Hello the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse. Regards Raymond Macharia
the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse.
Not sure what proposal this article refers to. The proposals under discussion are posted here <http://www.arin.net/policy/proposals/proposal_archive.html> Note that all kinds of crazy ideas get proposed to ARIN and most of them never see the light of day. If you carefully read the Adopted and Abandoned proposals listed on the ARIN page you will see that the Adopted ones tend to be incremental improvements, not radical change. In the case of this "address commodity" idea, most people believe that it will not do much, if anything, to mitigate the effects of running out of IPv4 addresses which will hit us in about two years from now. It is too costly for companies to free up addresses and put them on the market, meaning prices will be very high, and supply will be very low. This means that there will not be enough liquidity for a true "market" to form, just occasional transactions not much different from the grey market today. And IPv6 addresses will remain 100% free of charge which will reduce demand, i.e. potential buyers might prefer to spend their cash on IPv6 capability knowing that it frees them from the need to participate in a dubious IPv4 market scheme. Not to mention the negative effects of an increase in the number of small routes in the Internet routing table. You might pay half a million dollars for a /24 only to find that ISPs will not accept your route announcements. Yet another reason why a registered IP address block, in itself, is not valuable enough to buy or sell. In any case, nothing is decided before the meeting in Denver on the 6th to 9th of April 2008. In fact, even then it won't be decided, just past the first hurdle. If you or anyone really is interested in such things, pro or con, then you should join the ARIN PPML list here <http://lists.arin.net/mailman/listinfo/ppml> --Michael Dillon
On Mon, Feb 18, 2008 at 02:11:57PM -0000, michael.dillon@bt.com <michael.dillon@bt.com> wrote a message of 47 lines which said:
The proposals under discussion are posted here <http://www.arin.net/policy/proposals/proposal_archive.html>
The ones that were already screened. The policy the OP is talking about seems to be this one, not yet screened: http://lists.arin.net/pipermail/ppml/2008-February/009978.html
The ones that were already screened. The policy the OP is talking about seems to be this one, not yet screened:
http://lists.arin.net/pipermail/ppml/2008-February/009978.html
That's the one. And if you go down to the very bottom of the message and click on [thread] you will be able to read through all the discussion to date regarding this proposal. --Michael Dillon
On Feb 18, 2008, at 6:11 AM, <michael.dillon@bt.com> wrote:
the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse.
Not sure what proposal this article refers to. The proposals under discussion are posted here <http://www.arin.net/policy/proposals/proposal_archive.html>
Only after they are accepted by the AC as a formal proposal and assigned a number. The article is referring to a proposal submitted by the ARIN AC which, if adopted, would allow for organizations to transfer address space to other organizations under dramatically different circumstances than are allowed today. That policy proposal is available here: http://lists.arin.net/pipermail/ppml/2008-February/009978.html
Not to mention the negative effects of an increase in the number of small routes in the Internet routing table. You might pay half a million dollars for a /24 only to find that ISPs will not accept your route announcements. Yet another reason why a registered IP address block, in itself, is not valuable enough to buy or sell.
Actually, the proposed policy provides significant limits to the de-aggregation which can be done in its current form.
In any case, nothing is decided before the meeting in Denver on the 6th to 9th of April 2008. In fact, even then it won't be decided, just past the first hurdle.
True. However, there is also a NANOG BoF where several members of the ARIN AC will be present to get input from the community on the subject of this proposal, and, other ideas/strategies for IPv4 after IANA free pool exhaustion. The BoF will be in the Crystal Room from 4:00 PM to 5:30 PM on Tuesday, Feb. 19. Owen DeLong ARIN AC
Distribution: This "idea" comes from clueless individuals who want to know "who owns the Internet"? When I worked at Enron Broadband Services "the crooked E", management wanted to buy PSInet so that "we" could developed a trading desk for IP address blocks. We informed management that neither EBS or PSInet owned their IP addresses but rented them from ARIN. And when the organization indicates that IP addresses are no longer needed, they can be returned to ARIN or ARIN can come and get them from the organization per ARIN AUP and other policies that users signed when making a request to ARIN. (Review a court case several years ago, about a company going into bankruptcy, I believe, claiming that "their" IP addresses were part of the assets of the company...) Now for those who could not follow the last paragraph, the analogy is when you were young and renting your apartment or house and you wanted to make money selling one of the rooms of your rented apartment or house. So anyone with spare /16 or larger send the blocks back to ARIN so they can be good stewards of the diminishing resource. John (ISDN) Lee I Still Don't kNow It Suites Dennis's Needs ________________________________ From: owner-nanog@merit.edu on behalf of Raymond Macharia Sent: Mon 2/18/2008 8:39 AM To: 'NANOG list' Subject: IPV4 as a Commodity for Profit Hello the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse. Regards Raymond Macharia
Hi John, I think that comment is way out of line. In fact, I met at LINX one of authors of a trading proposal. They are smart, well educated individuals. Markets have proven to be excellent mechanisms for allocating resources fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses. Nor is the trading of IP addresses inconsistent with ARIN ownership. Regards, Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com Wireless: 1-212-444-8829. Landline: 33-1-4346-3209. French Wireless: 33-6-14-33-48-97. AOL Messenger: GlobalBandwidth rod.beck@hiberniaatlantic.com rodbeck@erols.com ``Unthinking respect for authority is the greatest enemy of truth.'' Albert Einstein. -----Original Message----- From: owner-nanog@merit.edu on behalf of John Lee Sent: Mon 2/18/2008 4:45 PM To: Raymond Macharia; NANOG list Subject: RE: IPV4 as a Commodity for Profit Distribution: This "idea" comes from clueless individuals who want to know "who owns the Internet"? When I worked at Enron Broadband Services "the crooked E", management wanted to buy PSInet so that "we" could developed a trading desk for IP address blocks. We informed management that neither EBS or PSInet owned their IP addresses but rented them from ARIN. And when the organization indicates that IP addresses are no longer needed, they can be returned to ARIN or ARIN can come and get them from the organization per ARIN AUP and other policies that users signed when making a request to ARIN. (Review a court case several years ago, about a company going into bankruptcy, I believe, claiming that "their" IP addresses were part of the assets of the company...) Now for those who could not follow the last paragraph, the analogy is when you were young and renting your apartment or house and you wanted to make money selling one of the rooms of your rented apartment or house. So anyone with spare /16 or larger send the blocks back to ARIN so they can be good stewards of the diminishing resource. John (ISDN) Lee I Still Don't kNow It Suites Dennis's Needs ________________________________ From: owner-nanog@merit.edu on behalf of Raymond Macharia Sent: Mon 2/18/2008 8:39 AM To: 'NANOG list' Subject: IPV4 as a Commodity for Profit Hello the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse. Regards Raymond Macharia
On Feb 18, 2008 11:05 AM, Rod Beck <Rod.Beck@hiberniaatlantic.com> wrote:
<snip>
Markets have proven to be excellent mechanisms for allocating resources
fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses.
Nor is the trading of IP addresses inconsistent with ARIN ownership.
Regards,
Roderick S. Beck
Markets have a history of efficiently allocating resources, this much is true. My concern is when IP allocations are based on fiscal merit instead of technical merit. Also, don't forget speculators within a market. Do you really want the price of IP blocks bid up by "IP day traders"? -brandon
Hi Brand, You want some speculators to create liquidity. At the same time you want to avoid the excessive trading that leads to speculative bubbles, whether is the equity bubble of 2000 or the real estate bubble of 2007. I think the answer is that you need the ability to impose a transaction tax. For example, there is no real estate bubble in France because the one time costs of buying and selling are quite high. A transaction tax would discourage excessive trading. Bear in mind that I have devoted zero time to thinking how to construct such a market. :) Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com Wireless: 1-212-444-8829. Landline: 33-1-4346-3209. French Wireless: 33-6-14-33-48-97. AOL Messenger: GlobalBandwidth rod.beck@hiberniaatlantic.com rodbeck@erols.com ``Unthinking respect for authority is the greatest enemy of truth.'' Albert Einstein. -----Original Message----- From: Brandon Galbraith [mailto:brandon.galbraith@gmail.com] Sent: Mon 2/18/2008 5:18 PM To: Rod Beck Cc: John Lee; Raymond Macharia; NANOG list Subject: Re: IPV4 as a Commodity for Profit On Feb 18, 2008 11:05 AM, Rod Beck <Rod.Beck@hiberniaatlantic.com> wrote:
<snip>
Markets have proven to be excellent mechanisms for allocating resources
fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses.
Nor is the trading of IP addresses inconsistent with ARIN ownership.
Regards,
Roderick S. Beck
Markets have a history of efficiently allocating resources, this much is true. My concern is when IP allocations are based on fiscal merit instead of technical merit. Also, don't forget speculators within a market. Do you really want the price of IP blocks bid up by "IP day traders"? -brandon
Markets have a history of efficiently allocating resources, this much is true.
Its only true when those markets have sufficient liquidity. We have gotten used to regulated markets where the SEC and others, make sure that liquidity is maintained and block trading of instruments when liquidity is insufficient. In order to be liquid, a market has to have a sufficiently stable supply. If it is a commodity market, where the instruments being traded are then consumed by the buyer in producing their products or services, then the only way to maintain liquidity is to have a steady supply of fresh commodities. This is not too hard when talking about orange juice, pork bellies or even gold. But they are not making any more IPv4 addresses. The supply is fixed and the problem which we face is that the global unallocated supply is shrinking to the point of exhaustion in about two years. In this situation, I can't see that it would be possible to have a liquid IPv4 address market for more than a short period of time. Since IPv4 trading is not regulated by the SEC (unless it comes under some umbrella provision) the way is open for some group of SMART, WELL-EDUCATED INDIVIDUALS to rig the market and make off with lots of cash when it collapses. Everyone in the network operations sphere is a rank amateur when it comes to making markets, maintaining liquidity and regulation of trading. We cannot possibly hope to create a market within the next two years which is beneficial to the Internet network operations industry. --Michael Dillon
Michael, On Feb 18, 2008, at 10:15 AM, <michael.dillon@bt.com> <michael.dillon@bt.com
wrote: We cannot possibly hope to create a market within the next two years which is beneficial to the Internet network operations industry.
A market will exist whether or not "we" want to create it and it doesn't matter how long lived it is. Without some form of regulation (a bit hard since it would need to be applied globally), it is almost certain it will be extremely painful and folks who "shouldn't" make lots of money will. So it goes. Regards, -drc
On Mon, 18 Feb 2008 12:18:10 -0800 David Conrad <drc@virtualized.org> wrote:
Michael,
On Feb 18, 2008, at 10:15 AM, <michael.dillon@bt.com> <michael.dillon@bt.com
wrote: We cannot possibly hope to create a market within the next two years which is beneficial to the Internet network operations industry.
A market will exist whether or not "we" want to create it and it doesn't matter how long lived it is. Without some form of regulation (a bit hard since it would need to be applied globally), it is almost certain it will be extremely painful and folks who "shouldn't" make lots of money will. So it goes.
Yah. A market exists today, though it's perforce sub rosa. An interesting operational question is how to prevent deaggregation as a result of a market. If, say, a company isn't using half of its address space, could it sell that half, to several other parties? Can that be prevented by market means? See http://www.cs.columbia.edu/~smb/papers/piara/index.html for a paper I and some others wrote some years ago on these topics. --Steve Bellovin, http://www.cs.columbia.edu/~smb
Steve,
Yah. A market exists today, though it's perforce sub rosa.
An interesting operational question is how to prevent deaggregation as a result of a market. If, say, a company isn't using half of its address space, could it sell that half, to several other parties? Can that be prevented by market means?
This is a strong argument for regulation of the market. A regulated market could provide liquidity needed by those who would otherwise find <unregulated> means to accomplish their ends (such as making private deals that are perhaps undetectable). It provides an incentive for people to do the right thing, assuming the cost of doing so is not prohibitive (a big assumption if ever there was one). Eliot
On Tue, 19 Feb 2008 08:23:53 +0100 Eliot Lear <lear@cisco.com> wrote:
Steve,
Yah. A market exists today, though it's perforce sub rosa.
An interesting operational question is how to prevent deaggregation as a result of a market. If, say, a company isn't using half of its address space, could it sell that half, to several other parties? Can that be prevented by market means?
This is a strong argument for regulation of the market. A regulated market could provide liquidity needed by those who would otherwise find <unregulated> means to accomplish their ends (such as making private deals that are perhaps undetectable).
I have no problem with regulating markets -- I tend to think they work better that way. (He ducks, fending off the attacks of maddened libertarians...)
It provides an incentive for people to do the right thing, assuming the cost of doing so is not prohibitive (a big assumption if ever there was one).
That's something the market will do very well: balancing the profits against the cost of renumbering. --Steve Bellovin, http://www.cs.columbia.edu/~smb
Steve, On Feb 18, 2008, at 11:33 PM, Steven M. Bellovin wrote:
This is a strong argument for regulation of the market. A regulated market could provide liquidity needed by those who would otherwise find <unregulated> means to accomplish their ends (such as making private deals that are perhaps undetectable).
I have no problem with regulating markets -- I tend to think they work better that way.
Given the decentralized nature of the Internet and the lack of a single legal/policy regime that covers it, the challenge is in identifying viable candidates for the regulator. Regards, -drc
David Conrad wrote:
Steve,
On Feb 18, 2008, at 11:33 PM, Steven M. Bellovin wrote:
This is a strong argument for regulation of the market. A regulated market could provide liquidity needed by those who would otherwise find <unregulated> means to accomplish their ends (such as making private deals that are perhaps undetectable).
I have no problem with regulating markets -- I tend to think they work better that way.
Given the decentralized nature of the Internet and the lack of a single legal/policy regime that covers it, the challenge is in identifying viable candidates for the regulator.
That is a big challenge in this space, considering that there is no particular national alignment in the address structure. The approach in the proposals so far is to attempt to create a framework around recognition of market-based outcomes that act as implicit constraints on the market. The proposals so far for recognition of address transfers in ARIN, RIPE and APNIC all attempt to define a class of transactions that would be regarded as eligible for entering in their respective address registries. That said, they each propose a different set of constraints! Geoff
Brandon, On Feb 18, 2008, at 9:18 AM, Brandon Galbraith wrote:
Markets have a history of efficiently allocating resources, this much is true. My concern is when IP allocations are based on fiscal merit instead of technical merit. Also, don't forget speculators within a market. Do you really want the price of IP blocks bid up by "IP day traders"?
Presumably, the market would occur when the IPv4 address free pool has been exhausted. Without a market, there will be no IPv4 address space. With a market, IPv4 address space will be available at a price. That price will be constrained by the cost the IPv4-desirous ISP would face in deploying IPv6 + NAT-PT. If IPv6 + NAT-PT isn't sufficient, the IPv4-desirous ISP will face a choice of paying whatever the market will bear or doing without (implying double/triple NAT, etc.) since there won't be any other alternatives. It's not clear to me how many people actually understand the implication of IPv4 free pool exhaustion... Regards, -drc
On 18 feb 2008, at 21:06, David Conrad wrote:
Presumably, the market would occur when the IPv4 address free pool has been exhausted. Without a market, there will be no IPv4 address space. With a market, IPv4 address space will be available at a price.
I wouldn't be so sure. How many millions of addresses do the Comcasts of this world use up every year? 2? 5? 8? (That is PER large ISP, NOT for all of them together.) When trying to obtain such a number of addresses immediately after the RIRs are out will almost certainly be possible, but at what price? The likes of HP will have to spend a lot of money auditing their networks or take huge risks freeing up millions of addresses. (How do you know there isn't some 15-year-old legacy system whose address is hardcoded all over the place (no DHCP back then! Even DNS wasn't ubiquitous in the early 1990s) in a given address block?) I'm not sure if they'll be prepared to do this for a price that the big ISPs will find affordable. But even if this works out the first round, supply can only go down and the price can only go up the second round. If I were an ISP, I wouldn't start a process like this that can only end in tears a few years down the road, but rather, go for the alternatives where I don't have to obtain fresh IPv4 space immediately. It only makes sense to buy address space in bulk for large ISPs if they're caught by surprise and need time to implement kicking the IPv4 habit. As for those of us who aren't ISPs connecting hundreds of thousands of customers per year: that single /8 a year that make up 90% of the requests can probably be accommodated from the normal return of address space, and if not, people who need a /24 can afford to pay a whole lot more than those who need a /10, so supply and demand should work fine here. By the way, we already have a perfectly functioning IPv4 address market. But it's not about owning, but about renting. Buy IP transit service and you'll get a bunch of IP addresses thrown in.
I am not sure it's a perfectly functioning market. The whole point of a market is to penalize the holding excess inventory of IP addresses. There is no penalty today because there is no opportunity cost to holding excess inventory. :) What's amazingly ironic is how the free market guys suddenly vanish when one wants to apply free markets to their industry ... :) Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com Wireless: 1-212-444-8829.
It's good that this discussion is happening now. To make the discussion as productive as possible, it's probably a good idea to clarify assumptions and terms. We all know what "market" means -- but in all likelihood many of the things we all "know" do not overlap, and some are probably mutually contradictory. If thinking about IPv4 addresses as a "commodity" has any validity, it comes from the assumption that making them subject to "market pricing" will increase supply, i.e., incentive current surplus holders to make that surplus available to would-be buyers. In other "commodity" markets, the connection between market pricing and increased supply is *production* -- i.e., when the revealed price of a commodity goes up, those who are capable of making it are motivated to make more, or to jump into the market for the first time. In other commodity markets, that motivation is bounded by the threat of alternative suppliers, by the impracticality of hoarding, and by the inability of the potential seller to use more of the commodity directly. In other words, the existence or potential emergence of alternative producers/suppliers tends to discourage hoarding to maximize prices (because there's no guarantee that prices will stay high, much less go even higher), and the lack of direct "use value" reduces any countervailing incentive that the prospective seller to just hold the assets in perpetuity, until they can be used in-house. In the case of IPv4 addressing, none of these bounding conditions apply. No more IPv4 addresses can be produced, and they're almost certain to have unique (if not irreplaceable) use value, at least for some classes of ISPs that exist today, for at least a decade or more (or as long as those kinds of ISPs exist, whichever is shortest). That means the potential price is always going to be higher tomorrow than it is today, right up to the day before the last day that IPv4 becomes useless. Which means hoarding is going to continue to be the most sensible behavior for all surplus holders -- even those that no longer have any Internet-related ops or business interests. This countervailing incentive is much stronger for surplus holders that *do* still have such interests. Knowing that IPv4 addresses that they might need in the future will certainly cost more (maybe lots more) than whatever price they could command for surplus IPv4 today, growing ISPs are not likely to contribute much to the salable, "liquid" address pool. Worse still, so long as IPv4 continues to be a non-substitutable, must-have input for certain kinds of ISPs, ISPs like that will know that the threat of competition from existing or hypothetical future competitors will be absolutely limited by the availability of IPv4 address space. For them, making IPv4 address space unavailable to competitors is a perfectly sensible "use", and one with quite a lot of value. An unmediated market is not going to "work", for almost any meaning of that term. Get over it. Anyone who disagrees should point out anything disputable in the above, or else clarify what they actually think/hope that an IPv4 address market will achieve. TV On Feb 19, 2008, at 12:04 PM, Rod Beck wrote:
I am not sure it's a perfectly functioning market.
The whole point of a market is to penalize the holding excess inventory of IP addresses.
There is no penalty today because there is no opportunity cost to holding excess inventory. :)
What's amazingly ironic is how the free market guys suddenly vanish when one wants to apply free markets to their industry ...
:)
Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com Wireless: 1-212-444-8829.
I suppose that this is easier said than done. I doubt whether those "hording" IP space all allocate them in a uniform contiguous manner so even if there was sufficient monitory incentive I wonder whether it would be worth it trying to renumber an entire network so as to "sell" a whole contiguous block. This introduces the complexity that not just any IP block will do but it has to satisfy certain conditions, which complicates things even further. The amount of disruptions in the process of renumbering would erode the perceived gains very quickly. But let us see what the meeting will yield. Raymond Macharia Head of Engineering AccessKenya Email: engineering@accesskenya.com Web: http://www.accesskenya.com/ Communications Solutions Ltd 4th Floor | Museum Hill Centre P.O. Box 43588 - 00100 | Nairobi From: Rod Beck [mailto:Rod.Beck@hiberniaatlantic.com] Sent: 2008-02-18 20:06 To: John Lee; Raymond Macharia; NANOG list Subject: RE: IPV4 as a Commodity for Profit Hi John, I think that comment is way out of line. In fact, I met at LINX one of authors of a trading proposal. They are smart, well educated individuals. Markets have proven to be excellent mechanisms for allocating resources fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses. Nor is the trading of IP addresses inconsistent with ARIN ownership. Regards, Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com Wireless: 1-212-444-8829. Landline: 33-1-4346-3209. French Wireless: 33-6-14-33-48-97. AOL Messenger: GlobalBandwidth rod.beck@hiberniaatlantic.com rodbeck@erols.com ``Unthinking respect for authority is the greatest enemy of truth.'' Albert Einstein. -----Original Message----- From: owner-nanog@merit.edu on behalf of John Lee Sent: Mon 2/18/2008 4:45 PM To: Raymond Macharia; NANOG list Subject: RE: IPV4 as a Commodity for Profit Distribution: This "idea" comes from clueless individuals who want to know "who owns the Internet"? When I worked at Enron Broadband Services "the crooked E", management wanted to buy PSInet so that "we" could developed a trading desk for IP address blocks. We informed management that neither EBS or PSInet owned their IP addresses but rented them from ARIN. And when the organization indicates that IP addresses are no longer needed, they can be returned to ARIN or ARIN can come and get them from the organization per ARIN AUP and other policies that users signed when making a request to ARIN. (Review a court case several years ago, about a company going into bankruptcy, I believe, claiming that "their" IP addresses were part of the assets of the company...) Now for those who could not follow the last paragraph, the analogy is when you were young and renting your apartment or house and you wanted to make money selling one of the rooms of your rented apartment or house. So anyone with spare /16 or larger send the blocks back to ARIN so they can be good stewards of the diminishing resource. John (ISDN) Lee I Still Don't kNow It Suites Dennis's Needs ________________________________ From: owner-nanog@merit.edu on behalf of Raymond Macharia Sent: Mon 2/18/2008 8:39 AM To: 'NANOG list' Subject: IPV4 as a Commodity for Profit Hello the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse. Regards Raymond Macharia
Thanks Rod, The traders at Enron are/where PhDs with MBAs from Wharton, Harvard and Oxford, UK so well educated and smart was not the point of the comment. ARIN (and other RIRs) and the rules of use of IP address were specifically setup to allow global communications around the world with a large number of entities on an equal basis. IPv4 would be a fixed size pool of a commodity and from my training at Enron in trading (they were the largest in the world by number and volume of trades) it only works under several rules. You have a group of suppliers and consumers, you have a measurable commodity and a standard way of measuring it, you have a standard set of commodities, you can assign a value to the commodity etc. If I grow corn or drill for "Texas Light Crude" there standards in place so that if you drill in the Middle east, China or Russia you produce the same "Texas Light Crude" to trade. Trading of IP address does violate ARIN rules as ARIN has explained to me since to "Trade" something is to have title or ownership of that item and that ownership belongs to the cognizant RIR not to a company or person. Regards John PS: The RIRs are community driven and so if the community wants to become a market place, they can petition ARIN have a vote and change if the majority of the community wants to. ________________________________ From: Rod Beck [mailto:Rod.Beck@hiberniaatlantic.com] Sent: Mon 2/18/2008 12:05 PM To: John Lee; Raymond Macharia; NANOG list Subject: RE: IPV4 as a Commodity for Profit Hi John, I think that comment is way out of line. In fact, I met at LINX one of authors of a trading proposal. They are smart, well educated individuals. Markets have proven to be excellent mechanisms for allocating resources fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses. Nor is the trading of IP addresses inconsistent with ARIN ownership. Regards, Roderick S. Beck Director of European Sales Hibernia Atlantic 1, Passage du Chantier, 75012 Paris http://www.hiberniaatlantic.com <http://www.hiberniaatlantic.com/> Wireless: 1-212-444-8829. Landline: 33-1-4346-3209. French Wireless: 33-6-14-33-48-97. AOL Messenger: GlobalBandwidth rod.beck@hiberniaatlantic.com rodbeck@erols.com ``Unthinking respect for authority is the greatest enemy of truth.'' Albert Einstein. -----Original Message----- From: owner-nanog@merit.edu on behalf of John Lee Sent: Mon 2/18/2008 4:45 PM To: Raymond Macharia; NANOG list Subject: RE: IPV4 as a Commodity for Profit Distribution: This "idea" comes from clueless individuals who want to know "who owns the Internet"? When I worked at Enron Broadband Services "the crooked E", management wanted to buy PSInet so that "we" could developed a trading desk for IP address blocks. We informed management that neither EBS or PSInet owned their IP addresses but rented them from ARIN. And when the organization indicates that IP addresses are no longer needed, they can be returned to ARIN or ARIN can come and get them from the organization per ARIN AUP and other policies that users signed when making a request to ARIN. (Review a court case several years ago, about a company going into bankruptcy, I believe, claiming that "their" IP addresses were part of the assets of the company...) Now for those who could not follow the last paragraph, the analogy is when you were young and renting your apartment or house and you wanted to make money selling one of the rooms of your rented apartment or house. So anyone with spare /16 or larger send the blocks back to ARIN so they can be good stewards of the diminishing resource. John (ISDN) Lee I Still Don't kNow It Suites Dennis's Needs ________________________________ From: owner-nanog@merit.edu on behalf of Raymond Macharia Sent: Mon 2/18/2008 8:39 AM To: 'NANOG list' Subject: IPV4 as a Commodity for Profit Hello the article here http://www.networkworld.com/news/2008/021308-ipv6-delay.html is an interesting read given the current state of IPv4 depletion/IPv6 conversion operational climate. As it is indicated, it's a proposal and there are considerations as to whether it makes things better or worse. Regards Raymond Macharia
John, On Feb 18, 2008, at 9:48 AM, John Lee wrote:
ARIN (and other RIRs) and the rules of use of IP address were specifically setup to allow global communications around the world with a large number of entities on an equal basis.
More IP addresses were allocated prior to the creation of the RIRs than since. The terms under which those early allocations were made is a bit fuzzy (to put it mildly). ARIN is attempting to remedy this fuzziness by creating a "legacy RSA" in which you give up some potential rights in exchange for some potential rights (it also asserts ARIN has the right to decide what happens with all the legacy space which causes some concern internationally, but that's not relevant here). See http://www.arin.net/registration/legacy/ index.html for more information.
PS: The RIRs are community driven and so if the community wants to become a market place, they can petition ARIN have a vote and change if the majority of the community wants to.
The question really isn't whether the ARIN community will want a market to exist. A market, albeit black or grey, exists already. The question is how ARIN will deal with the market after the IPv4 free pool exhausts. Ignoring the market will likely result in the marginalization of ARIN for services such as registration of address space (for good or ill). Not ignoring the market will likely result in all sorts of 'fun', in the worst case similar to what has occurred in the domain name market. "Choose wisely." Regards, -drc
David Conrad wrote:
John,
On Feb 18, 2008, at 9:48 AM, John Lee wrote:
ARIN (and other RIRs) and the rules of use of IP address were specifically setup to allow global communications around the world with a large number of entities on an equal basis.
More IP addresses were allocated prior to the creation of the RIRs than since. The terms under which those early allocations were made is a bit fuzzy (to put it mildly).
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly. That will likely affect the value equation even before pool exhaustion. Furthermore, I expect more credence will be lent to the reclaiming efforts, and pre-RIR swamp space has lots of candidates. Class-E, rfc3330 and similar reclamation might occur as well. "effort not worth the return" objections, maybe its true, not under current allocation guidelines and resulting burn rate. Therefore, I would favor not performing reclamations until/unless those are changed. Consider under new drastically more miserly policies, likely return will be worth the effort, if not right away, then in a bit in the future. ipv4 experimental space has no reason to exist, experiments should all be ipv6 anyways.
ARIN is attempting to remedy this fuzziness by creating a "legacy RSA" in which you give up some potential rights in exchange for some potential rights (it also asserts ARIN has the right to decide what happens with all the legacy space which causes some concern internationally, but that's not relevant here). See http://www.arin.net/registration/legacy/ index.html for more information.
PS: The RIRs are community driven and so if the community wants to become a market place, they can petition ARIN have a vote and change if the majority of the community wants to.
The question really isn't whether the ARIN community will want a market to exist. A market, albeit black or grey, exists already.
And we should be taking into account the value addresses have in this market as opposed to the "white" market, on an ongoing basis.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts.
I expect the value will skyrocket, whether on the black, grey or white market.
Ignoring the market will likely result in the marginalization of ARIN for services such as registration of address space (for good or ill). Not ignoring the market will likely result in all sorts of 'fun', in the worst case similar to what has occurred in the domain name market.
If there is a market, it makes little sense to ignore it, gonna be all sorts of consequences to doing so.
"Choose wisely."
Regards, -drc
Joe, On Feb 19, 2008, at 4:28 AM, Joe Maimon wrote:
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly.
No. You might see a push towards this, but it will take far longer to get policies modified than there will be time left and there will be increased 'competition' among the RIRs that will strongly discourage this course of action (as someone who has proposed a policy that would impose more restrictions on v4 allocations, I have already heard the "if we modify our policies to be more conservative, then the folks in other RIRs will get an advantage" several times). The RIR bureaucracy is a ponderous ship that turns very slowly and has multiple captains who do not necessarily agree on the direction to turn. IPv4 allocation policy revisions aren't going to save us.
Furthermore, I expect more credence will be lent to the reclaiming efforts, and pre-RIR swamp space has lots of candidates.
What incentive to a holder of early allocations is there to return address space voluntarily?
Class-E,
Efforts to redefine class E have stalled because there is simply no way it can be used for anything other than private space. There are too many implementations out there that will never be modified (e.g., Windows 98) on which you can't even configure class E space.
rfc3330 and similar reclamation might occur as well.
IANA recently reclaimed 14/8. I think that added 3 _weeks_ to the expected runout date. That was likely the last "easily" reclaimable block.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts. I expect the value will skyrocket, whether on the black, grey or white market.
Yep. And the question is: as an ISP or other address consuming organization, what will you do when the cost of obtaining IPv4 addresses skyrockets? So far, as far as I can tell, the answer to that question (in most cases) has been putting hands over ears and saying "La la la" loudly. See <http://www.networkworld.com/news/2008/020608-ipv4-address-depletion.html
.
Regards, -drc
David Conrad wrote:
Joe,
On Feb 19, 2008, at 4:28 AM, Joe Maimon wrote:
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly.
No.
You might see a push towards this, but it will take far longer to get policies modified than there will be time left and there will be increased 'competition' among the RIRs that will strongly discourage this course of action (as someone who has proposed a policy that would impose more restrictions on v4 allocations, I have already heard the "if we modify our policies to be more conservative, then the folks in other RIRs will get an advantage" several times).
Things might get different when the end is staring us in the face.
The RIR bureaucracy is a ponderous ship that turns very slowly and has multiple captains who do not necessarily agree on the direction to turn. IPv4 allocation policy revisions aren't going to save us.
RIR's have bylaws about emergency policies, dont they? Its not about saving, its about prolonging the end and how long that migh be expected to last.
Furthermore, I expect more credence will be lent to the reclaiming efforts, and pre-RIR swamp space has lots of candidates.
What incentive to a holder of early allocations is there to return address space voluntarily?
None, but the nice thing about being a registry is that reclaiming things is as simple as allocating it to somebody else. Buyer beware and all that. And in the absence of any other method of obtaining ipv4, I would expect RIR mebership to push for aggressive reclamation, with policy change to make it worthwhile.
Class-E,
Efforts to redefine class E have stalled because there is simply no way it can be used for anything other than private space.
Amazing that so much effort can go into ipv6 but nobody can spare a few hours per product to remove a couple lines of code?
There are too many implementations out there that will never be modified (e.g., Windows 98) on which you can't even configure class E space.
Faced with a choice of ipv6 and no ipv4 or ipv6 and class-e ipv4, which would you choose? Not like windows98 (if there are any still around that mean anything to anybody) has ipv6 either.
rfc3330 and similar reclamation might occur as well.
IANA recently reclaimed 14/8. I think that added 3 _weeks_ to the expected runout date. That was likely the last "easily" reclaimable block.
Reclamation efforts without policy change isnt profitable and would only become so if miser mode is in effect.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts.
I expect the value will skyrocket, whether on the black, grey or white market.
Yep. And the question is: as an ISP or other address consuming organization, what will you do when the cost of obtaining IPv4 addresses skyrockets?
Pass it on to the customer. Reclaim. Scavenge. Engineer more nats and workarounds while accelerating ipv6. Get budget and manpower to actually make changes. Drag the users kicking and screaming, cause thats what it will take.
So far, as far as I can tell, the answer to that question (in most cases) has been putting hands over ears and saying "La la la" loudly. See <http://www.networkworld.com/news/2008/020608-ipv4-address-depletion.html
.
Things will likely be different in 2010
Regards, -drc
On Feb 19, 2008, at 8:47 AM, Joe Maimon wrote:
David Conrad wrote:
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly. No. You might see a push towards this, but it will take far longer to get policies modified than there will be time left and there will be increased 'competition' among the RIRs that will strongly discourage this course of action (as someone who has proposed a
Joe, On Feb 19, 2008, at 4:28 AM, Joe Maimon wrote: policy that would impose more restrictions on v4 allocations, I have already heard the "if we modify our policies to be more conservative, then the folks in other RIRs will get an advantage" several times).
Things might get different when the end is staring us in the face.
By then, the policy process will take too long to be meaningful.
The RIR bureaucracy is a ponderous ship that turns very slowly and has multiple captains who do not necessarily agree on the direction to turn. IPv4 allocation policy revisions aren't going to save us.
RIR's have bylaws about emergency policies, dont they?
Yes, but, I think it is unlikely, at least in ARIN's case that the BoT will consider runout an emergency.
Its not about saving, its about prolonging the end and how long that migh be expected to last.
Prolonging the end in terms of tightening requirements is just a question of deciding who to fail to serve. Do you fail to serve those who came first, or, do you punish those who come first and serve only those who meet some other arbitrary criteria? What arbitrary criteria would you suggest be used to decide who should not receive service?
Furthermore, I expect more credence will be lent to the reclaiming efforts, and pre-RIR swamp space has lots of candidates. What incentive to a holder of early allocations is there to return address space voluntarily?
None, but the nice thing about being a registry is that reclaiming things is as simple as allocating it to somebody else. Buyer beware and all that.
This would create significant legal challenges and costs. Likely, the legacy holders would prevail in many cases, and, you just might find that becomes the excuse that congress needs to hand over management of the IP space to the ITU. I don't see that as a good scenario at all. By the time the legal challenges were resolved, the ISPs receiving such allocations would be long-since out of business due to the inability to provide reliable service to their customers.
And in the absence of any other method of obtaining ipv4, I would expect RIR mebership to push for aggressive reclamation, with policy change to make it worthwhile.
The RIR membership doesn't necessarily have standing to do much about legacy holders other than what the legacy holders themselves choose to agree to. You are assuming that the RIR has power that is, as yet, untested, unproven, and, unlikely.
Class-E, Efforts to redefine class E have stalled because there is simply no way it can be used for anything other than private space.
Amazing that so much effort can go into ipv6 but nobody can spare a few hours per product to remove a couple lines of code?
Different entities and a belief that IPv6 is the correct solution on the part of those in a position to do so. Class E would actually not buy very much time, either.
There are too many implementations out there that will never be modified (e.g., Windows 98) on which you can't even configure class E space.
Faced with a choice of ipv6 and no ipv4 or ipv6 and class-e ipv4, which would you choose? Not like windows98 (if there are any still around that mean anything to anybody) has ipv6 either.
True. So, this will probably create the mandate for W98 to go away. I don't see this as a bad thing. As it stands now, anyone who wants can try and use class-e IPv4. However, I don't expect any RIR to be handing it out with guaranteed uniqueness any time soon.
rfc3330 and similar reclamation might occur as well. IANA recently reclaimed 14/8. I think that added 3 _weeks_ to the expected runout date. That was likely the last "easily" reclaimable block.
Reclamation efforts without policy change isnt profitable and would only become so if miser mode is in effect.
I haven't seen you propose a policy change that would affect this. While you're too late for the Denver meeting, you are welcome to submit a policy to ARIN if you think policy can resolve this.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts.
I expect the value will skyrocket, whether on the black, grey or white market. Yep. And the question is: as an ISP or other address consuming organization, what will you do when the cost of obtaining IPv4 addresses skyrockets?
Pass it on to the customer. Reclaim. Scavenge. Engineer more nats and workarounds while accelerating ipv6. Get budget and manpower to actually make changes. Drag the users kicking and screaming, cause thats what it will take.
Yep... All of these will probably occur.
So far, as far as I can tell, the answer to that question (in most cases) has been putting hands over ears and saying "La la la" loudly. See <http://www.networkworld.com/news/2008/020608-ipv4-address-depletion.html
.
Things will likely be different in 2010
True. The question is not whether things will change, but, how they will change. This is a much harder question. For now, all that we know is that few people are paying attention to the problem, and, that the problem will get progressively harder to solve the longer that behavior persists. Owen
Owen DeLong wrote:
This would create significant legal challenges and costs.
I dont understand how they could prevail. Suppose I decided to tell people that as far as I was concerned this 32 bit number up to that 32bit number would be available for their use in whatever manner they wanted, like printing them on t-shirts, and that they would have the privilege of paying me for the guarantee that I would do so uniquely. Barring a prior agreement, what grounds does any third party have to object? For that matter, aside from consensus and inertia, what would stop the operator community as a whole from setting up shop with a "forked" registry that had no contractual agreements with anybody prior?
On Feb 19, 2008, at 11:08 AM, Joe Maimon wrote:
Owen DeLong wrote:
This would create significant legal challenges and costs.
I dont understand how they could prevail. Suppose I decided to tell people that as far as I was concerned this 32 bit number up to that 32bit number would be available for their use in whatever manner they wanted, like printing them on t-shirts, and that they would have the privilege of paying me for the guarantee that I would do so uniquely.
Barring a prior agreement, what grounds does any third party have to object?
The argument could be made that some form of prior agreement exists that the legacy holders have already received that assignment under those terms and that the RIRs have accepted a burden to preserve that. Thus, your "Barring a prior agreement" condition is not met.
For that matter, aside from consensus and inertia, what would stop the operator community as a whole from setting up shop with a "forked" registry that had no contractual agreements with anybody prior?
Nothing. However, do you really think that is viable? 1. Consensus would be very hard to achieve. 2. Identifying a single entity to manage such a "forked" registry vs. a bunch of islands of registration would be even harder. 3. Much breakage and instability would likely result. 4. Do you have any illusion that this would do anything other than beg government(s) to try and get involved in regulating address space? Owen
Owen DeLong wrote:
Barring a prior agreement, what grounds does any third party have to object?
The argument could be made that some form of prior agreement exists that the legacy holders have already received that assignment under those terms and that the RIRs have accepted a burden to preserve that.
Well I suppose that would be an interesting question that would depend on examining a whole bunch of agreements that may or may not exist and that may or may not have any legal ramification in some jurisidiction or another and that may or may not apply to successors in interest who may or may not actually be successors in interest. Legally speaking, if the registries voluntaried disbanded, thus requiring a new unencumbered entity to rise from the ashes, how could any prior agreement be considered binding?
Thus, your "Barring a prior agreement" condition is not met.
For that matter, aside from consensus and inertia, what would stop the operator community as a whole from setting up shop with a "forked" registry that had no contractual agreements with anybody prior?
Nothing. However, do you really think that is viable? 1. Consensus would be very hard to achieve.
If the current registries are not fulfilling the needs and have become irrelevant, consensus for forking becomes more likely. The only technical lockin I can spot is reverse dns.
2. Identifying a single entity to manage such a "forked" registry vs. a bunch of islands of registration would be even harder.
All such islands would have a vested interest to work together, and thus, they might actually do so, just like all internet network islands do so today.
3. Much breakage and instability would likely result.
Likely not, since any registry wishing to be successfull would be trying their best not to break anything, in other words, RIR allocations would likely be honored/duplicated, but swamp would become fair game.
4. Do you have any illusion that this would do anything other than beg government(s) to try and get involved in regulating address space?
If there is a breakdown in supply, than there will be an opportunity for an entity to step forward, if they can promise supply. If iana free pool runs out and registries cant offer any new ones, whos to say goverments wont start stepping in and "eminent domain"ing address space and setting up registry shop themselves? Consensus is still required. Otherwise its just a national private network, with or without nat. I think the takeaway is that the registries better remain relevant to ipv4 so long as ipv4 is relevant.
Owen
On Feb 19, 2008, at 11:51 AM, Joe Maimon wrote: > > > Owen DeLong wrote: > > >>> Barring a prior agreement, what grounds does any third party have >>> to object? >>> >> The argument could be made that some form of prior agreement >> exists that the >> legacy holders have already received that assignment under those >> terms and that the >> RIRs have accepted a burden to preserve that. > > Well I suppose that would be an interesting question that would > depend on examining a whole bunch of agreements that may or may not > exist and that may or may not have any legal ramification in some > jurisidiction or another and that may or may not apply to successors > in interest who may or may not actually be successors in interest. > > Legally speaking, if the registries voluntaried disbanded, thus > requiring a new unencumbered entity to rise from the ashes, how > could any prior agreement be considered binding? > You'd probably need to disband the following entities without successor arrangements in order to accomplish this: 1. The RIRs 2. IANA/ICANN 3. USDoC The third is extremely unlikely. The second is unlikely. The first is very unlikely. At a certain point, the courts will apply the reasonable and prudent test to the question and likely determine that someone who received an assignment from SRI-NIC or NSI-NIC had a reasonable expectation to be able to use that address space in perpetuity and that whatever registry had reasonable duty not to duplicate said assignment. Very likely any ISP routing the assignment to the new holder would be part of the lawsuit and would get enjoined from doing so. >> Thus, your "Barring a prior agreement" condition is not met. >>> For that matter, aside from consensus and inertia, what would >>> stop the operator community as a whole from setting up shop with >>> a "forked" registry that had no contractual agreements with >>> anybody prior? >> Nothing. However, do you really think that is viable? >> 1. Consensus would be very hard to achieve. > > If the current registries are not fulfilling the needs and have > become irrelevant, consensus for forking becomes more likely. > I think it is unlikely that the current registries will become less relevant that alternative registries. > The only technical lockin I can spot is reverse dns. > That's substantial, but, having multiple registries competing to assign the same addresses in an uncoordinated manner is not likely to be a useful or successful model in any case. >> 2. Identifying a single entity to manage such a "forked" >> registry vs. a bunch >> of islands of registration would be even harder. > > All such islands would have a vested interest to work together, and > thus, they might actually do so, just like all internet network > islands do so today. > How do you see this working? Who would determine which addresses went to which islands for assignment/allocation? >> 3. Much breakage and instability would likely result. > > Likely not, since any registry wishing to be successfull would be > trying their best not to break anything, in other words, RIR > allocations would likely be honored/duplicated, but swamp would > become fair game. > You still haven't explained who would have the authority to divide up the swamp between the competing organizations, all of whom would likely claim full control of the entire swamp, but, certainly there would be overlapping conflicts. >> 4. Do you have any illusion that this would do anything other >> than beg >> government(s) to try and get involved in regulating address >> space? > > If there is a breakdown in supply, than there will be an opportunity > for an entity to step forward, if they can promise supply. > Promising supply is an act of fiction. > If iana free pool runs out and registries cant offer any new ones, > whos to say goverments wont start stepping in and "eminent > domain"ing address space and setting up registry shop themselves? > Could be interesting, indeed. > Consensus is still required. Otherwise its just a national private > network, with or without nat. > Yep. > I think the takeaway is that the registries better remain relevant > to ipv4 so long as ipv4 is relevant. > I agree that is the ideal outcome. The bigger question is how best to achieve this. Owen
On 19 feb 2008, at 21:54, Owen DeLong wrote:
At a certain point, the courts will apply the reasonable and prudent test to the question and likely determine that someone who received an assignment from SRI-NIC or NSI-NIC had a reasonable expectation to be able to use that address space in perpetuity and that whatever registry had reasonable duty not to duplicate said assignment.
I'm sorry to have to say this, but that's all a load of crap. People get their street addresses changed when there is a need. Phone numbers are changed when this is required to keep the numbering plan working. Why would people who by the policies that have been in effect for a decade don't qualify be able to keep using unreasonably large amounts of address space if this blocks others from connecting to the network? ARIN/IANA/whatever should have had the stones to first put a large amount of pressure on the legacy class A holders and then take them to court. Declaring defeat before any action is taken is not a reasonable course of action. Now it's too late, of course: the lawsuits would take years, renumbering too. By the way, I sat down on the couch and turned on the NANOG channel to watch the IPv6 hour, but the video was fairly flakey. What was it that Randy found so cool?
ARIN/IANA/whatever should have had the stones to first put a large amount of pressure on the legacy class A holders and then take them to court. Declaring defeat before any action is taken is not a reasonable course of action.
Not naming names, but I recall one member of an institution of higher education stating at a meeting that from their (his?) perspective, they would give up their class A only if they calculated that the cost of defending against any potential lawsuit (which they expected they would win) would exceed the cost of renumbering (and they had done a back of the envelope calculation that it was in the millions to renumber). This is a clear profit and loss point of view. Why change voluntarily if there is no profit to be made? We can all talk about "the good of the community", but true altruism is rare (although we all want to see it in others).
At 8:20 AM -0800 2/19/08, David Conrad wrote:
What incentive to a holder of early allocations is there to return address space voluntarily?
No incentive at all, unless they don't need them anymore. In that case, RFC2050 applies (which you and Kim and Mark and Daniel and Jon Postel coauthored, right?): "IP addresses are valid as long as the criteria continues to be met. The IANA reserves the right to invalidate any IP assignments once it is determined the the requirement for the address space no longer exists. In the event of address invalidation, reasonable efforts will be made by the appropriate registry to inform the organization that the addresses have been returned to the free pool of IPv4 address space." I understand that quite a few allocations were made prior to RFC 2050, but they certainly don't predates Jon's involvement as the IANA or the original collaborative spirit of the Internet community. To be clear, many of those who received early allocations are making use of them in some manner, and I haven't seen in any policies or proposals any suggestion that organizations actually using addresses shouldn't keep on using them. The discussion has been on why most of those who haven't been making use of their early assignments haven't actually returned them to the free pool (we've had some successes over the last ten years, including the recent address returns that Leo discusses here: http://blog.icann.org/?p=271) but these add up to a very small percentage of the potentially unused space in the early allocations. What we now face is the simply reality of whether enabling a financial incentive for those who don't adhere to the community spirit incentive is overall worthwhile for the Internet. Some will say this has certain unsavory aspects to it (such as discouraging altruistic return of address space) but that shouldn't preclude it from being considered at all. /John
John, On Feb 19, 2008, at 9:17 AM, John Curran wrote:
What we now face is the simply reality of whether enabling a financial incentive for those who don't adhere to the community spirit incentive is overall worthwhile for the Internet.
Not really. I figure "community spirit" as a significant motivator died long, long ago and even if it did continue to exist, it certainly wouldn't survive the upcoming free-for-all in two or three years. The simple reality is that businesses who require IPv4 addresses to continue operations will do what is necessary to obtain them, regardless of what an informational document published over a decade ago or informal agreements with individuals sadly passed away might say. ARIN and the other RIRs can continue to try to ignore that reality, but the almost certain end state of that action is to make ARIN and the other RIRs irrelevant in IPv4 registration management (who would be relevant is left to the reader as an exercise).
Some will say this has certain unsavory aspects to it (such as discouraging altruistic return of address space) but that shouldn't preclude it from being considered at all.
The question really isn't whether or not financial incentives should be considered. They already are (as you well know). The question really should be how (or even if) the existing policy bodies can impose some form of self-regulation to keep the inevitable market behavior from completely running amok. Quoting sections of 2050 written when the Internet was quite different and free IPv4 space was plentiful, particularly given subsequent agreements between ICANN and the RIRs removed IANA from any significant address management role, is merely wasting time. Regards, -drc
At 11:00 AM -0800 2/19/08, David Conrad wrote:
The simple reality is that businesses who require IPv4 addresses to continue operations will do what is necessary to obtain them, regardless of what an informational document published over a decade ago or informal agreements with individuals sadly passed away might say. ARIN and the other RIRs can continue to try to ignore that reality, but the almost certain end state of that action is to make ARIN and the other RIRs irrelevant in IPv4 registration management (who would be relevant is left to the reader as an exercise).
David - I imagine that there are many potential outcomes. For example, in a world where ICANN/IANA (who seems to very much want to be in charge of all this) actually did IP block revocation of unused blocks per RFC2050, we'd likely not be having any discussion of a relaxed transfer policy, as a result of the complete lack of need. Would the ISP community support adherence to RFC 2050 and route accordingly? It certainly has to date, and nearly every RIR policy has been build accordingly. It might result in some legal work, but that's a small price to pay to further operational stability of the Internet. /John p.s. ICANN seems to have no problem with asserting the informal DNS agreements from the same time period (with entire teams of lawyers) so maybe we just need to wait until they're free to pay attention to IP resources? Will that be soon?
John, On Feb 19, 2008, at 11:40 AM, John Curran wrote:
I imagine that there are many potential outcomes. For example, in a world where ICANN/IANA (who seems to very much want to be in charge of all this)
This is almost amusing given the _mutually_ agreed role for ICANN/IANA in address allocation, but I guess we all need our bogeymen.
actually did IP block revocation of unused blocks per RFC2050, we'd likely not be having any discussion of a relaxed transfer policy, as a result of the complete lack of need.
To my knowledge, ICANN has neither the ability (given the mutual agreements between ICANN and the RIRs) nor the desire to intrude upon the RIR's bailiwick in this way. I find it somewhat surprising that after 10 or so years of trying to ensure ICANN didn't get into the RIRs business that you're suggesting they do so now (I seem to remember someone telling me "IANA should be a black box where the RIRs crank the handle and /8s pop out, nothing more"). However, this misses the point. The people who control the Internet address space are not the RIRs or ICANN. Control is vested in the ISPs who decide what is routed or not routed. In effect, the ISPs have agreed to use the RIRs as a neutral meeting point to avoid negotiating a myriad agreements on who has the right to announce what. This only works as long as it is in the best interests of most (in particular, the big parties) to play along. As soon as policies begin to impact those best interests negatively, the policies will either be modified or ignored. If they are ignored, the vacuum will be filled by someone (whether a private entity or governments/ITU isn't clear at this point). A market already exists. Whether it is done by buying/selling a company for its IP address assets or just simply buying the address space outright and paying an ISP to not look at a RIR whois server isn't particularly relevant. As the IPv4 free pool exhausts, that market is going to get much bigger, much faster. It would be nice if this market were somehow self-regulated by the industry players involved since failing that implies something I suspect none of us want. However, the current path appears to be to not do anything until it is too late. Perhaps we could agree that not doing something until it is too late would be bad?
Would the ISP community support adherence to RFC 2050 and route accordingly? It certainly has to date, and nearly every RIR
policy has been build accordingly.
And IPv4 address space has been essentially free to date. That _is_ changing as you well know.
It might result in some legal work, but that's a small price to pay to further operational stability of the Internet.
As far as I can tell, the best way to ensure instability is for ISPs and other address consuming organizations to continue ignoring the issue. Promoting the idea that 2050 is somehow still applicable to the post IPv4 free pool world would seem to support that action. Not sure how that helps.
p.s. ICANN seems to have no problem with asserting the informal DNS agreements from the same time period (with entire teams of lawyers) so maybe we just need to wait until they're free to pay attention to IP resources?
I'm afraid you're confused. The informal agreements I presume you're referring to are national sovereignty issues and ICANN has essentially no role (and certainly no role for ICANN's lawyers). FWIW, ICANN's lawyers are largely consumed in dealing with new, formal, contractual agreements. However, continue to make spurious accusations, it undoubtedly makes the various parties feel better. Regards, -drc
At 3:39 PM -0800 2/19/08, David Conrad wrote:
Perhaps we could agree that not doing something until it is too late would be bad?
Absolutely. Quite a few things are being done (hopefully not too late) including looking into a more flexible transfer policy and encouraging return of IP space which is no longer needed. You asked what "incentive to a holder of early allocations is there to return address space voluntarily?" The first answer ("it's the right thing to do") has clearly worked in some cases). A second answer ("because it's required by RFC 2050 if the addresses are not needed") is also available, presuming that the community wish that to be the rule. As you stated quite clearly, ultimately the control is vested in the ISPs who decide what is routed or not routed. /John
David Conrad wrote:
John,
On Feb 19, 2008, at 11:40 AM, John Curran wrote:
I imagine that there are many potential outcomes. For example, in a world where ICANN/IANA (who seems to very much want to be in charge of all this)
This is almost amusing given the _mutually_ agreed role for ICANN/IANA in address allocation, but I guess we all need our bogeymen.
actually did IP block revocation of unused blocks per RFC2050, we'd likely not be having any discussion of a relaxed transfer policy, as a result of the complete lack of need.
To my knowledge, ICANN has neither the ability (given the mutual agreements between ICANN and the RIRs) nor the desire to intrude upon the RIR's bailiwick in this way. I find it somewhat surprising that after 10 or so years of trying to ensure ICANN didn't get into the RIRs business that you're suggesting they do so now (I seem to remember someone telling me "IANA should be a black box where the RIRs crank the handle and /8s pop out, nothing more").
However, this misses the point.
The people who control the Internet address space are not the RIRs or ICANN. Control is vested in the ISPs who decide what is routed or not routed. In effect, the ISPs have agreed to use the RIRs as a neutral meeting point to avoid negotiating a myriad agreements on who has the right to announce what. This only works as long as it is in the best interests of most (in particular, the big parties) to play along. As soon as policies begin to impact those best interests negatively, the policies will either be modified or ignored. If they are ignored, the vacuum will be filled by someone (whether a private entity or governments/ITU isn't clear at this point).
A market already exists. Whether it is done by buying/selling a company for its IP address assets or just simply buying the address space outright and paying an ISP to not look at a RIR whois server isn't particularly relevant. As the IPv4 free pool exhausts, that market is going to get much bigger, much faster. It would be nice if this market were somehow self-regulated by the industry players involved since failing that implies something I suspect none of us want. However, the current path appears to be to not do anything until it is too late.
Perhaps we could agree that not doing something until it is too late would be bad?
Would the ISP community support adherence to RFC 2050 and route accordingly? It certainly has to date, and nearly every RIR
policy has been build accordingly.
And IPv4 address space has been essentially free to date. That _is_ changing as you well know.
It might result in some legal work, but that's a small price to pay to further operational stability of the Internet.
As far as I can tell, the best way to ensure instability is for ISPs and other address consuming organizations to continue ignoring the issue. Promoting the idea that 2050 is somehow still applicable to the post IPv4 free pool world would seem to support that action. Not sure how that helps.
p.s. ICANN seems to have no problem with asserting the informal DNS agreements from the same time period (with entire teams of lawyers) so maybe we just need to wait until they're free to pay attention to IP resources?
I'm afraid you're confused. The informal agreements I presume you're referring to are national sovereignty issues and ICANN has essentially no role (and certainly no role for ICANN's lawyers). FWIW, ICANN's lawyers are largely consumed in dealing with new, formal, contractual agreements. However, continue to make spurious accusations, it undoubtedly makes the various parties feel better.
Regards, -drc
Agree with David. If it isn't the structural change proposal (scrap the original constituencies for contractual and non-contractual, with subdivisions within contractual for type of contract, and within non-contractual for type of non-contractual), then no one who cares about the balance of power within the GNSO will give it a moment's thought. If it isn't IDN (more personal caveats than most), then no one who cares about ICANN's next couple of years will give it a moment's thought. So, no. As fun as the v4 pool looks to be, the g-side, the cc-side, and the more-bits-than-7-side have wicked bigger fish to fry. Have fun, Eric
Policy, Regulations, Community support and ,yes that word again, Market all factored in there are some things that would be noteworthy to look at 1. Where is the current demand for IPv4 coming from? Plenty of analysis here. 2. Can the new demand be steered towards IPv6 therefore slow down IPv4 depletion. For example in the developing world a lot of the demand is brand new and therefore can it be satisfied with IPv6? Of course incentives are in order though it's much easier than asking the behemoths to give up "their" /16s etc. Best Regards Raymond Macharia -----Original Message----- From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On Behalf Of Eric Brunner-Williams Sent: 2008-02-20 05:20 To: David Conrad Cc: John Curran; Nanog Subject: Re: IPV4 as a Commodity for Profit David Conrad wrote:
John,
On Feb 19, 2008, at 11:40 AM, John Curran wrote:
I imagine that there are many potential outcomes. For example, in a world where ICANN/IANA (who seems to very much want to be in charge of all this)
This is almost amusing given the _mutually_ agreed role for ICANN/IANA in address allocation, but I guess we all need our bogeymen.
actually did IP block revocation of unused blocks per RFC2050, we'd likely not be having any discussion of a relaxed transfer policy, as a result of the complete lack of need.
To my knowledge, ICANN has neither the ability (given the mutual agreements between ICANN and the RIRs) nor the desire to intrude upon the RIR's bailiwick in this way. I find it somewhat surprising that after 10 or so years of trying to ensure ICANN didn't get into the RIRs business that you're suggesting they do so now (I seem to remember someone telling me "IANA should be a black box where the RIRs crank the handle and /8s pop out, nothing more").
However, this misses the point.
The people who control the Internet address space are not the RIRs or ICANN. Control is vested in the ISPs who decide what is routed or not routed. In effect, the ISPs have agreed to use the RIRs as a neutral meeting point to avoid negotiating a myriad agreements on who has the right to announce what. This only works as long as it is in the best interests of most (in particular, the big parties) to play along. As soon as policies begin to impact those best interests negatively, the policies will either be modified or ignored. If they are ignored, the vacuum will be filled by someone (whether a private entity or governments/ITU isn't clear at this point).
A market already exists. Whether it is done by buying/selling a company for its IP address assets or just simply buying the address space outright and paying an ISP to not look at a RIR whois server isn't particularly relevant. As the IPv4 free pool exhausts, that market is going to get much bigger, much faster. It would be nice if this market were somehow self-regulated by the industry players involved since failing that implies something I suspect none of us want. However, the current path appears to be to not do anything until it is too late.
Perhaps we could agree that not doing something until it is too late would be bad?
Would the ISP community support adherence to RFC 2050 and route accordingly? It certainly has to date, and nearly every RIR
policy has been build accordingly.
And IPv4 address space has been essentially free to date. That _is_ changing as you well know.
It might result in some legal work, but that's a small price to pay to further operational stability of the Internet.
As far as I can tell, the best way to ensure instability is for ISPs and other address consuming organizations to continue ignoring the issue. Promoting the idea that 2050 is somehow still applicable to the post IPv4 free pool world would seem to support that action. Not sure how that helps.
p.s. ICANN seems to have no problem with asserting the informal DNS agreements from the same time period (with entire teams of lawyers) so maybe we just need to wait until they're free to pay attention to IP resources?
I'm afraid you're confused. The informal agreements I presume you're referring to are national sovereignty issues and ICANN has essentially no role (and certainly no role for ICANN's lawyers). FWIW, ICANN's lawyers are largely consumed in dealing with new, formal, contractual agreements. However, continue to make spurious accusations, it undoubtedly makes the various parties feel better.
Regards, -drc
Agree with David. If it isn't the structural change proposal (scrap the original constituencies for contractual and non-contractual, with subdivisions within contractual for type of contract, and within non-contractual for type of non-contractual), then no one who cares about the balance of power within the GNSO will give it a moment's thought. If it isn't IDN (more personal caveats than most), then no one who cares about ICANN's next couple of years will give it a moment's thought. So, no. As fun as the v4 pool looks to be, the g-side, the cc-side, and the more-bits-than-7-side have wicked bigger fish to fry. Have fun, Eric
On 20 feb 2008, at 8:13, Raymond Macharia wrote:
1. Where is the current demand for IPv4 coming from? Plenty of analysis here.
Hm, let's see: mysql> select rir, country, descr, num, day from addrspace where day
= 20060101 and type = 'ipv4' order by num desc limit 5; +---------+---------+-------------+---------+------------+ | rir | country | descr | num | day | +---------+---------+-------------+---------+------------+ | ripencc | FR | 90.0.0.0 | 8388608 | 2006-03-02 | | apnic | CN | 116.128.0.0 | 4194304 | 2007-05-11 | | apnic | CN | 117.128.0.0 | 4194304 | 2007-07-17 | | apnic | IN | 117.192.0.0 | 4194304 | 2007-08-01 | | arin | US | 75.0.0.0 | 4194304 | 2006-02-28 | +---------+---------+-------------+---------+------------+
inetnum: 90.0.0.0 - 90.0.0.255 role: Wanadoo France Technical Role address: FRANCE TELECOM/SCR inetnum: 116.128.0.0 - 116.191.255.255 descr: New Guoxin Telecom Corporation inetnum: 117.128.0.0 - 117.191.255.255 descr: China Mobile Communications Corporation inetnum: 117.192.0.0 - 117.255.255.255 descr: NIB (National Internet Backbone) AT&T Internet Services SBCIS-SBIS-6BLK (NET-75-0-0-0-1) 75.0.0.0 - 75.63.255.255 Answer: ISPs/telcos.
2. Can the new demand be steered towards IPv6 therefore slow down IPv4 depletion. For example in the developing world a lot of the demand is brand new and therefore can it be satisfied with IPv6? Of course incentives are in order though it's much easier than asking the behemoths to give up "their" /16s etc.
Well, you could give IPv6 instead of IPv4 but only if they can talk to the rest of the world over IPv6. So everyone who has IPv4 today would have to add IPv6, but if they were prepared to do that, running out of IPv4 space wouldn't be an issue in the first place. Also, who cares about /16s? We currently have 16848 /16s left and we're using up approximately 8 - 9 of them each day. Postponing the inevitable by 3 hours really isn't all that useful... Can we please finish off the remaining IPv4 address space already so we can get on with our lives?
What incentive to a holder of early allocations is there to return address space voluntarily?
Zero. In fact there is a disincentive since it costs money to return address space. Most of these organizations will not have central control and management of IPv4 addresses with an automated system that can clearly and confidently report what addresses are not needed. Only ISPs implement systems like this and not all of them do so. Most organizations have pieced together spreadsheets, scripts, network discovery tools and departmental IP address management systems to do the job. The effort required to just get a clear and trustable report of which addresses are surplus will cost a considerable amount of money. This all comes at a time when the same resources (cash and people's time) should be working on planning how and where to deploy IPv6 within the organization for the greatest benefit and least disruption. The potential losses risked by a company for not being able to deploy IPv6 when and where they need it, are far greater than the potential benefits of selling a few IP address blocks. Even if you are talking about figures of $500,000 dollars for a /16 block, you still have to deduct the cost of figuring out that this /16 can be released without harming the organization's network. There might not be much profit left. This whole issue (IPv4 depletion) is likely to be written up in one or more B-school cases because this is classic MBA territory. There are no right answers. There is no solution to the problem. There are only courses of action which carry less risk and lead to more profitable outcomes. In my opinion, none of those courses of action includes participation in an IPv4 address market.
Efforts to redefine class E have stalled because there is simply no way it can be used for anything other than private space. There are too many implementations out there that will never be modified (e.g., Windows 98) on which you can't even configure class E space.
Class E could have been used if we had started a year ago. But now it is simply to late to do anything about it. The window of opportunity has passed. I'd still support moving Class E space to being ordinary unicast IPv4 space because it makes no sense to keep it special, but I don't believe that anyone will be able to use these addresses until well after we are into the mixed IPv6/IPv4 Internet phase.
IANA recently reclaimed 14/8. I think that added 3 _weeks_ to the expected runout date. That was likely the last "easily" reclaimable block.
Easily!? I'm not so sure that is a correct description of the process. Perhaps Leo Vegoda could comment on this point since he had more first hand involvement than I did. --Michael Dillon
On 19/02/2008 09:57, "michael.dillon@bt.com" <michael.dillon@bt.com> wrote: [...]
Easily!? I'm not so sure that is a correct description of the process. Perhaps Leo Vegoda could comment on this point since he had more first hand involvement than I did.
On the whole, these assignments were for one or two /32s. They were unlikely to be useful for anything else. The one network that renumbered didn't have a problem finding a couple of spare /32s from their RIR space. I think I would have seen significantly more difficulties if the assignments had mostly been used or been for /24 or shorter prefixes. The difficulty I experienced in reclaiming 14/8 was finding out who to contact. Once I'd found someone who knew about the addresses there wasn't really any resistance to returning them. That being said, the process took about a year of calendar time, which is more than I had expected. I have been looking at other /8s for about the same time with less success. Regards, Leo
David Conrad wrote:
Joe,
On Feb 19, 2008, at 4:28 AM, Joe Maimon wrote:
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly.
No.
...
The RIR bureaucracy is a ponderous ship that turns very slowly and has multiple captains who do not necessarily agree on the direction to turn. IPv4 allocation policy revisions aren't going to save us.
A collaborative bottom-up consensus-based policy determination framework is a ponderous ship that turns ... [etc]. The problem is not necessarily in the machine room that implements these address allocation policies but in the process of determining policies that all interested parties can live with. It takes time. Probably more time than you have left. So even if there are a flurry of last minute policy proposals to salvage the situation it may well be a case of too little too late.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts.
I would suggest that the real question is "How will industry deal with the situation when the current supply streams for IPv4 vaporize?" And the secondary question is "Will the industry's reaction to this shift in the supply of addresses destroy the integrity and utility of the entire IPv4 space?" What I'm gettting at is that if there is no mechanism in whatever industry does for address supply after the unallocated pool exhausts to preserve the essential attribute of the address system, namely uniqueness of use, and we start to see competing claims to be able to use addresses without any agreed framework of resolution, then what happens to the Internet? Do we all just originate whatever addresses we feel like on the day in to the routing system?
Yep. And the question is: as an ISP or other address consuming organization, what will you do when the cost of obtaining IPv4 addresses skyrockets? So far, as far as I can tell, the answer to that question (in most cases) has been putting hands over ears and saying "La la la" loudly. See <http://www.networkworld.com/news/2008/020608-ipv4-address-depletion.html>.
indeed. Geoff
I would like to encourage everyone engaging in this discussion here to move the discussion over to the ARIN PPML and talk about the policy. It's great to talk about it on NANOG , and, indeed, many members of the ARIN AC are on this list. However, the rest of the ARIN community should also see your comments, and, that is technically the correct forum for addressing ARIN policy matters. Owen On Feb 19, 2008, at 2:42 PM, Geoff Huston wrote:
David Conrad wrote:
Joe, On Feb 19, 2008, at 4:28 AM, Joe Maimon wrote:
When IANA free pool exhaustion happens or even appears to be imminent, one can expect push for allocation policies to be changed drastically towards the miserly. No. ...
The RIR bureaucracy is a ponderous ship that turns very slowly and has multiple captains who do not necessarily agree on the direction to turn. IPv4 allocation policy revisions aren't going to save us.
A collaborative bottom-up consensus-based policy determination framework is a ponderous ship that turns ... [etc]. The problem is not necessarily in the machine room that implements these address allocation policies but in the process of determining policies that all interested parties can live with. It takes time. Probably more time than you have left.
So even if there are a flurry of last minute policy proposals to salvage the situation it may well be a case of too little too late.
The question is how ARIN will deal with the market after the IPv4 free pool exhausts.
I would suggest that the real question is "How will industry deal with the situation when the current supply streams for IPv4 vaporize?"
And the secondary question is "Will the industry's reaction to this shift in the supply of addresses destroy the integrity and utility of the entire IPv4 space?"
What I'm gettting at is that if there is no mechanism in whatever industry does for address supply after the unallocated pool exhausts to preserve the essential attribute of the address system, namely uniqueness of use, and we start to see competing claims to be able to use addresses without any agreed framework of resolution, then what happens to the Internet? Do we all just originate whatever addresses we feel like on the day in to the routing system?
Yep. And the question is: as an ISP or other address consuming organization, what will you do when the cost of obtaining IPv4 addresses skyrockets? So far, as far as I can tell, the answer to that question (in most cases) has been putting hands over ears and saying "La la la" loudly. See <http://www.networkworld.com/news/2008/020608-ipv4-address-depletion.html
.
indeed.
Geoff
On Mon, 2008-02-18 at 17:05 +0000, Rod Beck wrote:
Markets have proven to be excellent mechanisms for allocating resources fairness is a distinct issue) and might be the medication required given the apparent hoarding of IP addresses.
Growth is king, also in networking. How can a v4 market meet the demand of an expanding global network beyond a short-lived gold-rush? A price-tag may create an incentive to sell, but doesn't create more units or magically solve other problems (e.g. fragmentation). Many are those who look forward to a v4 market. Not to invest in in, but because will be the most powerful catalyst driving the transition to v6. //per
Per Heldal wrote:
Growth is king, also in networking. How can a v4 market meet the demand of an expanding global network beyond a short-lived gold-rush? A price-tag may create an incentive to sell, but doesn't create more units or magically solve other problems (e.g. fragmentation). Many are those who look forward to a v4 market. Not to invest in in, but because will be the most powerful catalyst driving the transition to v6.
I personally agree with all that you say, but it doesn't mean that a market isn't useful. In particular, can it be useful in a transition from IPv4 to IPv6 to those who are not in a position to easily move from one to the other? They would pay a premium to move based on scarcity already, but if there is no motivation to bring unused blocks into the market, then they won't show up. And that is sufficient motivation for a black market, a market that governments themselves couldn't play a constructive role in (buying OR selling). Eliot
Per, On Feb 19, 2008, at 12:22 AM, Per Heldal wrote:
A price-tag may create an incentive to sell, but doesn't create more units or magically solve other problems (e.g. fragmentation).
It doesn't create more units, but it does increase the incentive to find ways to be more efficient in use. Does MIT really need a /8? Does InterOp? Does HP need 2 plus a bunch of /16s? Etc. In the extreme, does any reasonably sized organization really _need_ more than a few /32s (which could be allocated out of PA space thereby reducing fragmentation) for their NAT gateway and public facing servers? How many ISPs still allocate from a small set of fixed size block to customers regardless of what the customers actually _need_, simply because that's what their backend systems were written to do?
Many are those who look forward to a v4 market. Not to invest in in, but because will be the most powerful catalyst driving the transition to v6.
That's the optimistic view... Regards, -drc
On Tue, 19 Feb 2008 07:56:44 PST, David Conrad said:
Does InterOp? Does HP need 2 plus a bunch of /16s? Etc. In the extreme, does any reasonably sized organization really _need_ more than a few /32s (which could be allocated out of PA space thereby reducing fragmentation) for their NAT gateway and public facing servers?
Only if you think "everybody hiding behind a NAT" is workable - in particular, if you've got enough hosts that you're using up most of 2 /16s, you probably have enough machines that might want an *inbound* SYN packet once in a while that NAT isn't a really good idea. Maybe if you're a corporation where 98% of the machines really shouldn't be accessing the Internet at all *anyhow*, but if you're someplace where "Yes, you can get to the Internet" is policy, you shouldn't be getting into the "but we're not letting you to *all* of the Internet, just the things NAT works with" food-fight.
participants (20)
-
Brandon Galbraith -
Buhrmaster, Gary -
David Conrad -
Eliot Lear -
Eric Brunner-Williams -
Geoff Huston -
Iljitsch van Beijnum -
Joe Maimon -
John Curran -
John Lee -
Leo Vegoda -
michael.dillon＠bt.com -
Owen DeLong -
Per Heldal -
Raymond Macharia -
Rod Beck -
Stephane Bortzmeyer -
Steven M. Bellovin -
Tom Vest -
Valdis.Kletnieks＠vt.edu