- but it isn't because of the speed of light delay. It just gets worse when you go to the outer planets.

So we said: OK, this is an example of what we'll call variable delay and disruption. The reason it is disrupting is that the planets are rotating and


So if you're talking to something on the surface of the planet that's rotating, eventually it's on the wrong side of the planet and you can't talk to it until it comes back around again.

So we had to develop a new suite of protocols that we call delay and disruption tolerant networks.

These protocols have now been tested terresterially.

We had the reindeer herders in northern Sweden helping us test these things by having all-terrain vehicles with the DTM protocols in them and, say, we would approach a village with a WiFi server, we would dump things off and put things onto the all-terrain vehicles.

We have had tests in the International Space Station. We have 3D devices that are on board using these new protocols. We put the protocols on board what was called the deep impact spacecraft several years ago, which rendezvoused with Tempel 1 and fired a probe into the comet. That spacecraft revisited another comet recently called Hartley 2 and we uploaded our protocols to that spacecraft which is in eccentric orbit around the sun.

As the 2011 year unfolds, I'm hoping to have multi-node testing of the new interplanetary protocols.

Why would we do all this? What we are hoping in the long run is to grow an interplanetary backbone both to support manned and robotic exploration. If all of these space-faring nations of the world adopt the new interplanetary protocols, then they will be compatible with each other, they will be able to exchange data and when a spacecraft has completed its primary mission, it could be repurposed to become part of a node of an interplanetary backbone.

We don't suggest that we should literally build an interplanetary network and hope that someone will use it. Rather, we're saying, let's grow one, mission by mission, each mission taking place because of scientific interest, but leaving behind a potential element of an interplanetary backbone.

I won't live to see the end of this. Many of you will see more than I will. But I think that it's very exciting to be part of a project which takes what sounds like crazy idea and turns it into an engineering reality.

That brings us back to APAN and APRICOT, where ideas become engineering reality.

Thank you very much.


Charles Mok: Thank you, Vint, for most practical yet visionary remarks.

And to get us back on earth, it's my great honour to introduce our second keynote speaker, Dr Zhang Ya-Qin, the Corporate Vice-President of Microsoft Corporation and Chairman of the Microsoft Asia Pacific R&D Group.

Dr Zhang is well known for his leadership in driving Microsoft's overall R&D efforts in the AP region. He has published more than 500 papers in leading journals around the world and been granted 50 US patents in digital media, internet, multimedia, wireless, satellite, satellite communications. And many of these technologies he has developed has become the basis of many start-up ventures, commercial products and international standards.

He's also an advisor to five provincial and also Central Government agencies in China. He has been an honorary and guest and adjunct professorship he has held for over 20 universities in China, Australia, Hong Kong and the US. He's been working very closely with the HKUST in Hong Kong.

He is highly regarded as a technology and industry visionary and received many awards, such as the IEEE Industry Pioneer's Medal and the Centennial Medal.

His pioneering contribution in digital video, mobile communications and internet technology has made him one of the visible industry leaders in China and has he the distinction of being included in the top 10 CEOs 2007, IT leaders of the year 2005 and top 10 innovators of the year 2007.

Let us all welcome Dr Zhang.


Ya-Qin Zhang: Thank you very much for the kind introduction. Good afternoon, everyone.

Very honoured to be here. It's actually a big challenge to talk after Vint Cerf. I remember it was probably around 10 years ago, he and I had exactly the same sequence in globe come and I felt the pressure then, I feel the pressure now.

But I promise I will talk about the things on earth.

I want to thank the APRICOT and APAN for inviting me here. This is indeed a very impressive event.

This is my first time and I truly enjoy interaction with friends and people from industry, academia and also the policymakers.

Some of you know I happen to spend this month in University of Science and Technology. In Microsoft, they give two month sabbatical every seven years to some of the R&D executives, so I'm spending this entire month in Hong Kong.

Through the month, I truly learn quite a great deal from the professors, the policymakers and I'm really impressed about the quality of the professors and also the overall enthusiasm of the internet industry and the permanent secretary, Elizabeth Tse just gave a very great summary of the internet in Hong Kong, it's all about. I am very impressed with the number of hotspots, what is it, almost 10,000, broadband penetration and the advanced infrastructure.

When I talked with the Chairman of Internet Society here in Hong Kong, Charles Mok, about my topic and he told me, first, talk about the state of China internet and then you talk about the future of internet, what you envision the technology is going to be in 5, 10 years and you should talk about Microsoft and you should do that within 30 minutes, plus it cannot be too technical, because the breadth of audience. So I will do just that, try to accommodate all this within 25, 30 minutes.

First, what's happening in China's internet? It's just so exciting. Nothing like that, industry like China's internet. Right now, China has the largest population, over 400 million users, number 1, in the mobile internet and number 2 on PC and it's going to take over US next year to become the largest PC users in the world.

The second, it has a very different user habits and demographics. The average Chinese user is 25 years old versus the US, which is 42 years old.

A lot of people use internet from a cafe, about 40 per cent. In US, it's mainly from home and some from work.

Also, if you look at the top three applications of internet, it's quite different. In China, it's music, video, some communication chatting, and news.

A lot of gaming.

In the US, it's mainly search, IM, there is shopping, it's very different.

Also, in China, one thing that's unique is the domestic companies are really doing very well, very different from other countries.

The multinationals, like Google, like Microsoft, and Amazon, we are actually working really hard, struggling to an extent.

There are a lot of reasons. If you look at -- I list just three. The main reason is really the lack of agility. It takes too long for us to take decision. We will improve, but if you look at the top 3 Chinese internet companies, Tencent and Baidu and Alibaba, the market happens over a hundred million and it's growing.

There's a lot of criticism in the China internet.

Despite the tremendous market success, there are a lot of people saying there's not much innovation, it's all copycat. The business model and technology are imitations.

I personally disagree with that comment, statement. Indeed, China has learned a great deal from the US in the last 10 or 15 years, pretty much everything. But there was a lot of innovation, innovation especially how to adapt to local market and how to provide better service, so I would characterise is probably 80 per cent learning and 20 per cent innovation and innovation going to become a lot more enhanced in the future.

Some very good examples. People use Weibo and learn from Twitter, but it goes way beyond. Tencent virtual goods and Sohu IME are some very good examples. In fact, if you look at the features, the capability are probably better than some of the tools, technology from multinationals.

Competition is also very intense, fierce. I am going to talk more about that, but a good quote from the Mercury News: the Chinese internet is like gladiatorial, a no holds barred fight to the death.

There is no ground rules. Sometimes companies do get carried away.

As you know, the recent war between Tencent and 360, some of the Chinese audience probably know this story very well. I'm not going into the details, but this is a typical case that two companies competed and somehow lost the rationale why they competed. Both had regret it and they probably should have done things differently.

Also, for people who use Kaishin, there are two Kaishins in China some time ago. One, the real Kaishin and the other was fake. This was the Kaishing domain name, but then it kind of completely copied the games and a lot of people got confused.

How come there are two exactly the same company?

Also, imitations between the Chinese companies.

I'm hiding on the right, I hid the company name.

The one company came out with this browser and pretty cool experience and the other, there's exactly the same thing, same UI, same experience and much bigger audience.

Also, the business models in the China internet are quite creative. Gaming is a huge thing in China. You can buyamunition, you can buy weapons to become more powerful, more skilled. You can get reach the level.

There are companies you can hire to do that for you. They charge probably 15 per cent of the goods you buy, but you can get elevated instantly.

Also, there is a company Kaishing doing the social gaming with stealing vegetables, parking lots, a lot of people in China get addicted to it. It became instant success, probably three or four months, almost 200 million users.

At one point, when I came to my office, and I see a lot of people just playing this game during daytime, during work day, and I made a tough decision.

First time in my life, to censor this particular site, actually block it for three months, which I normally wouldn't go, but there are a lot of the young kids.

I have about 3,500 engineers working in Beijing and Shanghai, a lot of people just get so addicted to that game, so I decided to do that. It's not a good example, but sometimes you have to make tough decision.

I recently also trying to sign up to the IM service. I'm not going to mention the company.

When you sign up, it's free. So they provide you all this beautiful and nice fashionable clothes and a large selection.

Then the second month, they ask you to pay 5RMB.

If you don't, they are going to scale you down to somebody who is not as cool.

The third month, if you refuse to pay, they are going to take off your clothes.

Imagine the next month you don't pay.

I didn't use my real name, by the way.

The other thing is the internet is really a reflection of the society. There are a lot of issues, growing pains in China. So internet really function as a mirror of what the society is all about.

Obviously, in China a lot of people don't like corrupt government officials and on the right picture, that was a director of housing and real estate in Nanjing and he made some remarks which I forgot the remarks he made, in fact, was probably the housing price should be this or that. Then people didn't like that remark and within24 hours, people started to try to find how much his watch worth, how much a pack of cigarettes he was smoking and then they concluded that as a government official, his income would not be able to support that kind of spending.

This guy right now is removed from office, government has started a very serious investigation and he's gone. He's probably in jail right now.

Another case which is very famous in China recently. I'm not going to go through it, but "My father is Li Gong". A lot of people here know that story.

People also want instant celebrities, especially for the. On the left, it's actually a beggar who looks really cool and then they made him really celebrity on the internet and right now, he comes to talk shows and make probably a lot of money, I don't know.

Other thing is Calvin Klein model, this picture is probably fake, I don't know, but people want the disadvantages to become winners, people don't like corrupt government officials, people don't like a lot of rich people who get rich overnight and this all reflects in the internet.

More recently, way would or the Twitter, microblogger, had become a major phenomenon. This took off literally within 18 months.

If you look at the Weibo or the Twitter, I really shouldn't call it Twitter, because it's quite different from original Twitter.

It really fits into Chinese culture really well.

It's a combination of short message and emails, social networking and it's very convenient for people to communicate.

In China, a lot of people are using short message.

This basically meets, fits their habit, plus can also allow you to interact with a lot of people. So it became an instant phenomenon.

I have two Weibo accounts, one in Tencent and one in Sino. I find it's actually quite effective. Any remarks you make will reach a lot of people.

Especially in Tencent, I have almost 10 million followers. In fact, if we write something, a lot of people know. It could be good and it could be dangerous.

Some of you probably know I got emails from you guys, too. A few days ago, I just made a simple Because Weibo allow 140 characters, so I actually, I had to separate my message into two parts, so the first Weibo, what I said, is with the exhaustion of the v4 addressing, it's a necessity to migrate into v6. I made a few comments, without the advantage of v6.

The second post, I said: on the other hand, people should not panic, because v4 still has some mileage and there are a lot of people working on the migration.

Then you see two headlines in Sina and Tencent.

One says: Microsoft China said, you know, v6 is necessary. The other said it's complete opposite, but it's the same message, so got to be very careful about this Weibo posting. It's really becoming immediate.

Also, friends, they don't read the report any more. Weibo is becoming their primary source for information and also primary source for communication. Watch out, this is going to be huge.

The second part of my talk is really about the future, what's going to happen in the internet in the technology space.

For people who are researchers, engineers, we really have four parameters, we play computing, we look at bandwidth, we look at storage, we also look at power.

I'm not going to go deep into this curve, but pretty much the computing storage or bandwidth all follow the law of somehow modified. In computing, we are getting into the exabyte era. IP traffic doubles every two years and especially in China, 80 per cent of this traffic is video.

People watch a lot of video.

Storage, get into zettabyte. Zettabyte is 10 to the power of 21. That's basically we are able to record the entire human history many times over.

This doubles every year, means next year, we actually, the amount of data we can storage is over all summation of all the data, all the storage capacity we have.

Power-wise, is a little bit different. Overall power consumption is going up linearly, although the efficiency for data centre and for your CPUs are coming down. But this is going to be a major challenge for the computing.

What is going to happen next few years? What are the technologists going to drive the growth of internet?

In fact, other keynote speakers have talked about some of this. I think, first, it's driven by the advanced in user interface. We call it natural user interface. All the smart devices, the new computing model, needs to have a better user interface. I'll talk more about this. But this for somebody who work on computer science, it's really how we define the generations of computing.

Mobile internet, where you have so many different smart devices and you have a network and how you are going to access the internet.

Last is the platform for cloud. It's a new architecture and completely changed the business model.

In the next few years, there will be three battles to be fought. I mentioned briefly in the press conference. The first battle is for cloud. Which company or which companies will be the winners of cloud computing?

There will be a lot of collaborations, Vint Cerf says, but the competition is getting really intense in this space.

The second is mobile platform. We use smart phones and we have a lot of different systems, which system is going to become dominant in space or which system?

The other battle, which is probably more fierce, is between the PC and the mobile. If you look at technology, there is no reason why PC cannot be applied to mobile and there's no reason why mobile cannot go to PC.

In the next five years, you are going to see those battles and whoever wins those battles will become the leader in the next generation.

Let me talk about the natural user interface.

User interface is always a basic research. Bill Gates said it is the area that requires the highest IQ. It started with text based, then move into graphics user interface in the 1980s, 1990s and then we keep enhancing the graphic user interface with better voice, with touch, with digital handwriting and now we enter a new stage, what I call natural user interface, in a sense you don't have an interface.

Meaning the UI contextual, is aware of the environment, the ambience, understanding the spoken language, it's able to do what people really do on a daily basis.

A key example is actually a very preliminary application for natural user interface, is touch surface table. The other example is Kinect, is a video game console that's peripheral to the Xbox 360, that is controller free with 3D and a depth sensor, motion capture and face recognition and it's really changed how we play games and it change how we interact with machines. It's not only for games, it's really a new metaphor that changes the interaction with computing.

Let me just play a short video. Some of you may have already used this device, the Kinect. It's actually something we currently selling around the world and we sold over 8 million units in the last two months. It's actually one of the best selling consumer products ever.

My team in China is really working hard and we are actually working with our partners in Guangdong around the clock trying to fill all these back orders.

Video played.

It doesn't need a lot of explanation, it's self-evident. This is not only for gaming, it's really a new way for us to control TV, to control any device. There's no controller. You are the user interface.

We are working very hard to add more features to this and actually in a few months, we are coming up with what we call the avatar connect and then you will be able to control the facial expression, the body motion, with all the cool features that you can interact in a very compelling manner.

My team in Beijing insist me put this slide, because these guys are working so hard on this technology. Obviously, you know, the bulk of work is done from Redmond, a lot of work fromCambridge, but I have a team in Beijing. They work probably for at least 10 years on different parts of computer vision, graphics, sensoring, recognition. So I actually often joke with those guys and say you guys probably saw the papers produced a lot of demos, but I haven't seen your technology put in use, so this is probably the first time we see the computer vision technology. Some of the recognition technology come to mainstream.

Let me get to the mobile internet. The best chart to talk about is actually ... you look at several different stages and concluded the mobile internet is probably five times or even 10 times bigger and I completely agree with her. In fact, if you look at China, it's already beginning to happen. Last quarter, the smartphone shipment was already bigger than PCs, over 100 million. This is worldwide.

Today's smartphone is a lot more powerful than my PC a couple of -- 10 years ago.

You see a lot of penetration of 3Gs in the LTEs and obviously, the interface is going to help move the smartphones into main stage.

I'm going to skip this, but with phone in fact we have a lot more things to leverage than your PC.

The location information. Some of the directional information which actually don't have on your PC.

One of the projects we work on is how to combine the current browser with new information, using augmented reality, multi-layers, browsing and navigation, in fact, it's one of the directions, one of the things we are going to have some products.

The other area is cloud computing. We are going to collect a large amount of data, we need computing, we need to store it and you need to communicate. Where can you store this data? How are you going to compute? Cloud computing is really a new infrastructure, new architecture that really is going to change the game.

As Vint mentioned, there are a number of different definitions for cloud computing. But it really happens in three different levels. The infrastructure, which you have to provide the basic facility to host the cloud. And you need a platform which people can write codes, people can deliver the service. Of course, you can provide the software as a service on a third level, which is application and data.

Cloud computing is really and all those things combined, it is data, it is the software, it is platform, and it's infrastructure, then delivered through a new service model.

The cloud is really a new vision that enables IT, just like water, electricity, as a utility.

A key element of cloud is the data centre. I'm not going to go to detail, but if you look at the progress over the last few years, the power efficiency, the overall scalability and the performance just have improved tremendously.

Right now, we are deploying data centre around the world, in fact, we have a data centre in Hong Kong that connects to mainland China and the rest of Asia, with many, many servers, I can't disclose the number. But Hong Kong is a very interesting place, it provides some of these services.

A key almost of data centre is product utilisation to 1.1 and obviously, the perfect stage is 1, which you have no loss of efficiency. 1.1 is really the state of art.

In China, cloud computing took place a little bit slower and it's more of a top down initiative. In the last two years, it suddenly became a major campaign. In fact, probably went a little bit overboard. Every city you visit, every place you talk, it is about cloud. There are clouds in five cities that make billions of investments and especially with the new initiative, I think you can see a lot of excitement in the Chinese cloud.

Microsoft will work with the government, and industry very closely to build the China Innovation Centre, focused on cloud.

Some of you are probably aware of the Windows Azure, that is the platform that we provide in an open way to the industry. A lot of people have already used the Azure, either as a service or to build their software application on top of that.

I have a video I'm going to skip, but if you look at computing, it's really very interesting.

In the first 30 years, there were many and the mainframe. In the 1980s, you get into the PC. It's more distributed. Right now, we getting into more of cloud and client with smart devices, with new natural user interfaces.

If you look at China, it's actually more dynamic and more exciting. I often see 21st century, there are two major phenomenon. One is the proliferation of IT and the other is the rise of China.

In the first 30 years, China was absent from all these advances. In the second 30 years, China has been an active participant. In the next 30 years, I think China will become one of the leaders, along with the rest of the world.

Remember, 10 years ago, Vint and I were giving a speech, people were asking about the internet, what is the future. And he said: the future is great, it's bright, you have not seen anything yet.

He's right. The future of the internet is bright and the next 10 years, you are going to see a lot more than the last 10 years. China, Hong Kong, Asia will certainly be a major part of that.

Thank you very much.


Lily Kung: Thank you, Dr Zhang.

Indeed, the internet community is all about participation. Take ICANN NomCom as an example. It is a process by which everyone, including all of you here, can get involved to help build. So we would like to invite Dr Gazak, from the ICANN NomCom, to the stage to say a few words, please.

Speaker: Good evening to all of you. I think we had a great presentation by Vint Cerf, as well as Dr Zhang.

I think both of them spoke about the immense power of technology. I remember Prof Harold Linstone from Portland once said: we are approaching 21st century technologies with 20th century governance processes and 18th century governance structures.

So that is a big gap we are really looking at, the technology racing to be met by governance models.

You all know that ICANN, quite silently in the last one decade, is one of the interesting studies for anybody in governance research a bottom-up approach, where a lot of participants or a particular firm which may have around 2 billion as its stakeholders.

This must be the largest company, in one sense, if you look in the world of the netizens, I would call it.

What does NomCom do? It invites and encourages you to become part of this governance process, great amount of diversity is required when we are really talking about global internet. So let me just have a few slides, as I said. Don't think it's a military slide it's more a voluntary slide asking you to know about we are doing at NomCom for leadership purposes, especially this year.

Every year, ICANN's nominating committee selects outstanding individuals as members of the ICANN board of directors, GNSO council, CCNSO council and the ALAC.

The 2011 NomCom invites statements of interest and candidate recommendations from the internet community for these key leadership positions. The deadlines for submissions is 4 April 2011 and selected candidates will take up the positions at the end of ICANN's annual leading in October 2011.

Essential information about the open positions, the requirements and the procedures are available at NomCom.icann.org website.

The open leadership positions currently available are two members of the board of directors, three-year terms each member, three members of that large advisory committee, currently one from Africa and the second one from Asia, Australia Pacific and Latin America, Caribbean Islands regions. These are two-year terms invited from these regions. Two Organisation is open. This is a two-year term and one member of the Council of the Country Code Name Supporting Organisation, which is a three-year term.

The deadline for submissions is 4 April 2011. Don't forget that. Please go do back and see.

The selection criteria is integrity, objectivity, intelligence, sound judgment, open mind and capacity for thoughtful group decision making, understanding of ICANN's mission and a commitment to the ICANN model of participative governance and familiarity with the aspects of internet and practice. So many people do ask me that this is all very high positions or two types of questions come that am I qualified? Please, you don't know. I believe that you are qualified to apply. Please do try.

The second they ask me is how much time I need to spend on this? I always tell to my students always ask me in the university, there's no time. I always tell, you always have time in the world to do things which you love and if you love the internet community and to be part of this governance, I'm sure time is not a constraint and many of us do that.

Willingness to serve as a volunteer, without compensation, I mean, monetary, but a freight compensation of satisfaction, other than reimbursement of certain expenses which is part of this service. Ability to work and communicate in English and NomCom seeks to reflect the internet community's cultural, gender and geographic diversity in its appointments. I have to tell that Asia Pacific region, if I understand, China and India put together is about 2.5 billion population, roughly out of the 7 billion population. The 7th billion, I believe, interestingly, is going to be born in India, the 7th billion citizen of the world.

So this is 2.5 and then you put across the remaining countries in Asia, it is certainly the diversity of people applying from Asia Pacific into various things, that's why I'm representing here.

And very importantly, it is a very well-known fact there is a very huge gender imbalance currently, in terms of the people applying. So we encourage women to apply more to these positions. So kindly go back and think that you would like to serve this, this is a great opportunity for you.

For more information, you can submit your application directly at this site or rather if you think somebody you want to nominate, you can always go, there is another place where you can suggest the names if you think NomCom and please do it before 4 April 2011 and thanks for this opportunity to present here.

Thank you.


Desiree Ho: A very big thank you to all the guests who have spoken with us today. Finally, I would like to wrap up with some of the housekeeping notes for your attention. Firstly, there will be a 10 minute coffee break before we begin the next plenary session here and that is the APOPS plenary IPv6 operations.

Secondly, this evening's social event will take place at Cafe Deco on the peak at 7 pm and buses will leave this building at 7.15 pm -- the first bus will leave at 6.15 pm and the last bus will leave at 7.15 pm. So please arrive punctually if you need transportation to the opening social.

The buses from the opening social back to Wan Chai will depart from The Peak at 9 pm to 10 pm. Thank you very much for your attention and I hope you will all have a very fruitful and enjoyable stay here in Hong Kong.

Thank you.