[rpd] [Community-Discuss] Cloud Innovation Displays Very Poor, If Not Criminal, Netizenship

[Owen DeLong]

The opinions below are mine and mine alone. They are not official positions of any organization I may be affiliated with…(Several of which probably prefer I not post them)


> On May 31, 2020, at 14:37 , Noah <noah at neo.co.tz> wrote:

> 

> 

> 

> On Sun, May 31, 2020 at 10:02 PM Owen DeLong <owen at delong.com <mailto:owen at delong.com>> wrote:

> 

> 

>> On May 31, 2020, at 11:40 , Noah <noah at neo.co.tz <mailto:noah at neo.co.tz>> wrote:

> 

> 

> See inline....

>  

>> We were not faced with a choice between “Implement transfer markets or not.” We were faced with a choice of “Recognize transfer markets and

>> regulate them or ensure that they are black markets and that the RIR system and its IPv4 policies become irrelevant to the actual operation of

>> the internet.”

>> 

>> And black markets still exist nonetheless (a failure of responsible audits and accountability) irrespective of existing transfer markets imho.

> 

> Most of the places where black markets exist with any significance are places with unreasonably restrictive transfer policies. It’s not so much a failure of responsible audits and

> accountability as a failure of policy to adapt to reality.

> 

> The AfriNIC region has an IPv4 transfer within the AfriNIC region policy [link 1]  where under section 5.7 there are provisions that permit transfers of IPv4 address space within Africa, therefore one would not claim that there are unreasonable restrictions. 


This depends on your definition of reasonable restrictions, I suppose.


> But if responsible audits were to be attempted to ensure compliance and accountability that would go to resolve the apparent black market.


Would it? I can think of several ways to legitimately transfer the use of addresses without necessarily transferring or altering the registration at AfriNIC, none of which would be a violation of AfriNIC policies as currently written.

My definition of a black market is any effective transfer which is not recorded in the RIR database.

Perhaps your definition is any effective transfer which violates policy.

Those are very different definitions. The former will not be prevented or addressed in any effective way through auditing. The latter is a much narrower definition, and would only catch people incapable of minimal creative thought.


> People are seeing the current overly restrictive transfer policy in the AfriNIC region as damage and routing around it.

> 

> Well the AfriNIC Bylaws [link 2] Section 3 [Objectives of the Company], sub section (iii) below ensures;

> 

> (iii) to promote responsible management of Internet resources throughout the African region, as well as the responsible development and operation of Internet infrastructures;

> 

> Surely, AfriNIC as a company can not just  forget its own responsibility as bestowed to her in the Bylaws. The people who are seeing this as overly restrictive should understand that we can not break our own rules/laws and I don't see how such restrictions are a damage if they are meant to ensure that number resources within the AfriNIC region are put to good use for the purpose they were requested for in order to develop the Internet in Africa.


I’m not suggesting that they should. We have agreed to disagree on what constitutes compliance with that requirement in the past. I see no reason that will change in the foreseeable future.


>>> So now we have folks with capital spending most of their energy moving IPv4 address space all over the place since its a currency that ensures serious economic benefits. 

>> 

>> If you know of a way to stop this, I’m all ears.

>> 

>> Impossible to stop economic activities. 

> 

> Thus, we felt it was better in the ARIN region to provide reasonable accommodation while still preserving useful aspects of regulation. I think we have achieved

> a good balance there and that there’s relatively little black market movement of IPv4 resources in the ARIN region which has also made it possible to have

> better accountability (for example the recent reclamation of a large quantity of improperly transferred address space).

> 

> What works in ARIN may not necessarily work in the AfrNIC region but there are currently some policy proposals the working group are looking at within AfriNIC. We shall see how that goes as we adhere to our Bylaws. 


Sure… Conversely, however, just because it’s Africa doesn’t necessarily mean that what works elsewhere won’t work there. Far more similarities exist than differences.


>  

>> Faced with the business externalities that they are, they really have no choice but to try

>> and acquire enough IPv4 to support customer demand for as long as possible. I can assure you that each of them would love to see customer demand for IPv4 go away.

>> 

>> Hence my point that both shall co-exist for years. So we cant devalue IPv4 in an attempt to promote IPv6 and this is why responsible management of IPv4 space must be ensured.

> 

> Nobody said that devaluing IPv4 was a way to promote IPv6… We said that deploying IPv6 would devalue IPv4.

> 

> Ok understood, but my point was that the deployment of IPv6 is not a clear cut process as folks seem to suggest since the decisions are internal for each entity.


Oh, it is a clear cut process. The timetable is unclear because each entity gets to implement on their own schedule, but the process is very clear cut and the sooner more entities do deploy IPv6, the better for everyone else.

Nonetheless, regardless of the timeframe, the fact remains that the best and most effective way to devalue IPv4 is to deploy IPv6.


>> OTOH, IPv6 is available to the vast majority of eyeballs in the US. Comcast has 100% IPv6 coverage, as do most of the major cellular carriers. AIUI, the other major

>> eyeball ISPs in the US are fast approaching that.

>> 

>> One can attribute different factors to such outcomes beyond just the US.

> 

> I cited the US as an example because it is the market with which I have the greatest familiarity. If you want another good example outside of the US, look at the current rate of IPv6 adoption around India.

> 

> Globally, Google is seeing more than 30% of their traffic via IPv6. The prime laggards according to their map are northern Asia, Russia, Greenland, the Middle East, and the vast majority of Africa.

> 

> In fact, Africa is by far, the least deployed continent for IPv6, with notable exceptions in Togo and Gabon and problematic deployments in Kenya and Burundi.

> 

> I think you will see spontaneous deployment of IPv6 across China in the near future. Being as all the major ISPs in China are essentially one organization owned by the government, when the mandate finally comes, implementation and deployment will be quite rapid.

> 

> I like the "you will see spontaneous deployment of IPv6" part of your response but IPv4 is still useful for a while.


Nobody said it wasn’t.

Nonetheless, it is a fact that IPv4 is:
	+	Increasing in cost both acquisition cost and operational cost.
	+	Decreasing in functionality
	+	At some point will be capable of reaching a declining subset of the full internet.


>>> Atleast this is how I see it. Capitalism at its best.

>> 

>> Or one of the finest examples of how capitalism is nearly as flawed as the alternatives.

>> 

>> No system is perfect after all.

> 

> Agreed… But people love to tout th failures of socialism while often ignoring those same failures in a different form in capitalism.

> 

> In socialism, the lack of reward for effort and the lack of incentive to rise above is cited quite often.

> In capitalism, the failure to tie reward and/or incentive to a greater good is mostly overlooked. The concept of perverse incentives occasionally gets mentioned, but usually in arguing for deregulation which often exacerbates the most harmful perverse incentives.

> 

> The take away for me is that both have pros and cons however, for any industry to thrive in either systems, raw-materials are extremely important since they form part of the means of production. Textile factories in both a socialist and capitalist systems would still require Cotton as a raw material.


A textile factory can use cotton. In many cases, it can also use polyester, rayon, dacron, nylon, hemp, wool, or a variety of other fibers. In Uganda, textiles have been made of fibers from tree bark.

There are many alternatives and if Cotton became scarce and expensive, most textile production would shift to other materials rather rapidly.


> In our case within the Internet Industry, IPv4 addresses (just like IPv6) are the scarce resources which form part of the means of production for the final service the Internet. As such, responsible management of INR is akeen to the promotion and development of the Internet Industry in Africa which so many on the continent have come to rely on to sell their labor, research & education, economic activities, social life etc. 


In our case, IPv4 and IPv6 are both raw materials. In an ideal world, by now, they would be functional equivalents and could be used interchangeably.

Unfortunately, that ship has sailed and we are where we are. As such, the textile comparison fails because we have cotton and wool, but currently only 40% of the population can wear wool, while 100% of the population can wear cotton, but we only have enough cotton to make garments for 1/5th of the worlds population, thus every person wanting to wear a cotton garment must share that garment with at least 4 other strangers.

Owen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20200531/aa10a12a/attachment-0001.html>