IGF 2020 - Day 7 - WS327 Believe it or not, the Internet Protocol is on Sale!

 

>> EDUARDO BARASAL MORALES: Thank you.  I'm going to share my screen.

Good morning, everyone.  Thank you, all, for coming to our workshop, Believe it or not, the Internet Protocol is on Sale.  We're very grateful to IGF for this opportunity to discuss the future of the internet.  I would like to thank the panelists for accepting our invitation and for supporting us in our research.  My name is Eduardo Barasal Morales and together with my colleagues, Tiago Jun Nakumura and Andrea Komo, we'll be the moderators for this workshop.

We are here to debate the impacts of the market for the internet.  Let me introduce our four panelists.  First we'll have two representatives from the technical community, then we'll have two private secretary viewers, Mr. Lee Howard from the United States and also India will be represented.

Let's start our introduction, remember, the basic, in order to access the internet, each equipment has to have an IP address.  Currently, we have two versions of IP, version 4, IPv4, exhaustion issues or IPv6, the newer version to develop to substitute the 4.  Although 6 has been around for more than 20 years, it is still variable according to measurements made by some internet companies, approximately 33% of the traffic is running over IPv6 because this is worse because IPv4 days are numbers.  There is a small number of numbers left in each region.  As you see, all of the regions, they're suffering from exhaustion.

In our region, our LACNIC region, this year, we ran out of the IPv4 addresses in August, this caused a huge increase in prices, last week, the LACNIC price on IPv4/23 was the same as IPv6/32, $600.  Now that LACNIC can no longer assign the address, the only way to get through those is through a broker that will charge you $12,000, an increase of almost 2,000%.  IPv6 has increased its price and it now costs over $2,000.

A high price, but still lower than IPv4.  However, the number of IPv4 transfers continues to increase and that's the reason why we want to discuss the IPv4 market we want to talk about the digital inclusion and the future of the internet.

Now let's start our questions and debate.  I'll kindly ask you all to type in your browser Slido and enter the code IP for sale all in upper case and we'll start our quiz.

>> ANDREA ERINA KOMO: Hello, everybody. 

We'll start with question number one, it proposes what is the correlation between IPv4 market and IPv6 adoption?  You have 2 minutes to answer this question, please.

>> EDUARDO BARASAL MORALES: Type in your browser Slido and then enter the code, IP for sale.

>> ANDREA ERINA KOMO: Now let's discuss this question, first with Mr. Moreiras.

>> MARCO HOGEWONING: Hello.

Interesting results.  It seems that we have quite a bit of IPv6 fans here.  There is this ‑‑ there seems to be an overall sort of lean to this is a false impression that you do not need IPv6.  I'm not so sure in that sense, by the time you enter the markets, by the time you pay that amount of money for IPv4, I kind of assume I would assume that you would look at the alternatives that you have seen.  You would think that people would reconsider that if they see that price tag coming with IPv4, at the same time, yes, it is still needed until we all, the world has moved on to IPv6, we need some IPv4 and that means that some of us, especially newcomers, need to find a way to obtain this.

I'm a bit surprised there is sort of still like a quarter of you saying they're still not correlated.  I would love to come back to that if we have time to discuss and why you think there isn't a connection.  In my mind, there certainly is unless we all move to IPv6, yes, we have to do IPv4 as Eduardo Barasal Morales explained, and the same goes with the IPv6 service region, yeah, we basically have no more IPv4 to give out.  If you need more addresses, then the way forward, you find someone with a surplus and have an agreement to use those addresses.  It is interesting.

I'm not surprised by sort of the responses that I have seen in front of me.

I'm not sure that people just sort of look at the IPv4 market, yeah, there is this market, I don't need IPv6 and there may be a few there, but they're naive and I think if you see the price tag, you should be aware that there are alternatives and it puzzles me and it has been puzzling me for years why people are paying so much money to continue with a protocol that we know will never fit.  The internet is much bigger than the 4 billion addresses we have in IPv4.  Yeah.  We should move.

Now, I see a lot of familiar faces here in the chat.  I know many of them have already moved.  The big question on the table, I think also for IGF in general, it is what can we do to help and move the others.  Again, I'm happy to leave room for discussion and get feedback from the audience on what you think is sort of the key area and what we can do to unlock it apart from working until the ‑‑ it is ‑‑ I leave it here as an opening remark.  I leave time for some discussion with the participants.

>> ANDREA ERINA KOMO: Thank you, Marco Hogewoning.

Next, please, Mr. Antonio Marcos Moreiras.

>> ANTONIO MARCOS MOREIRAS: Looking on one side, I see that the market, there's certain kinds of reasons to get the IPv4, they need to exist and to operate to date.

On the other side, we see IPv6 adoption, it is still very low.  For instance, in commerce, for instance in government and other kinds of reasons as well.  We see some kind of business such as ISPs that connect to the end users, transferring IPv4 and don't use, don't deploying IPv6, this is very worrying.  For instance, we have the data from LACNIC, we have had 25 transfers, and just 7 of the companies are deploying IPv6, just 7 of the 22 transfers, they have IPv6 together with it.  32%, I think it is very, very worrying.  It's not ‑‑ I can't not say it for certain, but since having an IP in the market, it is jeopardizing the IPv6, companies are not deploying IPv6 because they have the possibility of still getting IPv4 for price.  This is my view, that the market, it is not good for IPv6 adoption.

>> ANDREA ERINA KOMO: Thank you.

Next, please, Mr. Howard.

>> LEE HOWARD: Thank you.

I'm glad to be here.

So I did some preparation for this, I wanted to see what kind of correlation there is between the address market and rate of IPv6 adoption.  It is not an advertisement, I did a blog post, I have some charts that you can go see.  What I found, it was the rate of acquisitions by IPv4 addresses on the internet, it is linear, it is a straight line and the rate of IPv6 adoption at this point, it is very linear, a straight line and the two lines, they're almost perfectly parallel which says to me, there is none of the answers, it is not exactly correct, they are perfectly correlated, they are perfectly correlated at exactly ‑‑ this is my perspective, my opinion.  It is that they're perfectly correlated at exactly the rate of growth of the internet.

As organizations are bringing new users or devices online, they're having to adopt either IPv4 or IPv6 or both.  I have done multiple presentations over the years, including at Nanog and LACNIC talking about the price of IPv4 addresses and comparing that to the price of deploying network address translation and the cost of IPv6 and what I have found is that at a certain point it becomes ‑‑ it is cheap tore buy the IPv4 addresses than to deploy others.  In the market mostly, what we're see, it is that IPv6 is another way to reduce the cost of Nat and we have seen several companies using it that way, you need IPv4 addresses because the internet is not completely IPv6 yet but you want IPv6 so you have fewer IPv4 addresses for outside of your net.  That's what I'm seeing so far.  I'm sure we'll have a lot more to talk about.

Thank you.

>> ANDREA ERINA KOMO: Next, we have Mr. Rajesh Chharia.

>> RAJESH CHHARIA: Thank you.

It is a very good question, a good answer.  What I find, helping the adoption of IPv6 and IPv4 market will jeopardize the market of IPv6, what other are the two parts of the coin.  In one way, no doubt IPv4 market will be helping the start‑up to start the network but at the same time, if the market is there, it will not encourage them to adopt IPv6 from the initial time.  What they find, the IPv4 deployment is very easy version easy to remember and IPv6 is very very tough to remember and you have to upgrade your hardware, equipment a lot.  That's the reason why this difference is here.

In India, what I'm finding, the IPv6 deployment, it is the highest in India if you compare with the world and the network operator, the large network operator, even the small network operator, they're ready to deploy the IPv6 and providing IPv6 to customers.  At the same time, we have to understand one thing, in point of India, in India, the lifetime of the equipment is very long.  We are using the equipment until the time that the equipment says sorry, throw me away, my retearing costs would be much more higher than the new equipment and then only are we changing the equipment and that's on the rural side.  India consists of around 65% of the world and the equipment that's getting obsoleted in the city, it is traveling back to the rural side, that's the reason why you find that the deployment of IPv6 in the rural area is small in comparison to the metro city.  IPv4 market is there, it helps us also to make the difficulty of deploying the IPv6 in a proper way.

>> ANDREA ERINA KOMO: We'll go to the next question.  Thank you.

What is the impact of IPv4's markets to the distribution of IPv4 addresses.

To discuss, please.

>> MARCO HOGEWONING: Not a big surprise.  The audience speaks for itself.  I would sort of have a contrary view from a technical perspective, it is good to see that resources that basically lay there on the shelf eating dust get used.  There are still, as we discussed previously, there are still regions to do IPv4.  We don't have to.  As long as not everybody is on board with IPv6 it is good.  Yeah.  Obviously the price tag that you see, the prices that you are floating around for the resources, they are high and they're only getting higher and in that sense, yeah, it certainly is for the very few that can afford it.  In that sense again, you know, if we're all about sort of affordability for all of us, inclusivity, the sooner we go to IPv6 the better it is.  It is not free.  I mean, it is probably a lot more affordable than continuing this weird spiral upwards in market and certain brokers.  Forgive me for saying that.  Yeah.  If we all want the world to enjoy the internet then we should just move to IPv6.  That's the only way to make that happen.

ANDREA ERINA KOMO: Thank you.

Next, Mr. Antonio Marcos Moreiras.

>> ANTONIO MARCOS MOREIRAS: This is welcome, giving us flexibility to the current situation, together with the recovery process that are made ‑‑ that are made from EIRs.  The market depends on the economic power to buy the address, not just the need of the others.  Maybe things don't come together necessarily.  For instance, in the for organizations such as community networks, new community networks, maybe they could not come to exist, come to operate without new other, new IPv4 for others, it is not okay to have just a IPv6 only network, it is almost impossible.  They need IPv4 for others, they would be able to buy it in the market, I'm not sure, in a small way ‑‑ for example, in Brazil, we have 80,000 and 50,000 autonomies systems but we have 50,000 ISPs.  Only half of the ISPs are not autonomies systems.  There are ISPs operating in a small geographic region or in some cities, and they don't have IP and they don't know how to get them from NI Ryan Nugent‑Hopkins, from IRR, so they would be able to have that in the market, probably not because they are very small.  They don't have economic research to do it.

The transfers, they're a very good thing.  They have flexibility, but the market, I'm not sure it is the best way to solve the current situation and we should foster IPv6 adoption and make it faster better, but I'm not sure if the market will help with that, will help with the distribution of not using address and things like that.

>> ANDREA ERINA KOMO: Thank you.

Next, please.

>> LEE HOWARD: It is a really interesting question.  I want to start by responding to a couple of things that Antonio just said.  You said IPv6 is not possible to run a network at ISP at IPv6 only, which is mostly true but for mobile companies, several mobile companies are IPv6 only inside of their networks, they only get those addresses and they do the nat64 or other technology on the edge when the users need to get to something with IPv4.  It is possible.  Is also Facebook, it is famously IPv6 only.  Everything in their network runs IPv6 only and only have IPv4 in translation on the edge.  It is possible for a data company A contact come to run IPv6 only.  That said, bough of those, all of the operators I know of running IPv6 only have also had to buy IPv4 addresses at some point so they still needed addresses to grow large enough to reduce the need there and that's dependent on the rest of the internet ecosystem, dependent on the content providers, the users they're trying to reach or in Facebook, the users trying to reach them for the IPv6 for the IPv6 investment to be worthwhile..

A bit of a joke, you said we should make IPv6 faster and it is faster, it is lower latency than IPv4, we see that on the internet, that's for another panel.

For this particular question, what strikes me, the place to look for information for how to really answer it, it is the policies, so transfers, for a long time, LACNIC was ‑‑ for a long time, it was not allowing transfers of IPv4 addresses outside of the region and that's partly related to other policies outside of the region and that changed a few months ago and we have begun to see addresses transferring in and out of the organization and we have seen an increase in the LACNIC region to buy addresses, it is more expensive than before.  That's a downside and we have seen prices increasing a lot over the course of this year.  So that's making it more and more difficult.  Economists would say, well, it is not, how deep someone's pockets are in terms of how they're able to pay for address, but it is even somebody with billions of dollars, they don't want to spend hundreds of dollars on something they don't need, but economists say those spending more for addresses will put them to a better, higher use, I don't entirely believe that.  They think there is a need for accredits while people deploy IPv6, it is still as an IPv6 advocate, it is still easy to say that the way to reduce the cost for IPv4 is to deploy more IPv6 and to tell friends and colleagues that they need to deploy IPv6.

That's enough for now.  More for the next question.

ANDREA ERINA KOMO: Next, Mr. Rajesh Chharia, please.

>> RAJESH CHHARIA: The IPv4 market is making the situation, they have the financial muscle, it will enter into this market.  I feel that the start‑ups, the new intranet, they're finding it very difficult because if they start investing a lot of money into the purchasing of the IPv4, the IPv4 is necessary for the network, so the whole idea is going into that, so one thing, what I feel, it is very ‑‑ I'm not appreciating that, initially, the historical allotment of IPv4 addresses, they were allocated to some of the research companies, to some of the educational institutes, and to some of the organization NGOs and things, and at that time, because the IPv4 at that time had no value and the location, it was there, and they were taking the location data and now after this market, those NGOs, the educational institutes, they're the research organizations, they have taken this IPv4 address for the purpose of no start date, sending that into the market, into the market price and I'm not feeling good for this.  This is not at all what we thought when these were allocated to these researches or the educational institutes.  In my opinion, if we have any I have addresses for any of this, and if the research, if it is completed, then those IPs should come back to the IRR or come back so that the new intranet or the start‑up, they should be getting that IPv4 address for running their network properly..

14.6 million IPv4 addresses has come into the market through the tender process.  That's very long.  In the tender processes, those blocks are so big, someone will not be able to participate in that tender because that sender requires a huge muscle power to participate in that tender and that makes the initial investor in the network a problem for them to even pay for that only the big company, the big organizations, with plenty of dollars, they have the deep pockets, they have already cornered those IP and for the general user, for the general network operator, ESP, for the late intranet into the internet countries, the late entrants, they're finding it very difficult.  That's my thinking that the IPv4 market is only for the financial muscle power, nobody, general public, general network operator, they cannot use this IPv4.  Thank you.

>> ANDREA ERINA KOMO: Let's go to the next question..

What is the correlation between I have market and digital inclusion.  One minute for responding.  Let's discuss this.

>> MARCO HOGEWONING: Yes.  It is ‑‑ the question, it is as we have seen before, it is interesting that there seems to be a bit of a race going on between different answers.  As we have heard from previous speakers and we see ourself, of course, yeah, building Ned work, it is expensive.  Building networks in this day and age that has to support IPv4, it will make it only more expensive.  As pointed out, we have seen networks that are only IPv6, a small little block at the edge to keep connected to the IPv4 role, it is enough and a great way forward.  Yes.  As long as people have the ability to sustain that only operation, they're not helping, they're not helping us so I think in that sense, yeah, it is probably ‑‑ I wouldn't say there isn't very strict, yeah, the I have market, it is breaking inclusion, it is helping and in that sense, back to my earlier comment, it is what it is, there are still a large chunk of the internet and you earlier on mentioned also governments, that do not do IPv6 so no matter how you put it, you have to find a way to connect to that old‑fashion world of IPv4 and that means that you have to somehow obtain ownership, whether it ‑‑ I greatly respect the comment, yeah, of course, we're more than happy to take addresses back and use our waiting list distribution center to distribute them in a fair manner by the policies determined by the multistakeholder process.  Yeah.  Market, it is a tough one to crack.  You know.  You can hand them back to me, you can hand them to somebody who is handing you a bag full of dollars.  The choice is probably for a lot of people fairly simple.

It is not great that IPv4 is still around.  Let me put it that way if you look for digital inclusion.  Again, if you're the one to move forward, and I saw an attendee pointing it out in chat, there is enough barriers and obstacles to get everybody on the internet already.  We don't need another one.  Let's move to IPv6.

>> ANDREA ERINA KOMO: Thank you.  Next is Antonio Marcos Moreiras.

>> ANTONIO MARCOS MOREIRAS: Very well.

I think in the current situation, the market can be the only viable way to make new SPs or other internet‑related business, new community networks viable.

IPv4 is needed.

What Mr. Howard told about the networks being IPv6 only almost totally, just having some IPv4 on the edge, for the map, not for the 64, it is true.

There is still a need for this IPv4 for others to do that.  We have seen that most games don't do IPv6, most governments don't do IPv6, yet it is almost impossible to operate in a IPv6‑only environment, at least without some IPv4.  Yeah.  We see that ‑‑ we see that IPv4, the markets hate and the IPv6 hate and the internet growth, it is all the same, it is worrying.  The IPv6 growing hate should be bigger than the internet, it is the same, it seems that it is not improving, the percent of deploying.  Even if at some time should be IPv6 only and the way it is going, it will never be.

We see there's an important role in the digital inclusion in Brazil and the business, it is not viable any more, because of the block of IPv4.  Maybe the market will be better than before.  In my opinion, it was lost in the IPv6 adoption and it is a barrier for them.

It is also big for community networks.

I think the core relation is not strong, but there is a correlation, could be a correlation between the IPv6 format and digital inclusion and this format, maybe it can reduce the digital inclusion rate of growth.

Thank you.

>> ANDREA ERINA KOMO: Next, please, Mr. Howard.

>> LEE HOWARD: I agree with everything that Antonio just said.

The small ISP, community ISP, it is where I think much of the growth of the internet is coming from, where much of the additional inclusion of new users on the internet is coming from.  Being new, being in developing areas, they tend to be people and organizations with a smaller budget, just less money to spend on building the network and that's complicated.  I feel that the people that I have talked to, worked with that are working on these kinds of network, they're very smart people and they understand the trade‑off that they have to make and what decisions they can make.

I don't think that the problem is ‑‑ I think this is consistent with what Antonio just said, that the problem is not that the ISPs are unwilling to deploy IPv6 or unwilling to deploy both IPv4 and IPv6 and doing translation from private IPv4 to public IPv4 or from public or private IPv6.  The more of the problem is in the consumer electronic market that the wi‑fi router that may be someone's home, a wi‑fi access point in an area, it doesn't support IPv6 or a transition technology.  The consumer electronic, the smart TV, the refrigerator, the home security devices, most of those don't support IPv6 and until they do, the consumer can't live on IPv6 only.  We need a transition technology to get them there and so ‑‑ so the consumer market has not yet caught up to the ability to do IPv6 only and that's preventing a truly equal market for newcomers to the internet.

There is also ‑‑ even that, a better solution than it was a long time ago.  Translating from one protocol to another, or even from IPv4 public to private to IPv4 public, it is not a effecter experience.  It is a degraded internet experience.  Not everything works as well when you do that.  It works pretty well.  Basic web access generally works.

Things like gaming, certain kinds of innovation on the internet don't work as well and those are things where IPv6 can provide the possibility that sorry, I was looking at the comment on the panel, that his TV does support IPv6!  That's great!.

Innovation on the internet, if you don't have a true IP address from which you could do network innovation then you have a lesser experience than elsewhere and I think that's potentially a very bad thing for the internet as a whole and therefore for the global community.

>> ANDREA ERINA KOMO:  Next, Rajesh Chharia, please.

>> RAJESH CHHARIA: Thank you.

The question gives the answer.  This is the very good thing.  As I just go through a lot of international conferences, ICANN, APNIC, IGF, what I find, generally the discussion moves around the developed countries only.  Why discussing about any issue, and the discussion, the developed country, that's very predicting.  We also have to see the under developed country or the developing country also because the underdeveloped country, the developing country, it doesn't have the big pocket like the developing country.  They have only matured, they are in the way of getting maturer.  Here is the difference.

If I take the example of India, 65% population, they have a very big concern about the internet plan.  When the mobile has started in India, at that time, most of the calls to the website were by the missed call, the children of the rural people to the major city, the mother, the father, they give a miscall to the son and daughter and they ‑‑ they call, he now after almost the call is free but still this miscall process is going on.  The major concern is, the cost.

If IPv4 market, if any service provider, one more example, if the service provider, the big telcos are so serious on the penetration of the broadband in rural India, why still the rural India doesn't have more than 20% penetration and there is no business case to lay the networks over there and to give the services to the real people.

For the real people, the smaller ISP has to go over there and to give the services and the smaller ISP forgiving the services, they have to buy the IP report address from the local market and doesn't have the deep pocket and they will not be able to buy the IPv4 address.  What he will do, start doing, tell start doing the netting.  Now, the security doesn't allow that netting in I wanted I can't, for the security reason, because India, it is very badly hidden by the terrorism.  If I have to run the network, I have to buy the address.  That becomes a bigger thing rather than buying a router, rather than buying a server, rather than buying any other ‑‑ first the investor and the start‑up, they have to think that I have to buy IPv4 on the market because they will be getting only/24, or the 26 only, not more than that.

In/24 you cannot run a network.  That's the reason.  IPv4 market, it is just curtailing the digital inclusion because of the cost of the internet plan will be high, the cost of network will be high.

The cost will be high.  The user will not be very happy to take that as a ‑‑ you can say digital revolution in the country.

Thank you.

>> ANDREA ERINA KOMO: Thank you.  That's good.

Now let's go to the last question for a minute please.

The last question proposes how can internet governance be affected by the IPv4 market, taking into account the role of each stakeholder in the scenario..

Mr. Marco Hogewoning, please.

>> MARCO HOGEWONING: I would turn it around.  Think about what we hear with each of the stakeholders in respective roles, what they can do to affect this market?  We spend a lot of time talking about inclusion and talking about essentially the short vision of IP before and the prices that are currently on the free market for additional IPv4 and the rest forms a barrier of entry to people, to companies, and I think it is in this sense being with the IGF, it is a great place to discuss what our collective responsibility is to stop that from happening.  The internet governance will be effected by the IPv6 market I don't think as much, the only real impact there, yes, in trying to progress inclusion, trying to get people collected and connected to the interrupt, this is turning into a bigger problem.  I think in that sense, you turn it around.  What can we, as internet governance community do to make sure that this stops and that this sort of becomes a more fair, more equal scenario and that's probably IPv6.

>> Andrea:  Thank you.  Next, please, Mr. Antonio Marcos Moreiras..

>> ANTONIO MARCOS MOREIRAS: If IPs are now so ‑‑ they are good instead of being a resource, a resource that has to be very well managed.  This is the situation, if IPs are now kind of good, what the ARAs, the NIRs, what they were doing, all these years, it was essentially wrong.  They should not keep doing that with IPv6, should IPv6 also be managed since the beginning, I'm not sure.

I think that the market brings some questions about how we have managed IPs in the last year.  We should maybe talk about it.

I'm not sure that the market gets in the way of IPv6 but I'm also not sure if it is helping it, if it is helping the internet future and present.  I'm not sure.  I see there are very good points in the existence of the market.  It allows some business to exist and to operate today.  Maybe there is not good effect for the future of the internet.

I'm really not sure.

Maybe the question should be what should we do to help IPv6 adoption and helps the internet to go forward.

Thank you.

>> ANDREA ERINA KOMO: Thank you.

Next, Mr. Howard.

>> LEE HOWARD: It is a very large question.  The internet governance as we discovered through years of IGF, it is ‑‑ it is a multistakeholder but essentially every stakeholder.  It is everybody that was on the internet, everybody who wants to be on the internet.  I think I want to give credit to ‑‑ there is a model at NIC.br, they set a model for deploying IPv6 in a country, we have heard from people there today, that yeah, we wish it was more.  The training and support provided by NIC.br is good and strong.  I look at countries where IPv6 deployment is high.  Belgium, a country with high IPv6 deployment.  That's because the government was ready to ban NAT, we need to be able to ‑‑ I think it was what was said, we need to identify a user in case of terrorism or something by their IP address and ISP said we can't bend that, we're ‑‑ we'll go out of business and citizens will pay significantly more for internet access.  They limited.  You can only have 10, 16 users per IP address, that made it, that gave all of the ISPs incentive to deploy that without telling them how to ‑‑ or to buy addresses.  They had their choices.  Most chose to deploy IPv6.

I wrote a plan for a developing island nation several years ago.  I noticed that the government in that nation was hosting most of the government webpages offshore, actually in the U.S. and so for citizens to access government services they had to reach a data center that was in Los Angeles instead of in the island.  I knew there were connectivity problems where the entrenched market, the networks didn't connect to each other and I proposed that by simply bringing the web hosting on to their nation, into their nation, on to their island, and requiring and providing preferential treatment for those who hosted IPv6 and providing preferential treatment for buying internet access services to the ISP would provide them IPv6, they may build a long way to deploying the IPv6 and more importantly, toward equalizing internet access in the country, so that citizens are able to access the government services so much faster because it is local and so much faster because they would be getting IPv6, they ‑‑ they ended up ‑‑ there was an administration change and the government ended up not doing anything.

There ‑‑ they're still reaching government services a long way away.

Part of it may be because of the cost of hosting locally, it may be just too high.

I don't think governments should enter the market themselves.  I don't think that governments in every country should just go buy a large block of addresses and allocate them locally.  That's another option that's sort of ‑‑ it is the expensive option but certainly distributes the costs of addresses among the entire nation, not just among the new entrants to the market, to the internet.  There's a lot more things that can be done.  I think that IPv6 is probably not the solution but based on the growth of both, we'll be watching and seeing both protocols running for a long time.  I think we can see that IPv6 is a more equal protocol, that it provides a true substitution in the nearer future.

>> Andrea:  Thank you.  Next.

>> RAJESH CHHARIA: Internet governance, multistakeholder, words we discuss in the IGF, what I have observed, since the beginning I have attended all of the IGFs, these words are very common in this arena and at the moment we're going out of the IGF venue and we're in business and we have forfeited everything.  What's the main objective of the IGF, access.  How the access is possible if the IP addresses are not there and how the access will be possible.  Hence, IPv4 market, it is effecting IGF very strongly.  If we really want, IPv6 should be deployed and IPv6 should be successful then we have to take a call and the government has to take a call with all of the financial institutions including bank, airlines, government website, social networking site and all of this marketplaces should be shifted to the IPv6.  When these sites, when they're going to the IPv6 automatically user demand IPv6 from the network, then the network will also be deploying in a very serious way for the user, otherwise until the time the demand will not come from the user side you will not be able to deploy the IPv6 the way that we want.  Actually, when we're talking into the IGF, in the IGF, we should come out with a solution that all of these governments should be something mandated, otherwise they won't able to deploy out of the deep pocket organizations, and they'll still run into the IPv4 and that will make all of the smaller operators also to buy it from the market which is not good.  In my opinion, the IPv4 market, it is very much effecting the internet governance and they should think about that.

Thank you.

>> TIAGO JUN NAKAMURA:  I thank you for coming to this session.