Hello I support this proposal. If you are worry about running out of IPv4, you should stop those are registering new LIRs, transfer the IPs and close the LIR. Even 2015-01 is unable to stop them. If it is limited to those LIRs who didn't transfer any IP to other LIRs, I think it could reduce amount of unnecessary transfers. I heard a lot about this: "New entrants should have access to IPv4". I think you should say that in this way: " The new entrants should have access to the INTERNET". Don't you think the new LIRs need to run IPv6? or imagine we are in 2020 and we ran out of IPv4, what should we do? Do it now. I think we are just postpone switching to v6. -- Shahin Gharghi
On 10/21/2015 1:26 PM, Shahin Gharghi wrote:
Hello
I support this proposal. If you are worry about running out of IPv4, you should stop those are registering new LIRs, transfer the IPs and close the LIR. Even 2015-01 is unable to stop them. If it is limited to those LIRs who didn't transfer any IP to other LIRs, I think it could reduce amount of unnecessary transfers. I heard a lot about this: "New entrants should have access to IPv4". I think you should say that in this way: " The new entrants should have access to the INTERNET". Don't you think the new LIRs need to run IPv6? or imagine we are in 2020 and we ran out of IPv4, what should we do? Do it now. I think we are just postpone switching to v6.
If RIPE runs out of IPv4 sooner it will not help switching to v6 faster. It only means that the new entrants will have to buy the resources from the market at prices which will be obviously higher. Iran is a good example, as a country can be considered a relatively new entrant. What has happened over there ? 0.05% IPv6 adoption rate according to google stats and Iran is the #1 importing country of IPv4 resources. I don't understand why there's still this confusion that if RIPE's pool will be empty, many think there will be no more IPv4 available and everyone will go the next day to IPv6. It's totally wrong. Ciprian Nica
Hi, I would support this proposal with some amendments, first there should be a 'top-cap' on how many IPs a LIR can have and still request a new /22 every 18 months - for example if a LIR has an /20 equivalent or more already assigned this LIR could not request new allocation (this has already been pointed out on the list) and I think that there should also be a minimum RIPE free space cap, where if we would reach somewhere in the range of /12 only new LIRs are eligible to get new /22 allocation. This is to get new entrants in the market some more wiggle room against the old players. Sincerely, Uros On Wed, Oct 21, 2015 at 12:40 PM, Ciprian Nica <office@ip-broker.uk> wrote:
Hello
I support this proposal. If you are worry about running out of IPv4, you should stop those are registering new LIRs, transfer the IPs and close
On 10/21/2015 1:26 PM, Shahin Gharghi wrote: the
LIR. Even 2015-01 is unable to stop them. If it is limited to those LIRs who didn't transfer any IP to other LIRs, I think it could reduce amount of unnecessary transfers. I heard a lot about this: "New entrants should have access to IPv4". I think you should say that in this way: " The new entrants should have access to the INTERNET". Don't you think the new LIRs need to run IPv6? or imagine we are in 2020 and we ran out of IPv4, what should we do? Do it now. I think we are just postpone switching to v6.
If RIPE runs out of IPv4 sooner it will not help switching to v6 faster. It only means that the new entrants will have to buy the resources from the market at prices which will be obviously higher.
Iran is a good example, as a country can be considered a relatively new entrant. What has happened over there ? 0.05% IPv6 adoption rate according to google stats and Iran is the #1 importing country of IPv4 resources.
I don't understand why there's still this confusion that if RIPE's pool will be empty, many think there will be no more IPv4 available and everyone will go the next day to IPv6. It's totally wrong.
Ciprian Nica
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry) On Wed, Oct 21, 2015 at 12:14 PM, Uros Gaber <uros@ub330.net> wrote:
Hi,
I would support this proposal with some amendments, first there should be a 'top-cap' on how many IPs a LIR can have and still request a new /22 every 18 months - for example if a LIR has an /20 equivalent or more already assigned this LIR could not request new allocation (this has already been pointed out on the list) and I think that there should also be a minimum RIPE free space cap, where if we would reach somewhere in the range of /12 only new LIRs are eligible to get new /22 allocation.
This is to get new entrants in the market some more wiggle room against the old players.
Sincerely, Uros
On Wed, Oct 21, 2015 at 12:40 PM, Ciprian Nica <office@ip-broker.uk> wrote:
Hello
I support this proposal. If you are worry about running out of IPv4, you should stop those are registering new LIRs, transfer the IPs and close
On 10/21/2015 1:26 PM, Shahin Gharghi wrote: the
LIR. Even 2015-01 is unable to stop them. If it is limited to those LIRs who didn't transfer any IP to other LIRs, I think it could reduce amount of unnecessary transfers. I heard a lot about this: "New entrants should have access to IPv4". I think you should say that in this way: " The new entrants should have access to the INTERNET". Don't you think the new LIRs need to run IPv6? or imagine we are in 2020 and we ran out of IPv4, what should we do? Do it now. I think we are just postpone switching to v6.
If RIPE runs out of IPv4 sooner it will not help switching to v6 faster. It only means that the new entrants will have to buy the resources from the market at prices which will be obviously higher.
Iran is a good example, as a country can be considered a relatively new entrant. What has happened over there ? 0.05% IPv6 adoption rate according to google stats and Iran is the #1 importing country of IPv4 resources.
I don't understand why there's still this confusion that if RIPE's pool will be empty, many think there will be no more IPv4 available and everyone will go the next day to IPv6. It's totally wrong.
Ciprian Nica
-- Kindest regards, Tom Smyth Mobile: +353 87 6193172 --------------------------------- PLEASE CONSIDER THE ENVIRONMENT BEFORE YOU PRINT THIS E-MAIL This email contains information which may be confidential or privileged. The information is intended solely for the use of the individual or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this electronic transmission in error, please notify me by telephone or by electronic mail immediately. Any opinions expressed are those of the author, not the company's .This email does not constitute either offer or acceptance of any contractually binding agreement. Such offer or acceptance must be communicated in writing. You are requested to carry out your own virus check before opening any attachment. Thomas Smyth accepts no liability for any loss or damage which may be caused by malicious software or attachments.
On 10/21/2015 2:20 PM, Tom Smyth wrote:
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie
that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry)
So if some LIR needed IPs desperately enough to pay money for them, they don't deserve to receive a free allocation from RIPE. <s>Yes, that would make everyone love this policy.</s> Ciprian
My point was that if people have used mechanisims such as new lir +transfer /merge then they would not qualify for an additional alocation ... which in my opinion is fair enough...and would still conserve ip address space for new lirs in future ... Do you love it now ;) ? On 21 Oct 2015 12:28, "Ciprian Nica" <office@ip-broker.uk> wrote:
On 10/21/2015 2:20 PM, Tom Smyth wrote:
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie
that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry)
So if some LIR needed IPs desperately enough to pay money for them, they don't deserve to receive a free allocation from RIPE.
<s>Yes, that would make everyone love this policy.</s>
Ciprian
The way you said it was like you were referring to anyone that bought IPs from the market. What you are proposing is impossible to implement and still won't save this policy. It's difficult to find a "better fairness" today and I don't see a way this policy would help anyone except a few small...ish providers. On 10/21/2015 2:33 PM, Tom Smyth wrote:
My point was that if people have used mechanisims such as new lir +transfer /merge then they would not qualify for an additional alocation ... which in my opinion is fair enough...and would still conserve ip address space for new lirs in future ... Do you love it now ;) ? On 21 Oct 2015 12:28, "Ciprian Nica" <office@ip-broker.uk> wrote:
On 10/21/2015 2:20 PM, Tom Smyth wrote:
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie
that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry)
So if some LIR needed IPs desperately enough to pay money for them, they don't deserve to receive a free allocation from RIPE.
<s>Yes, that would make everyone love this policy.</s>
Ciprian
Point taken..Ciprian, however, just because a policy is difficult to police doensnt mean that it is not a valid policy... and i think practical policing options should be considered so that this policy will help isps, without rewarding isps that have perhaps being abusing loopholes. I think this would be helpful in aleviating IP conservation Concerns On Wed, Oct 21, 2015 at 1:00 PM, Ciprian Nica <office@ip-broker.uk> wrote:
The way you said it was like you were referring to anyone that bought IPs from the market.
What you are proposing is impossible to implement and still won't save this policy. It's difficult to find a "better fairness" today and I don't see a way this policy would help anyone except a few small...ish providers.
On 10/21/2015 2:33 PM, Tom Smyth wrote:
My point was that if people have used mechanisims such as new lir +transfer /merge then they would not qualify for an additional alocation ... which in my opinion is fair enough...and would still conserve ip address space for new lirs in future ... Do you love it now ;) ? On 21 Oct 2015 12:28, "Ciprian Nica" <office@ip-broker.uk> wrote:
On 10/21/2015 2:20 PM, Tom Smyth wrote:
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie
that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry)
So if some LIR needed IPs desperately enough to pay money for them, they don't deserve to receive a free allocation from RIPE.
<s>Yes, that would make everyone love this policy.</s>
Ciprian
-- Kindest regards, Tom Smyth Mobile: +353 87 6193172 --------------------------------- PLEASE CONSIDER THE ENVIRONMENT BEFORE YOU PRINT THIS E-MAIL This email contains information which may be confidential or privileged. The information is intended solely for the use of the individual or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this electronic transmission in error, please notify me by telephone or by electronic mail immediately. Any opinions expressed are those of the author, not the company's .This email does not constitute either offer or acceptance of any contractually binding agreement. Such offer or acceptance must be communicated in writing. You are requested to carry out your own virus check before opening any attachment. Thomas Smyth accepts no liability for any loss or damage which may be caused by malicious software or attachments.
On 10/21/2015 3:08 PM, Tom Smyth wrote:
Point taken..Ciprian, however, just because a policy is difficult to police doensnt mean that it is not a valid policy...
If it's difficult yes, but in my oppinion it would be impossible.
and i think practical policing options should be considered so that this policy will help isps, without rewarding isps that have perhaps being abusing loopholes. I think this would be helpful in aleviating IP conservation Concerns
Well, I've been working for ISPs and took care of the relation with ripe since about year 2000 so I don't think policies can ever be perfect. There will always be some that would find loopholes and exploit them. We should focus on what's really efficient. How to promote IPv6 deployment and how to convince the large operators to take the first steps towards it. According to some recent data (few days ago) that we have, there are 11668 organisations that hold ALLOCATED PA IPv4 resources in RIPE region. The number of LIRs is larger because some of them have already sold the resources so they are no longer relevant.
From theese, the top 1% PA resource holders (117 organisations) have allocations totalling 363,535,872 IPs out of the total 575,180,544 ALLOCATED PA IPs in RIPE region. (that is 63.20%)
If the remaining 99% percent would fully deploy IPv6 then the region's average would be at 36.80%, which would not be enough to make us forget about IPv4. Therefore, in my oppinion, this is not a problem that can be solved by the many but by the few top providers. Think about policies that would "drive them to drive us all" towards IPv6 adoption. That would really help. Ciprian
Am 21.10.2015 um 14:20 schrieb Ciprian Nica:
From theese, the top 1% PA resource holders (117 organisations) have allocations totalling 363,535,872 IPs out of the total 575,180,544 ALLOCATED PA IPs in RIPE region. (that is 63.20%)
If the remaining 99% percent would fully deploy IPv6 then the region's average would be at 36.80%, which would not be enough to make us forget about IPv4.
Therefore, in my oppinion, this is not a problem that can be solved by the many but by the few top providers. Think about policies that would "drive them to drive us all" towards IPv6 adoption. That would really help.
That's exactly my point. The current policy is mostly against new providers (I know many think it's a policy to help them). What about my previous suggestion, like a policy to force ipv4 space holders to return ex. 10% of their ipv4 per year. Money/ effort involved with the transition is no valid concern against it, because the same would hold true for the /21 limit for new LIRs. Corin
That's exactly my point. The current policy is mostly against new providers (I know many think it's a policy to help them). What about my previous suggestion, like a policy to force ipv4 space holders to return ex. 10% of their ipv4 per year. Money/ effort involved with the transition is no valid concern against it, because the same would hold true for the /21 limit for new LIRs.
I would support something like this but with a few changes. I would set some milestones, let's say by the end of 2016 you need to have 5% IPv6 adoption rate or you have to return 5% from the IPs that were allocated to you before the end of 2006 (so it would be a 10 year frame). Then at the end of 2017 they would need to have 10% IPv6 adoption rate or return 10% of the IPs allocated before the end of 2007 and so on. Something like this would push the "old" ISPs to make the first steps towards IPv6 and would not affect the relatively new entrants whom would not make sense returning 50-100 IPs from their /22. I would support this, although I know it would be very difficult to ask Telefonica to return Orange or Deutsche Telekom return millions of IPs. RIPE NCC is not the police but we can all think of a way to implement this, if the community would support it. Ciprian
Am 21.10.2015 um 14:40 schrieb Ciprian Nica:
I would support something like this but with a few changes. I would set some milestones, let's say by the end of 2016 you need to have 5% IPv6 adoption rate or you have to return 5% from the IPs that were allocated to you before the end of 2006 (so it would be a 10 year frame). Then at the end of 2017 they would need to have 10% IPv6 adoption rate or return 10% of the IPs allocated before the end of 2007 and so on.
What about a new policy proposal for that? Please let me know if you'd like to work on one together.
Something like this would push the "old" ISPs to make the first steps towards IPv6 and would not affect the relatively new entrants whom would not make sense returning 50-100 IPs from their /22.
The last /22 (or even /21, which everybody can get then easily) shouldn't be affected.
I would support this, although I know it would be very difficult to ask Telefonica to return Orange or Deutsche Telekom return millions of IPs. RIPE NCC is not the police but we can all think of a way to implement this, if the community would support it.
I'm no lawyer, but from a technical point of view it shouldn't be a problem for RIPE to disconnect them. Any I'm quite sure they'll act before that happens. Corin
On 10/21/2015 4:05 PM, Netskin NOC wrote:
Am 21.10.2015 um 14:40 schrieb Ciprian Nica:
I would support something like this but with a few changes. I would set some milestones, let's say by the end of 2016 you need to have 5% IPv6 adoption rate or you have to return 5% from the IPs that were allocated to you before the end of 2006 (so it would be a 10 year frame). Then at the end of 2017 they would need to have 10% IPv6 adoption rate or return 10% of the IPs allocated before the end of 2007 and so on.
What about a new policy proposal for that? Please let me know if you'd like to work on one together.
If there is support for the idea, then we can come up with a proposal after making a thorough analysis and think of some efficient percentages, terms, etc.
Something like this would push the "old" ISPs to make the first steps towards IPv6 and would not affect the relatively new entrants whom would not make sense returning 50-100 IPs from their /22.
The last /22 (or even /21, which everybody can get then easily) shouldn't be affected.
The idea is to promote IPv6 adoption but if we (the community) would get back enough resources then we can come up with policies to increase the "minimum guaranteed IPv4 slice" that gets out to everyone.
I would support this, although I know it would be very difficult to ask Telefonica to return Orange or Deutsche Telekom return millions of IPs. RIPE NCC is not the police but we can all think of a way to implement this, if the community would support it.
I'm no lawyer, but from a technical point of view it shouldn't be a problem for RIPE to disconnect them. Any I'm quite sure they'll act before that happens.
RIPE NCC can only do what RIPE decides but obviously it has to be legal so we'll have to consult on all aspects. Ciprian
On Wed, Oct 21, 2015, at 13:33, Tom Smyth wrote:
My point was that if people have used mechanisims such as new lir +transfer /merge then they would not qualify for an additional alocation ... which in my opinion is fair enough...and would still conserve ip address space for new lirs in future ... Do you love it now ;) ?
The issue of "multiple LIRs abuse" is much more complicated. If it isn't solved it's because it's too complex. If we take into account mergers and acquisitions (which is a policy in its own) things get even more complex. That part is more related to business processes than anything else. But I do agree that it's something that should be fixed at some point. -- Radu-Adrian FEURDEAN fr.ccs
I think it would be reasonable that if an entity has merged from another lir... they have already recieved one or more /22s over and above what ripe intended. So these entities have already benefited from gaining additional ips So it would be fair to exclude such lirs from getting another /22 under this policy proposal I would imagine that the merged lir option is straightforward to police ? Furthermore Is there any way straightforward way to determine the maximum impact of this policy if every lir in ripe was to request an additional /22 ? On 22 Oct 2015 07:16, "Radu-Adrian FEURDEAN" < ripe-wgs@radu-adrian.feurdean.net> wrote:
On Wed, Oct 21, 2015, at 13:33, Tom Smyth wrote:
My point was that if people have used mechanisims such as new lir +transfer /merge then they would not qualify for an additional alocation ... which in my opinion is fair enough...and would still conserve ip address space for new lirs in future ... Do you love it now ;) ?
The issue of "multiple LIRs abuse" is much more complicated. If it isn't solved it's because it's too complex. If we take into account mergers and acquisitions (which is a policy in its own) things get even more complex. That part is more related to business processes than anything else. But I do agree that it's something that should be fixed at some point.
-- Radu-Adrian FEURDEAN fr.ccs
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Good evening, On 22/10/2015 09:48, Tom Smyth wrote:
I think it would be reasonable that if an entity has merged from another lir... they have already recieved one or more /22s over and above what ripe intended. So these entities have already benefited from gaining additional ips
So it would be fair to exclude such lirs from getting another /22 under this policy proposal
I don't, because there is a confusion here between the LIR and the End User, the LIR being limited to a /22 whereas several End Users having the same LIR would be legitimate to have, say, a /24. So, some level of need-based criteria should not have been abandonned here and the trivial case where a LIR is a single entity that uses all of its allocations for itself should be questionned, since the PA and assignment system is clearly made to distinguish LIR and operator and is supposed to be the normal case. Therefore, creating and merging LIRs can be a very legitimate way to allow new entities to get a minimal IPv4 space, regarding the spirit of the last /8 policiy. Also, policies should not be studied from the only point of view of a LIR but also from the End User's viewpoint, which could be a much more legitimate approach. Best regards, S. Vallerot - -- http://www.opdop.fr - mutualiser et interconnecter en coopérative Opdop - Société Coopérative d'Interêt Collectif sous forme de SARL sur IRC réseau geeknode #opdop - tél: 0950 31 54 74, 06 86 38 38 68 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iF4EAREIAAYFAlYpR2QACgkQJBGsD8mtnRFbcwD9FYxfB1xUzWiJzIljySVPOJMi g5za8bCmCAvlFzzUJv4A/jKSpFup9xH/J+XlqNgN1ZuS6f1/4j9f8pfFO/kYr43X =K42N -----END PGP SIGNATURE-----
Hello Everybuddy, I propose we stop assigning IPv4 addresses, like ARIN did, even if we still have some in the pool. That way the whole policy issue will just go away. Having no policy is the best policy. Yours insincerely, -- Aleksi Suhonen () ascii ribbon campaign /\ support plain text e-mail
Dear chairs of the address policy working group, While I applaud your willingness to provide a platform to any and all policy proposal regardless of its merit, I must now strongly encourage you to start reconsidering this approach or to start actively moderating discussions on this mailing list. I'm appalled by the level of discussion, the ad hom attacks, fallacious argumentation (see http://www.don-lindsay-archive.org/skeptic/arguments.html for an exhaustive list), a lack of constructive attitude from both the proposers and respondents and the almost absent attempts to keep discussions focused. Given that this email list is the foundation of our fabled community-driven bottom-up policy process - and frankly, one of our few arguments to keep top-down decision makers out of numbers policy, I expect some standards to be maintained, and in my opinion they have been slipping, badly. Hoping for short term improvement, Remco (no hats) (and no Aleksi, I'm sure you had no ill intentions in your email, it was just the final straw for me)
Hello, On 10/23/2015 04:08 PM, remco van mook wrote:
I'm appalled by the level of discussion, the ad hom attacks, fallacious argumentation, a lack of constructive attitude from both the proposers and respondents and the almost absent attempts to keep discussions focused.
Hoping for short term improvement,
(and no Aleksi, I'm sure you had no ill intentions in your email, it was just the final straw for me)
I apologize for my indiscretion and I promise to never post anything on this PDP if nobody else does either. I will consider the PDP closed because there can clearly be no consensus on it, given how far apart the two factions are, and this PDP can thus never be accepted. Yours sincerely this time, -- +358 4567 02048 / http://www.trex.fi/ Aleksi Suhonen / TREX Regional Exchanges Oy `What I need,' shouted Ford, by way of clarifying his previous remarks, `is a strong drink and a peer-group.' -- Douglas Adams, Life the Universe and Everything
Hi Remco, On Fri, Oct 23, 2015 at 01:08:03PM +0000, remco van mook wrote:
While I applaud your willingness to provide a platform to any and all policy proposal regardless of its merit, I must now strongly encourage you to start reconsidering this approach or to start actively moderating discussions on this mailing list.
As much as I value your wisdom and contributions, *this* we totally cannot do - that is, decide "by order of the chair" which proposals have merit or not, or moderate the list (except in extreme cases). We do our best to focus the discussions and do call people to order if needed - OTOH, there are so many aspects to this particular topic (like: is there a real problem with IPv6 in Iran or not?) that it's not easy to declare something to be totally off-topic - and in particular, the argument "just stop wasting all our time on IPv4!" has been brought up before, and I can well see the merits of *this*. So, what do you want us to do? Disallow any discussions that touch IPv4, or have operational / monetary impact on people's networks? This would harm our open process much more than the occasional discussion that strays quite far from the original topic. (But anyway - Remco is right of course that some of the "contributions" to this discussions are *so* totally off-topic that even a very liberal interpretation won't find any useful content in there - like, signature flames going to the whole list - so, please refrain from adding extra noise to the discussion. You know who you are...) Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
< meta-discussion >
As much as I value your wisdom and contributions, *this* we totally cannot do - that is, decide "by order of the chair" which proposals have merit or not, or moderate the list (except in extreme cases).
thanks. i often tend to agree with remco, but not on this one. it's is the openness of the discussion (and yes, some of it is distasteful, welcome to mailing lists) which legitimizes the alleged bottom up coordination (i loathe the word governance). but i do take to heart remco's desire for a bit more civility, and will try to keep my side of the street a bit cleaner. randy
Dear Gert, Sander and others who responded either in private or in public, (more meta discussion)
On 23 Oct 2015, at 18:11 , Gert Doering <gert@space.net> wrote:
Hi Remco,
On Fri, Oct 23, 2015 at 01:08:03PM +0000, remco van mook wrote:
While I applaud your willingness to provide a platform to any and all policy proposal regardless of its merit, I must now strongly encourage you to start reconsidering this approach or to start actively moderating discussions on this mailing list.
As much as I value your wisdom and contributions, *this* we totally cannot do - that is, decide "by order of the chair" which proposals have merit or not, or moderate the list (except in extreme cases).
I think you already do - not consciously, maybe. And for years that has worked to mostly everyone’s satisfaction. What I do note is a general unwillingness to have any discussion that’s not part of the PDP (for those of you who are unfamiliar with it, you can find the most recent version here: https://www.ripe.net/publications/docs/ripe-642 <https://www.ripe.net/publications/docs/ripe-642> ). I understand that the PDP also very much functions as a safety net for the chairs as long as they follow it closely, but at the same time the chairs have been selected by the community itself because of their capabilities to lead a working group - so please, lead. That’s all I’m asking. Some people remarked that it wasn’t my place to make these comments, but given a lack of other people to stand up and say something, I consider it my obligation to this community that I’ve participated in for quite some time now to do get up and draw a line in the sand. The standards of discourse on this list (and I do appreciate a good bit of banter as much as everybody else) are causing people to either stay out, or walk out. How you fix this is up to you, but it is something that needs fixing, quite urgently. I would have left this list already if it wasn’t for some morbid sense of obligation. The argument “this is a mailing list, deal with it” would hold merit if the list had no formal purpose. As is repeated at the beginning of every single AP session at a RIPE meeting, this is the place where policy gets made, and consensus is reached. And nowhere else.
We do our best to focus the discussions and do call people to order if needed - OTOH, there are so many aspects to this particular topic (like: is there a real problem with IPv6 in Iran or not?) that it's not easy to declare something to be totally off-topic - and in particular, the argument "just stop wasting all our time on IPv4!" has been brought up before, and I can well see the merits of *this*.
I wasn’t talking about this topic in particular, although any discussion about how the final scraps of IPv4 are handled seems to bring out the worst in people. I’m very interested to hear about corner cases within our region where certain pieces of community policy are disallowed by local regulation or legislation, but the answer to those is not to then rewrite policy for the entire region. The RIPE region consists of, give or take, 76 countries and there’s no way we could accommodate every quirk or oddity. As for “Let’s stop talking about IPv4” - I think that’s too sweeping a statement to agree with. The focus, in my opinion, should have moved away from IPv4 a while ago but it’s impossible to tell if there’s not some corner case we might have forgotten about, Stating “Let’s only talk about IPv6 from now on” on the other hand als creates an interesting conflict with another working group of the RIPE community. Maybe we should think about a reshuffle?
So, what do you want us to do? Disallow any discussions that touch IPv4, or have operational / monetary impact on people's networks? This would harm our open process much more than the occasional discussion that strays quite far from the original topic.
As David remarked in his (otherwise very off-topic but brilliant post) we’re no longer in 1998. The impact of what happens in here is also a lot bigger: a potential multi-billion euro market for IPv4 transfers was created in here, other policies that reached consensus cost the industry as a whole tens of millions of euros in order to comply. I don’t think it’s too much to ask to have some observed rules around how discussions are being held in here. IETF working groups have a concept of drafts that are accepted by the working group to work on, that would be something to consider. RIPE meetings have a code of conduct these days, that’s maybe another thing.
(But anyway - Remco is right of course that some of the "contributions" to this discussions are *so* totally off-topic that even a very liberal interpretation won't find any useful content in there - like, signature flames going to the whole list - so, please refrain from adding extra noise to the discussion. You know who you are…)
Amen. Kind regards, Remco
On 23/10/15 13:23, Aleksi Suhonen wrote:
-- Aleksi Suhonen
() ascii ribbon campaign /\ support plain text e-mail
Hi, time traveller! Welcome to the future! I'm sure you're excited and eager to discover what has changed over the past 17 years. You probably find today's Internet a confusing place, much different from the one you left behind. From your e-mail, I deduct that you are completely puzzled by the current top-down approach to Internet governance and find yourself wishing to return to the grass-root pseudo-anarchy of yesteryear with which you're familiar. The evolution of the Internet may seem puzzling to you at first. Regulated, accountable, available universally and provided in some countries to all citizens as a basic human right, this always-on global network became an integral part of the world's economy. It is relied upon by billions for both personal and professional use, communication and entertainment, commerce and dating. To ease your transition, here are a few highlights of what has taken place in the time you skipped: CPU speeds have grown by a factor of x50; while your desktop is quite likely powered by an AMD K6-2 CPU running at 266Mhz, by 2003 a desktop CPU model ran at speeds of up to 2400Mhz, and today, it would not be uncommon to own a desktop CPU which runs at upwards of 3000Mhz and offers up to 4 computation cores, such as Intel's 6th generation i7 CPU. For simplicity, you can think of it as a 12000Mhz CPU available in both desktops and laptops (you may know them as "notebooks"). If you stop by an electronics retailer and pick up a new computer, however, you're probably going to find yourself somewhat disappointed. It will not feel considerably faster than the one you've left behind. This is because we've also increased the amount of abstraction on which we rely quite considerably. We no longer write any assembly, and rarely write C or C++. Instead, we rely on interpreted computer languages, or languages which are just-in-time bytecode compiled to be executed by a virtual machine. This may sound a little wasteful, but you'll learn to love it. It also makes optimization someone else's problem, and *everyone* loves that! You'll be happy to hear that this computational excess unleashed a flurry of new programming languages which made programming easier than ever. Programming no longer requires any understanding of the underlying hardware architecture on which the program will be executed, and more people are writing software now than ever before! By the way, you remember JavaScript? You may have used it to play a sound or make an image move when a mouse was hovered over a specific section of that GeoCities homepage you've built with Netscape Navigator Gold? It's grown quite a bit in popularity, and today it's used to write most software, large and small.
From demanding and scalable server software to games, there's hardly a problem node.js isn't an ideal tool for solving!
Storage costs have decreased and capacity has increased by more than x1000 as well! You probably own a 6.4GB drive for which you've paid about 330$. Today, that kind of money would buy you a 6TB drive, easily! You'll still be constantly running out of space, though. High definition video became commonplace, photography has been replaced with digital photography and stills are now taken with digital camera sensors capable of capturing tens of millions of pixels. Internet connectivity speeds have really boomed, increasing by as much as x2000! While you're probably used to a 56Kbps dial-up modem, today, for the same money you paid for your dial-up account, you can get 100Mbps+ service on DOCSIS 3.0 over a coaxial cable plant operated by a cable TV provider. Cable TV, by the way, isn't so hot any more. Now, all of these exciting changes bring us back to the subject of Internet governance. The RIPE NCC of your days operated not much different than, say, a wedding gift registry. It was a convenient arrangement that helped you avoid embarrassing yourself by showing up to the party with the exact same waffle iron someone else has already bought for the lucky couple. Adding and removing things to and from the registry was quite easy, and generally, everyone was quite happy if the registry was consulted at all. As the number of people and devices connected to the Internet increased, and most of these people and devices remained connected all the time, it became apparent that IPv4 addresses will soon run out. Initially, the reaction was that of total disbelief. Surely, 4 billion is a very large number that's almost indistinguishable from infinity? This layman argument resulted in the Internet community spending about a decade between 2000 and 2010 pretending that the problem does not, in fact, exist. In the meanwhile, a few people worked to introduce a new protocol known as IPv6, which aimed to both solve a wide range of problems and expand the number of addresses available. They spent about a decade being largely ignored and laughed at, and were repeatedly told by the Internet community that their new protocol changes too many things which work, and is too different from the IPv4 protocol to understand or implement. After they have, in despair, removed most of these features and made IPv6 look and work almost exactly like IPv4, the same Internet community that previously told them that IPv6 changed too many things began complaining that the IPv6 protocol is insufficiently revolutionary and does not address some of the core challenges which would justify the large expense of a transition. Unfortunately, by that time IPv4 addresses largely ran out and a transition was necessary anyway. In 2007, an Irishman introduced a RIPE policy proposal that became known as policy 2007-01, it granted RIPE NCC a wide range of powers which turned it from a beloved wedding cake registry to something akin to a regulator. RIPE NCC was bestowed with considerable new powers, which it proceeded to consolidate by aggressively reaching out to all resource holders of IPv4 addresses and AS numbers, binding them contractually directly under its control. The RIPE community then passed emergency measures to prevent the complete exhaustion of IPv4 addresses, and emphasized accountability and conservation on all fronts, as well as the need to transition to IPv6. As I write this, the transition from IPv4 to IPv6 isn't nearly complete. The IPv4 address pool is almost completely depleted. New solutions keep being proposed that range from the delusional to the discriminatory. Welcome to the future! On, and by the way, the ASCII ribbon campaign officially ended in June, 2013. We lost. Apparently, plain text e-mail just lacked enough "pizzazz". -- Best regards and success on your journey, David Monosov
Commenting on a signature is beyond off-topic, please stop.
Hi, While I'm sure this subtle point could easily get lost within the text, the primary intention was not so much to comment on a signature, but rather as a light hearted Friday-friendly manner of pointing out that the discussion has strayed in all directions and the positioning on the proposal and related issues is quite extreme and not conductive toward a consensus. A point which, I must admit, you have expressed considerably more concisely than myself. Apologies for the inconvenience. -- Sincerely yours, David Monosov On 23/10/15 16:10, Peter Hessler wrote:
Commenting on a signature is beyond off-topic, please stop.
Le 23/10/2015 16:04, David Monosov a écrit :
On 23/10/15 13:23, Aleksi Suhonen wrote:
-- Aleksi Suhonen
() ascii ribbon campaign /\ support plain text e-mail
As I write this, the transition from IPv4 to IPv6 isn't nearly complete. The IPv4 address pool is almost completely depleted. New solutions keep being proposed that range from the delusional to the discriminatory. Welcome to the future!
Nobody wants Ipv6, but only IPV4 CGN and ASN based routing. :) bst regards.
On, and by the way, the ASCII ribbon campaign officially ended in June, 2013. We lost. Apparently, plain text e-mail just lacked enough "pizzazz".
-- Best regards and success on your journey,
David Monosov
Just a little reminder, can we keep on the topic of this policy... (I agree with remco, and I suggested before as well, certain basic common sense of merderation are appreciated here) On Fri, Oct 23, 2015 at 3:14 PM, Frédéric <frederic@placenet.org> wrote: > > > Le 23/10/2015 16:04, David Monosov a écrit : > > On 23/10/15 13:23, Aleksi Suhonen wrote: > > > >> -- > >> Aleksi Suhonen > >> > >> () ascii ribbon campaign > >> /\ support plain text e-mail > > > > > > > > As I write this, the transition from IPv4 to IPv6 isn't nearly complete. > The > > IPv4 address pool is almost completely depleted. New solutions keep being > > proposed that range from the delusional to the discriminatory. Welcome > to the > > future! > > Nobody wants Ipv6, but only IPV4 CGN and ASN based routing. > > :) > > bst regards. > > > > On, and by the way, the ASCII ribbon campaign officially ended in June, > 2013. We > > lost. Apparently, plain text e-mail just lacked enough "pizzazz". > > > > -- > > Best regards and success on your journey, > > > > David Monosov > > > > > -- -- Kind regards. Lu
Hello Sylvain, The policy is centered around LIRs and granting additional IP space, If a Lir has merged "legitimately" with another lir, they have benefited from the addtitional space, from the merger, and it would be legitimate to say that the Merged Lir is getting the same space as a Lir that didnt merge with anyone else. I would be infavour of this policy if there are constraints, so that it does not give ip space to Lirs that used other mechanisims to increase their ip entitlements, -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Good evening, On 22/10/2015 09:48, Tom Smyth wrote:
I think it would be reasonable that if an entity has merged from another lir... they have already recieved one or more /22s over and above what ripe intended. So these entities have already benefited from gaining additional ips
So it would be fair to exclude such lirs from getting another /22 under this policy proposal
I don't, because there is a confusion here between the LIR and the End User, the LIR being limited to a /22 whereas several End Users having the same LIR would be legitimate to have, say, a /24. So, some level of need-based criteria should not have been abandonned here and the trivial case where a LIR is a single entity that uses all of its allocations for itself should be questionned, since the PA and assignment system is clearly made to distinguish LIR and operator and is supposed to be the normal case. Therefore, creating and merging LIRs can be a very legitimate way to allow new entities to get a minimal IPv4 space, regarding the spirit of the last /8 policiy. Also, policies should not be studied from the only point of view of a LIR but also from the End User's viewpoint, which could be a much more legitimate approach. Best regards, S. Vallerot - -- http://www.opdop.fr - mutualiser et interconnecter en coopérative Opdop - Société Coopérative d'Interêt Collectif sous forme de SARL sur IRC réseau geeknode #opdop - tél: 0950 31 54 74, 06 86 38 38 68 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iF4EAREIAAYFAlYpR2QACgkQJBGsD8mtnRFbcwD9FYxfB1xUzWiJzIljySVPOJMi g5za8bCmCAvlFzzUJv4A/jKSpFup9xH/J+XlqNgN1ZuS6f1/4j9f8pfFO/kYr43X =K42N -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 23/10/2015 16:51, Tom Smyth wrote:
The policy is centered around LIRs
Which is the basis of your further deductions, but I disagree with this first statement. I do not means it is not the case, nor deeply understood as this, I just disagree on the relevance of this lecture regarding to the spirit and goal of the last /8 policy. LIRs are delegated some authority from the RIR (that is delegated authority from ARIN), LIRs are not those who are supposed to use the ressources in the End. So distribution of ressources to LIRs is a wrong perspective, the goal being to have ressources available to End Users, the last /8 that limits available ressources to a /22 per LIR would be better deserving their goal by fixing some ressource quantity to be available to End Users. Of course, this is difficult to do, and most participants here seem not to consider the difference between LIR and operator/end user. So according to the last /8 policy goal (spirit if not letter), LIRs merging to get End Users to be able to access some little ressources is perfectly legitimate. We as a cooperative LIR do no use ressources for ourselves, but for End Users only, so as a LIR, being limited to a /22 is not relevant to us, because it just has the effect to limit the number of End Users that can have access to a minimal part of the IPv4 last bits to bootstrap. Is it the goal of this policy ? No it is not. So to allow new comers to emerge (with a single /24 sometimes) the only possible way today is (several of) them to create a new LIR together and later merge it to ours. And this does perfect sense if last /8 policy is there to allow newcomers to emerge. You thinking as LIR = End User having a /22 means a /22 per newcomer. When you have in mind that a /22 is a potential of 4 x /24 end users instead, then you deserve the last /8 policy 4 times as much. Maybe limiting the M&A to PAs containing space already assigned to enough independent (maybe even routable, with an ASN ?) operators, and garanteed to remain so for quite a long time) would be fine. Best regards, Sylvain - -- http://www.opdop.fr - mutualiser et interconnecter en coopérative Opdop - Société Coopérative d'Interêt Collectif sous forme de SARL sur IRC réseau geeknode #opdop - tél: 0950 31 54 74, 06 86 38 38 68 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iF4EAREIAAYFAlYqUq0ACgkQJBGsD8mtnRGnowEAkJ9DTr65tpRap+4tTLTfO+jK 2wXLItRWhxFWnw2t3U4A/j6d7Hb3nJKSQN72lSGCsEHq0QSxSFSIXPL9KvxGbIo8 =N+ff -----END PGP SIGNATURE-----
Sylvain, I wasnt lecturing, just outlining my postion, the point of the policy is that is is a form of reprieve to allocate more Ipv4 resources to Lirs, and I think that any Lir as I have said before that have benefited from merging with another lir to get Ip address space to assign to end users or other wise, shouldnt be allowed to benefit from this policy. I hope in this context you understand my position Thanks Tom Smyth On Fri, Oct 23, 2015 at 4:30 PM, Sylvain Vallerot < sylvain.vallerot@opdop.net> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 23/10/2015 16:51, Tom Smyth wrote:
The policy is centered around LIRs
Which is the basis of your further deductions, but I disagree with this first statement. I do not means it is not the case, nor deeply understood as this, I just disagree on the relevance of this lecture regarding to the spirit and goal of the last /8 policy.
LIRs are delegated some authority from the RIR (that is delegated authority from ARIN), LIRs are not those who are supposed to use the ressources in the End. So distribution of ressources to LIRs is a wrong perspective, the goal being to have ressources available to End Users, the last /8 that limits available ressources to a /22 per LIR would be better deserving their goal by fixing some ressource quantity to be available to End Users. Of course, this is difficult to do, and most participants here seem not to consider the difference between LIR and operator/end user.
So according to the last /8 policy goal (spirit if not letter), LIRs merging to get End Users to be able to access some little ressources is perfectly legitimate.
We as a cooperative LIR do no use ressources for ourselves, but for End Users only, so as a LIR, being limited to a /22 is not relevant to us, because it just has the effect to limit the number of End Users that can have access to a minimal part of the IPv4 last bits to bootstrap. Is it the goal of this policy ? No it is not.
So to allow new comers to emerge (with a single /24 sometimes) the only possible way today is (several of) them to create a new LIR together and later merge it to ours. And this does perfect sense if last /8 policy is there to allow newcomers to emerge.
You thinking as LIR = End User having a /22 means a /22 per newcomer. When you have in mind that a /22 is a potential of 4 x /24 end users instead, then you deserve the last /8 policy 4 times as much.
Maybe limiting the M&A to PAs containing space already assigned to enough independent (maybe even routable, with an ASN ?) operators, and garanteed to remain so for quite a long time) would be fine.
Best regards, Sylvain
- -- http://www.opdop.fr - mutualiser et interconnecter en coopérative Opdop - Société Coopérative d'Interêt Collectif sous forme de SARL sur IRC réseau geeknode #opdop - tél: 0950 31 54 74, 06 86 38 38 68 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1
iF4EAREIAAYFAlYqUq0ACgkQJBGsD8mtnRGnowEAkJ9DTr65tpRap+4tTLTfO+jK 2wXLItRWhxFWnw2t3U4A/j6d7Hb3nJKSQN72lSGCsEHq0QSxSFSIXPL9KvxGbIo8 =N+ff -----END PGP SIGNATURE-----
-- Kindest regards, Tom Smyth Mobile: +353 87 6193172 --------------------------------- PLEASE CONSIDER THE ENVIRONMENT BEFORE YOU PRINT THIS E-MAIL This email contains information which may be confidential or privileged. The information is intended solely for the use of the individual or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this electronic transmission in error, please notify me by telephone or by electronic mail immediately. Any opinions expressed are those of the author, not the company's .This email does not constitute either offer or acceptance of any contractually binding agreement. Such offer or acceptance must be communicated in writing. You are requested to carry out your own virus check before opening any attachment. Thomas Smyth accepts no liability for any loss or damage which may be caused by malicious software or attachments.
On Wed, Oct 21, 2015, at 13:20, Tom Smyth wrote:
Perhaps people would support the Proposal, if the there was a stricter condition on the transfers, ie
that the Lir has not had any transfered IPs added to its registry, (in addtition to the rule that the Lir has not transferred IPs out of its registry)
You have a point but: - not sure people will support it better with this condition added - there may be legitimate cases where you managed to have a inbound transfer and you still need adresses. Those being said, stricter criteria is under investigation. -- Radu-Adrian FEURDEAN fr.ccs
Am 21.10.2015 um 12:40 schrieb Ciprian Nica:
Iran is a good example, as a country can be considered a relatively new entrant. What has happened over there ? 0.05% IPv6 adoption rate according to google stats and Iran is the #1 importing country of IPv4 resources.
Imo the transfer market was a really bad idea. I assume there are many ISPs with lots of unused space just waiting for the prices getting even higher. Why return a free lunch? Transfer/ selling of ipv4 space should simply be forbidden. In addition big ipv4 space holders should be forced to return ex. 10% of their ipv4 per year. This way the BIG providers (new and existing) would be forced to act and implement ipv6. And these are the only one able to really drive ipv6 adoption. Corin
Transfer/ selling of ipv4 space should simply be forbidden.
Am 21.10.2015 um 13:34 schrieb Randy Bush:
Transfer/ selling of ipv4 space should simply be forbidden.
Most probably only for those who make a living out of this transfer business ;) Corin
So it would be ok to sell as long as you don't make a living out of it :) The idea was that you can't just forbid the wave to hit you. Nobody can control the entire community and decide what will happen. RIPE (community) has decided to allow transfers as it would help ease the pain on those who really need IPv4 resources. Do you think that UK's DWP would have ever given resources from their /8 to the community if there were no millions to make out of it ? Maybe some would be willing to give the IPs back to the community but usually there are some internal costs to release them. And if you are refering to brokers like me, we are only intermediating transactions, most of us are just creating links between buyers and sellers and assist them to properly complete the transactions. Yes we are making a living out of it but it's for the benefit of both seller and buyer and we don't make the decisions on their behalf. On 10/21/2015 2:40 PM, Netskin NOC wrote:
Am 21.10.2015 um 13:34 schrieb Randy Bush:
Transfer/ selling of ipv4 space should simply be forbidden.
Most probably only for those who make a living out of this transfer business ;)
Corin
Hi, On Wed, Oct 21, 2015 at 03:11:40PM +0300, Ciprian Nica wrote:
The idea was that you can't just forbid the wave to hit you. Nobody can control the entire community and decide what will happen. RIPE (community) has decided to allow transfers as it would help ease the pain on those who really need IPv4 resources.
Well, the most important reason was that we can't stop transfers from occuring (people will find ways...) but that *if* a resource is transferred, we must be able to document *where it went*. Since the IPv4 run-out, the primary function of the RIPE NCC has shifted from "hand out IPv4 addresses" to "registry", and this is more important than ever. (Of course one could frown on people making money on something they got for free - yes. But do we want to make money on something by selling under the hand, for a higher price, and no guarantees to the buyer?) Gert Doering -- NetMaster -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Am 21.10.2015 um 14:40 schrieb Gert Doering:
Well, the most important reason was that we can't stop transfers from occuring (people will find ways...)
Why not? RIPE is the official register and has to act on each transfer. Of course some might "lease" their space instead, but this puts a significant risk on the user/ customer. I assume many users wouldn't want to take that risk and so the majority of this business is blocked. Together with an enforced return of some percent of assigned ipv4 space every year this would be very effective to push ipv6 adoption significantly.
(Of course one could frown on people making money on something they got for free - yes. But do we want to make money on something by selling under the hand, for a higher price, and no guarantees to the buyer?)
That's why transfer/ selling should be completely forbidden. Corin
Hi, On Wed, Oct 21, 2015 at 03:17:08PM +0200, Netskin NOC wrote:
Am 21.10.2015 um 14:40 schrieb Gert Doering:
Well, the most important reason was that we can't stop transfers from occuring (people will find ways...)
Why not? RIPE is the official register and has to act on each transfer. Of course some might "lease" their space instead, [..]
You already described one possible way. Others exist, but I'm not going to try to list all of them.
(Of course one could frown on people making money on something they got for free - yes. But do we want to make money on something by selling under the hand, for a higher price, and no guarantees to the buyer?)
That's why transfer/ selling should be completely forbidden.
If you try to forbid it, prices will go up, and the registry "which network is where?" suffers as a consequence. Gert Doering -- NetMaster -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Am 21.10.2015 um 14:11 schrieb Ciprian Nica:
The idea was that you can't just forbid the wave to hit you. Nobody can control the entire community and decide what will happen. RIPE (community) has decided to allow transfers as it would help ease the pain on those who really need IPv4 resources.
Don't get me wrong, but I think we are in some kind of dilemma here. Many knew this would become a free lunch, so why vote against it?
Do you think that UK's DWP would have ever given resources from their /8 to the community if there were no millions to make out of it ? Maybe some would be willing to give the IPs back to the community but usually there are some internal costs to release them.
"With great power comes great responsibility.". Unused space has to be returned to the registry for free, as the initial requirements for the assignment are no longer met. RIPE should audit and enforce it - just start by grabbing all listings on the transfer market *g*.
And if you are refering to brokers like me, we are only intermediating transactions, most of us are just creating links between buyers and sellers and assist them to properly complete the transactions. Yes we are making a living out of it but it's for the benefit of both seller and buyer and we don't make the decisions on their behalf.
It's not against your business, it's against the current policies. Corin
Don't get me wrong, but I think we are in some kind of dilemma here. Many knew this would become a free lunch, so why vote against it?
Policies are adopted through consensus so we can asume that at the time of adoption any policy was what (most of) the community wanted.
"With great power comes great responsibility.". Unused space has to be returned to the registry for free, as the initial requirements for the assignment are no longer met. RIPE should audit and enforce it - just start by grabbing all listings on the transfer market *g*.
Yes, I'm waiting for Santa Claus, too. Unfortunately our capitalist society is mostly profit-driven so we have to live with it whether we like it or not. Go to Daimler and tell them that. Will they return any of the 16,7 million IPs that they hold (53.0.0.0/8) ? I remember (I hoep I'm not wrong) that RIPE NCC has contacted UK DWP a few years ago about their /8 and they said they can't return any of the IPs. This year they started to monetize a few million IPs already.
It's not against your business, it's against the current policies.
The policies are adopted by the community and represent the best decisions that could have been taken at that time. Of course things changes and sometimes policies need to be changed. However I don't think the current proposal would bring any good change. Ciprian
Am 21.10.2015 um 14:11 schrieb Ciprian Nica:
The idea was that you can't just forbid the wave to hit you. Nobody can control the entire community and decide what will happen. RIPE (community) has decided to allow transfers as it would help ease the pain on those who really need IPv4 resources.
Don't get me wrong, but I think we are in some kind of dilemma here. Many knew this would become a free lunch, so why vote against it?
Do you think that UK's DWP would have ever given resources from their /8 to the community if there were no millions to make out of it ? Maybe some would be willing to give the IPs back to the community but usually there are some internal costs to release them.
"With great power comes great responsibility.". Unused space has to be returned to the registry for free, as the initial requirements for the assignment are no longer met. RIPE should audit and enforce it - just start by grabbing all listings on the transfer market *g*.
And if you are refering to brokers like me, we are only intermediating transactions, most of us are just creating links between buyers and sellers and assist them to properly complete the transactions. Yes we are making a living out of it but it's for the benefit of both seller and buyer and we don't make the decisions on their behalf.
It's not against your business, it's against the current policies. Corin
Hi, On Wed, Oct 21, 2015 at 02:51:58PM +0200, Netskin NOC wrote:
Unused space has to be returned to the registry for free, as the initial requirements for the assignment are no longer met.
These are *allocations* and not assignments, and there has never been such a policy for allocations. Repeat: there is no policy that mandates return of unused allocations, and no mandate from the community for the NCC to go out and pester allocation holders to voluntarily return anything. Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Assignments are between LIR and end user and at this moment RIPE doesn't care much about them, only, as you mentioned, that they are properly reflected in the registry. If there were a policy already allowing RIPE to get back allocations, I think the situation would have been different (I can't stop thinking about my chinese "friend") Maybe it is time to create such policy. It will not be easy but maybe we are able to come up with some rules that NCC can implement and get back some of the space that we all know it's just waiting for a better price. Ciprian On 10/21/2015 4:06 PM, Gert Doering wrote:
Hi,
On Wed, Oct 21, 2015 at 02:51:58PM +0200, Netskin NOC wrote:
Unused space has to be returned to the registry for free, as the initial requirements for the assignment are no longer met.
These are *allocations* and not assignments, and there has never been such a policy for allocations. Repeat: there is no policy that mandates return of unused allocations, and no mandate from the community for the NCC to go out and pester allocation holders to voluntarily return anything.
Gert Doering -- APWG chair
Hi, On Wed, Oct 21, 2015 at 04:18:44PM +0300, Ciprian Nica wrote:
Assignments are between LIR and end user and at this moment RIPE doesn't care much about them, only, as you mentioned, that they are properly reflected in the registry.
If there were a policy already allowing RIPE to get back allocations, I think the situation would have been different (I can't stop thinking about my chinese "friend")
Maybe it is time to create such policy. It will not be easy but maybe we are able to come up with some rules that NCC can implement and get back some of the space that we all know it's just waiting for a better price.
I could point out that *this* would very much be "retroactively applying policy"... (And there would still never be sufficient IPv4, so I think that's why the community decided a few years ago to not bother going there - we've discussed this at RIPE meetings every now and then, and decided to better focus on making good IPv6 policies instead) Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Am 21.10.2015 um 23:00 schrieb Gert Doering:
(And there would still never be sufficient IPv4, so I think that's why the community decided a few years ago to not bother going there - we've discussed this at RIPE meetings every now and then, and decided to better focus on making good IPv6 policies instead)
The point is to speedup ipv6 adoption of the big ISPs by enforcing the return of existing ipv4 allocations over the next few years (tbd, like 10% of the allocated ipv4 space of each LIR per year). As a minor side effect everyone could get an /21 without any problems (which most probably wouldn't then be needed anymore anyway). Corin
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 21/10/2015 23:19, Netskin NOC wrote:
The point is to speedup ipv6 adoption of the big ISPs by enforcing the return of existing ipv4 allocations over the next few years (tbd, like 10% of the allocated ipv4 space of each LIR per year). As a minor side effect everyone could get an /21 without any problems (which most probably wouldn't then be needed anymore anyway).
I agree, it is a much better approach that 2015-5 : make the IPv4 dissappear OK but the large LIRs reserves should be drained at the same time. Unfortunately I suspect we lost the power to do so when we adopted the catastrophic "no need" policy, since big LIRs would easily declare fake assignments now to fill and reserve their already routed but yet unsed allocations. We would have to restore needs-based policy first. Best regards, Sylvain -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iF4EAREIAAYFAlYyK0oACgkQJBGsD8mtnRHaawD9EH+bBRnGt3xmGh9/pT370qdn hBChWiYaePdb3VZjDiwBAIQPKJD89dYihIG3CiXST4nk0hqHlaRLic2na0R2jRVM =Pyla -----END PGP SIGNATURE-----
On Thu, Oct 29, 2015, at 15:20, Sylvain Vallerot wrote:
Unfortunately I suspect we lost the power to do so when we adopted the catastrophic "no need" policy, since big LIRs would easily declare fake assignments now to fill and reserve their already routed but yet unsed allocations. We would have to restore needs-based policy first.
Things are a little bit more complex than that, but globally you are right, and we should have been more vocal years ago, when nonsense started to develop (no need, no new PI, PI trasnfer, ...). -- Radu-Adrian FEURDEAN fr.ccs
Hi, On Thu, Oct 29, 2015 at 09:27:03PM +0100, Radu-Adrian FEURDEAN wrote:
Things are a little bit more complex than that, but globally you are right, and we should have been more vocal years ago, when nonsense started to develop (no need, no new PI, PI trasnfer, ...).
The "nonsense", as you call it, has been the result of long discussions on this very mailing list, leading to community consensus that this is what we want our policy to be. If you don't understand the reasoning behind the changes, it would be useful to read up the discussions in the archives. It does not suit the proposer of a policy change very well to call earlier policy changes "nonsense". Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
Am 21.10.2015 um 15:06 schrieb Gert Doering:
These are *allocations* and not assignments, and there has never been such a policy for allocations. Repeat: there is no policy that mandates return of unused allocations, and no mandate from the community for the NCC to go out and pester allocation holders to voluntarily return anything.
Sorry, if I mixed it up. I really think it's time for such a policy then. Corin
This would make an important change to the current situation and we should think it very thorough. In less than a month it's RIPE 71 in Bucharest. Let's discuss this further on the mailing list and at RIPE71 and if there's support for the idea, we'll collect your opinions and come up with a feasible proposal. Ciprian On 10/21/2015 4:25 PM, Netskin NOC wrote:
Am 21.10.2015 um 15:06 schrieb Gert Doering:
These are *allocations* and not assignments, and there has never been such a policy for allocations. Repeat: there is no policy that mandates return of unused allocations, and no mandate from the community for the NCC to go out and pester allocation holders to voluntarily return anything.
Sorry, if I mixed it up. I really think it's time for such a policy then.
Corin
On Wed, Oct 21, 2015, at 13:34, Randy Bush wrote:
Transfer/ selling of ipv4 space should simply be forbidden.
Encouraging and stimulating it OTOH, could have been skipped/avoided. -- Radu-Adrian FEURDEAN fr.ccs
Transfer/ selling of ipv4 space should simply be forbidden. https://en.wikipedia.org/wiki/King_Canute_and_the_waves Encouraging and stimulating it OTOH, could have been skipped/avoided.
the true believers tried to pretend they could hold back the water for many years. some are still in denial. there is an ipv4 address space market. the rirs are effectively out of the free source of integers they rent to us. and world peace could use some work too. this ain't our ancestors' internet. we need to get over it. dealing with current reality is hard enough without fantasy and wishful thinking in the way. randy
participants (17)
-
Aleksi Suhonen -
Ciprian Nica -
Corin Langosch -
David Monosov -
Frédéric -
Gert Doering -
Lu Heng -
Netskin NOC -
Peter Hessler -
Radu-Adrian FEURDEAN -
Randy Bush -
Remco van Mook -
remco van mook
-
Shahin Gharghi -
Sylvain Vallerot -
Tom Smyth -
Uros Gaber