Hi Andrew,

Your apparent passion and commitment to the migration reminds me of Plato's Cave :) You have clearly seen the promised land of a native v6 world and it is beautiful.

Now, am hardly poet but am forced to say the above in appreciation of the amount of work that you have clearly put into this. You have also raised some very high level points that I will need some more time to fully research on. I do not have the answers at the moment.

First, I agree that the proposal of having a reserve block for new entrants is the most rational way forward. Still, some of the reasons why I would consider stopping the allocation of v4 are below.

1. These v4 IPs will eventually be exhausted anyway and the final goal is to stop using the current addressing scheme completely

2. Every single v4 IP that is currently/ to be allocated will have to be swapped eventually - more time and resources will be required

3. To expand on the above, this challenge grows exponentially with every v4-only device that is sold (and this mostly happens in developing countries)

4. There are RIRs that have long stopped v4 allocations (certainly not by choice) and I believe there are still new entrants into those markets and they still find a way to survive. Let us not underestimate human nature. A good point in case is that the exhaustion of v4 has been imminent for a long long long time.

5. For consumers( I include users, developers, Systems/IP guys - all of us), the costs also include the lost opportunities for better connections ( logical and social) and I can also add that we are also restricting innovation.

Consider the below especially as regards to 4 & 5. ( I do not necessarily agree with the simplistic opinions given by the author re ISPs as I know we all have different values and priorities - but then again I did not want to create any perception as being subjective on what content to highlight from the article, being in the industry myself)

"A lot of CGNs are being deployed. Some recent work I have been doing shows that upward of 3% of IPv4 users present on a different source IPv4 address within 10 seconds—i.e. a minimum of 3% of users lie behind CGNs with relatively aggressive address lease timeouts. Secondly, a lot of shared Web hosting is being deployed. It is evidently commonplace to see upward of 10,000 Web host environments co-existing on a single host IP server address."

ISPs are playing their cards close to the vest, but it looks like many of the ones that are planning to start rolling out IPv6 soon will be deploying IPv6 along with CGN-based IPv4 for new users. They're reluctant to change anything for existing users, because the first rule of being an ISP is "don't generate support calls." Providing broadband Internet access is a very profitable business, but the profit generated by a customer evaporates faster than you can say "have you tried rebooting your router?" when said customer calls for support. ISPs that started deploying IPv6 in past years had access to enough IPv4 addresses to give users their own along with a range of IPv6 addresses. That is no longer true, or will no longer be true as soon as ISPs use up their own stashes of IPv4 addresses.

The downside of NAT is that it only works well in one direction: from the inside to the outside. When connections must be set up from the outside to the inside, such as in the case of peer-to-peer audio or video conferencing, additional logic is necessary to find a way to the right internal system through the NAT. This is bad enough when two users are both behind their own home NATs so that two NATs must be bypassed, but it gets much worse as ISPs deploy CGNs, so now four NATs can be in the path. CGNs also can't open up ports as easily as home NATs. As long as it's not firewalled too severely, IPv6 has none of these issues; with 2^128 addresses there is no need for NAT. So it makes sense for ISPs to deploy IPv6 along with CGN-based IPv4. However, there are still ISPs that pooh-pooh IPv6. Huston again: "The pessimistic view is that so far nothing much has broken in IPv4-land, so there is still some more time left to do nothing!"

Unless the ISPs that have been ignoring IPv6 plan to just keep their existing customers and not sign up any new ones, those ISPs are still going to be bitten by the IPv4 address exhaustion and will almost certainly be forced to deploy CGN at some point. With no pressure relief valve in the form of IPv6, all user traffic will have to flow through the CGN, which can then easily become a bottleneck and a single point of failure. As a result, the quality of service delivered by different ISPs will diverge more and more, with the ones providing unshared public IPv4 addresses as well as IPv6 doing the best and the ones using CGNs with relatively many users per public IPv4 address and no IPv6 doing the worst.

The good news is that so far, the Internet has always managed to adapt just before collapse was imminent. In the late 1980s, TCP congestion control saved the Internet from massive congestion. In the 1990s, classless interdomain routing and route flap damping kept the routers going. This time we only have to turn on a feature that's been in our operating systems for a decade and maybe replace an aging modem or two. Call me an optimist, but I think it can be done. But only at the very last moment, of course.

(Source)

So yes v6 will happen with/out any drastic measures.

On the other hand, Africa finally has a chance to make a bold message that I can consider to be at the scale of the burning of Ivory in Kenya.

Before, stopping all allocations was impossible as the RIRs would never have been able to convince their customers or even each other to do so. But now that we are the only ones left with a substantive reserve of these (potentially toxic) IPv4 resources, we could agree to start moving in the other direction and let all the others follow this time. I would really be pained to hear the inevitable statements (with very correct language of course) that the reason v4 still exists is to accommodate the developing world.

Regards,

Kevin K.

+254720789158

On 13 October 2016 at 10:00, Andrew Alston <Andrew.Alston@liquidtelecom.com> wrote:

Hi Kevin,

Just another further note on what I said below.

To quote from Dave Michaud from Rogers Communications yesterday on another list (used with kind permission of the author who waived Chatham house rules that exist on the list it was posted to allow me to quote this):

Once we are done with this phase, we will start migrating some phones to IPv6-only operation. The transition will be executed per phone model by a logic implemented in a AAA server that is queried by the PGW when the PDP/PDN is established. The AAA decides based on IMEI if the phone should connect to the network using dual-stack or using IPv6-only with DNS64/CGN64.

He further sent me the following:

To clarify the v6-only service, we will be cherry picking devices for this service. Initially, it will be Samsung Galaxy S4, LG G4 and Nexus 5. After that, once we are comfortable, we will move more devices to IPv6-only operation (likely all newer Samsung, LG and Google/Nexus/Pixel devices). Moving forward, all new devices that we start selling will also be launched with IPv6-only.

Now – what does this mean. Firstly, the days of dual-stack as the only means are over. There *IS* a move towards single-stack v6 with only translation to v4. I believe (but am open to correction) that t-mobile is doing the same thing. Secondly – it emphasises the point of reserving a small block of space for new-comers that don’t have in order to facilitate the v6 -> v4 translation referred to above. Thirdly – the implications of the above are… quite mind blowing – because while NAT has certain ways to transverse back through it (UPNP etc) to enable certain applications to work ok behind standard v4 NAT, I am far from convinced the same will work with v6 to v4 NAT in the same manner. (Open to correction and would love to hear from someone more knowledgeable in this area). But what this means is – if people aren’t running v6 and pretty fast – life on the Internet is going to start getting really interesting – because v6 single stack isn’t a thing for tomorrow – it’s already happening.

Thanks

Andrew

From: Andrew Alston [mailto:Andrew.Alston@liquidtelecom.com]
Sent: 13 October 2016 09:45
To: Kevin Kamonye <kevin.kamonye@gmail.com>
Cc: General Discussions of AFRINIC <community-discuss@afrinic.net>; KICTAnet ICT Policy Discussions <kictanet@lists.kictanet.or.ke>; Barrack Otieno <otieno.barrack@gmail.com>

Subject: Re: [Community-Discuss] IPv6 Chapter 254

Hi Kevin,

I don’t think completely stopping v4 allocations right now would have the impact we’re looking for. If you look at the policy proposal I’ve put forward, I propose a /13 reserved entirely for new comers, people who had zero space from anywhere before this. I think this is still an adequate number and sufficient, but it is also critical. The reason for this is that it allows sufficient space for entities to do NAT64 / DNS64 for translation to legacy equipment in a single-stack v6 environment. I still believe this will be necessary for a few years to come – and I think the /13 reservation is sufficient.

I also think that at the rate of depletion – we won’t actually be gaining much time by stopping, and as you say, there are other considerations we have to keep in mind. Rather than focusing on the financial considerations, we have to consider the fact that the space that was given to AfriNIC by IANA was meant to the serve the people, and I’m pretty sure that if AfriNIC decided to just stop allocating and hold onto all of it they would run foul of the agreements under which they were given that space. (I could be wrong here, perhaps someone with more insight can comment).

What I’d like to see is a situation where those who need the v4 space today, for use on the continent, can get it, use it, and we deplete naturally. There is a lot of evidence that there is plenty of demand on the continent, and while some would say that large allocations indicate space flowing off the continent, I have yet to see any concrete evidence of this and in fact the allocation statistics seem to dispute this fact. (The majority of the really large allocations in recent months looking at the publically available data are tending to go to African countries that traditionally had far less space than other places, and an analysis of the BGP surrounding those allocations gives no indication that they have been moved off continent, though of course I say that BGP analysis and latency analysis of space to determine actual geographic location is a bit hit and miss and far from an exact science).

If we repeal the current soft landing policy and maintain a limited reservation strictly for new-comers (and I do believe a /13 is sufficient), this will achieve the necessary in my opinion. It will ensure the rapid depletion of v4 space on the continent, it will ensure that the space that is currently within AfriNIC is actually used for proper benefit, it will ensure that there is still space available for people who have absolutely none to use for single-stack v6 to v4 translation as necessary, and all in all, I believe that’s the best solution.

Thanks

Andrew

From: Kevin Kamonye [mailto:kevin.kamonye@gmail.com]
Sent: 12 October 2016 18:36
To: Andrew Alston <Andrew.Alston@liquidtelecom.com>
Cc: Mark Tinka <mark.tinka@seacom.mu>; KICTAnet ICT Policy Discussions <kictanet@lists.kictanet.or.ke>; General Discussions of AFRINIC <community-discuss@afrinic.net>; Barrack Otieno <otieno.barrack@gmail.com>
Subject: Re: [Community-Discuss] IPv6 Chapter 254

Hi Andrew,

Solid points all round.

I had really not grasped it properly before, but I can now see how the concept of actually encouraging the rapid exhaustion of v4 would certainly be a game changer.

To take it further, would you say that STOPPING the allocation of v4 starting NOW would have more impact? Of course this would have several downsides that would need to be mitigated. For instance, I can see that this would translate into financial challenges for Afrinic as they do rely (not sure about this) on the revenue from the sale of IPs to fund their operations. No one likes to lose money, not even a non-profit :)

I would really like to hear your thoughts on this.

Hi Mark, very true. v6 on mobile should be pretty much done by now. Also, I can already hear that the other big service providers are starting to stir due to this challenge from Liquid. Perhaps it will even turn into a race that makes us all winners.

@ Barrack - cheers mate.

Regards,

Kevin K.

+254720789158

On 12 October 2016 at 16:22, Andrew Alston <Andrew.Alston@liquidtelecom.com> wrote:

Hi Mark,

In the mobile space (LTE), and in the wireless space – while I can’t comment on specifics, watch this space.

In particular in KE and ZM dependent on which technology you’re referring to.

Thanks

Andrew

From: Mark Tinka [mailto:mark.tinka@seacom.mu]
Sent: 12 October 2016 15:55
To: Andrew Alston <Andrew.Alston@liquidtelecom.com>; Kevin Kamonye <kevin.kamonye@gmail.com>; KICTAnet ICT Policy Discussions <kictanet@lists.kictanet.or.ke>
Cc: General Discussions of AFRINIC <community-discuss@afrinic.net>; Barrack Otieno <otieno.barrack@gmail.com>
Subject: Re: [Community-Discuss] IPv6 Chapter 254

On 12/Oct/16 13:31, Andrew Alston wrote:

On this map, you will see there are only two countries in Africa that have in excess of half a percent v6 penetration levels. One is Sudan, and one in Zimbabwe. Zimbabwe currently runs at 4.76% penetration and climbing – beyond that the rest of Africa has effectively no real penetration. Now, compare that to the rest of the world where v4 is depleted, and you see a vastly different picture. The global average deployment rate is sitting at 12% and climbing, whereas all it took to *double* the aggregate penetration rate in Africa was the v6 enabling of 10 or 15 thousand FTTH users in Zimbabwe. This speaks volumes, we have v4, and its slowing us down in getting v6 deployed.

Given that consumers don't generally get a say in when IPv6 can be enabled, that helps a lot. Much of Europe, North America and Asia-Pac have sufficient broadband into people's homes that makes all the difference.

A number of major mobile operators in that part of the world have also turned on IPv6.

The majority of Internet access in Africa happens in the mobile space today. If we want to see the needle shift even a hair's width, mobile operators in Africa need to enable IPv6. As of today, I have neither seen nor heard of any plans from any major or small mobile network operator in Africa re: turning on IPv6, never mind have a strategy or plan.

If wire-line and non-GSM wireless service providers in Africa were to enable IPv6 for their broadband customers, there would be an improvement in the outlook (by your own experience in Zimbabwe), but not as much as if the mobile operators came to the party. It is absurd that there is no interest from this group, considering that the thinking is that it is cheaper to spend millions of $$ to sustain NAT444444444 than it is to roll out IPv6.

Mark.