Re: [address-policy-wg] IPv4 Maintenance Policy
At 02:42 PM 4/19/2012 +0000, Milton L Mueller wrote:
This aspect of the policy regarding legacy holders needs clarification:
"Leave data as it is in the RIPE Registry. The Legacy Resource Holder will not be able to add to or alter their data and will not have access to any RIPE NCC services such as reverse delegation and certification."
It is likely that in response to this policy legacy holders will choose to use an alternate registrar for the services you are precluding them from using (e.g., reverse delegation and certification). In that case RIPE NCC will need to negotiate an interoperability or interconnection agreement with these service alternate providers to ensure that a globally applicable unique registration occurs.
If RIPE NCC is not willing to do that, it appears to be attempting to leverage its monopoly to force legacy holders into purchase and use of their services, something that raises obvious competition policy issues. I wouldn't advise you to do down that path.
The problem is not so much the whois database but rather the delegated file which is the more "official" data for all RIRs: ftp://ftp.ripe.net/pub/stats/ripencc/delegated-ripencc-latest When ARIN and RIPE did their ERX thing in 2003, ARIN mistakenly transferred 4600+ IP blocks to RIPE with country=EU rather than country=xx. There are numerous software packages that use the delegated data to do geolocation. Examples: http://code.google.com/p/ci-geoip/ http://16bytes.com/geo_locating-html/ https://metacpan.org/module/IP::Country::DB_File::Builder http://www.codecodex.com/wiki/IP_Address_to_Country So your IP block would not say your country code but rather EU. There is one commercial firewall that uses the delegated data incorrectly as well and hides the details behind something more powerful than a firewall - a lawyerwall. RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR. -Hank
So your IP block would not say your country code but rather EU. There is one commercial firewall that uses the delegated data incorrectly as well and hides the details behind something more powerful than a firewall - a lawyerwall.
RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR.
we're strongly committed to registry accuracy, but shipping will cost you extra. randy
-----Original Message-----
RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR.
And I wonder how those who suggest there are no competition issues here, would explain that refusal?
Hi again Milton, On Fri, Apr 20, 2012 at 4:25 PM, Milton L Mueller <mueller@syr.edu> wrote:
-----Original Message-----
RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR.
And I wonder how those who suggest there are no competition issues here, would explain that refusal?
I would say that the RIPE NCC does what its members tell it to do. If the Db serves up the same data to all, there in no competitive advantage or disadvantage to any geo-location provider is there? If geo-location providers want NCC members to foot the bill for finer grained accuracy, I would say that is a bit cheeky. I'm not sure that I want much finer grained geo-location (from a privacy perspective at least). -- Cheers, McTim "A name indicates what we seek. An address indicates where it is. A route indicates how we get there." Jon Postel
On Fri, 20 Apr 2012, McTim wrote:
Hi again Milton,
On Fri, Apr 20, 2012 at 4:25 PM, Milton L Mueller <mueller@syr.edu> wrote:
-----Original Message-----
RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR.
And I wonder how those who suggest there are no competition issues here, would explain that refusal?
I would say that the RIPE NCC does what its members tell it to do.
Mctim, Can you point me at the membership request or the WG request to register all legacy IP blocks under a LIR? -Hank
If the Db serves up the same data to all, there in no competitive advantage or disadvantage to any geo-location provider is there?
If geo-location providers want NCC members to foot the bill for finer grained accuracy, I would say that is a bit cheeky. I'm not sure that I want much finer grained geo-location (from a privacy perspective at least).
On 21/04/2012 19:03, Hank Nussbacher wrote:
Can you point me at the membership request or the WG request to register all legacy IP blocks under a LIR?
The RIPE NCC is constituted to do what the RIPE community requests. The RIPE NCC members are a small subset of that community. It's clear that there is a problem here which needs to be fixed, and which balances both the concerns of the ERX holders and the RIPE community (+ by extension, the RIPE NCC). Nick
-----Original Message-----
I would say that the RIPE NCC does what [the voices] tell it to do.
Hi, call to order, please. Please debate the merits and shortcomings of the proposal at hand, instead of trying to voice some sort of carefully-tendered grudge against the RIR system. Gert Doering -- APWG chair On Mon, Apr 23, 2012 at 12:44:23AM +0000, Milton L Mueller wrote:
-----Original Message-----
I would say that the RIPE NCC does what [the voices] tell it to do.
-- 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 (89) 32356-444 USt-IdNr.: DE813185279
Gert: Learn to take a joke. And try to apply the same standard to RIR supporters as to critics, ok?
-----Original Message----- From: Gert Doering [mailto:gert@space.net] Sent: Monday, April 23, 2012 7:08 AM To: Milton L Mueller Cc: McTim; address-policy-wg@ripe.net Subject: Re: [address-policy-wg] IPv4 Maintenance Policy
Hi,
call to order, please. Please debate the merits and shortcomings of the proposal at hand, instead of trying to voice some sort of carefully- tendered grudge against the RIR system.
Gert Doering -- APWG chair
On Mon, Apr 23, 2012 at 12:44:23AM +0000, Milton L Mueller wrote:
-----Original Message-----
I would say that the RIPE NCC does what [the voices] tell it to do.
-- 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 (89) 32356-444 USt-IdNr.: DE813185279
Hi, On Tue, Apr 24, 2012 at 01:03:22PM +0000, Milton L Mueller wrote:
Gert: Learn to take a joke.
In the context this was posted, it wasn't obvious whether this was a joke or yet another nasty remark. But it was off-topic in any case, so it doesn't particularily matter.
And try to apply the same standard to RIR supporters as to critics, ok?
I do. If discussions stray too far from the topic at hand (which is "IPv4 maintenance policy", not "global policy forum" or "the RIRs are good/evil/..." whatever), I'll call to order. Now, if you have something to contribute, I'm all ears... 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 (89) 32356-444 USt-IdNr.: DE813185279
Gert: Let me get you corrected on the facts. You seem to have lost track of them and are confusing your animosity toward certain ideas/people with off-topic discussion. My goal is to encourage a more robust and substantive discussion of the reasons why RIPE (or any other RIR) finds it necessary to sacrifice registry accuracy in order to pressure legacy holders into contracts. The conversation in question began with a discussion of a specific sentence of the proposed IPv4 Maintenance policy, having to do with legacy holders. I expressed concern as to the competition policy implications of that. There was no response to this highly substantive question from you or anyone else associated with the development of that policy. There was also a concern expressed by Nussbacher, which was followed by a sarcastic one liner by Randy Bush which, I note, did not invoke any reprimand from you. In reply to my substantive comment, there was a statement from McTim that RIPE, like all RIRs, always follows the will of the community. This was a rather interesting comment given that the policy in question has not yet passed and thus we do not know whether it reflects anyone's will except that of the people who proposed it. But this statement that "the RIRs are always good" did not lead to any reaction from you. McTim's comment was followed by my humorous substitution of "community" with "the voices." This was followed by a reprimand. I think the pattern is clear.
I do. If discussions stray too far from the topic at hand (which is "IPv4 maintenance policy", not "global policy forum" or "the RIRs are good/evil/..." whatever), I'll call to order.
Another example of your discriminatory attitude. There is nothing in the messages under this heading about a "global policy forum." Moreover, this is an address policy forum and it is well within its remit to discuss global policy development and a global policy forum, if people here want to do so. The fact that you do not like the idea is not a justification to suppress discussion of it. Now, before you or anyone else accuses me of wasting everyone's time, let me just say that the real cause of the distraction here is your own (Gert's) attempt to arbitrarily single out a particular discussant for one (admittedly) sarcastic remark (although one in a veritable ocean of such remarks involving multiple people), a remark that was no more or less off topic than the comment that preceded it. Attempts to deal with important disagreements via intimidation and silencing won't work with this guy.
-----Original Message----- From: Gert Doering [mailto:gert@space.net] Sent: Tuesday, April 24, 2012 9:15 AM To: Milton L Mueller Cc: Gert Doering; address-policy-wg@ripe.net Subject: Re: [address-policy-wg] IPv4 Maintenance Policy
Hi,
On Tue, Apr 24, 2012 at 01:03:22PM +0000, Milton L Mueller wrote:
Gert: Learn to take a joke.
In the context this was posted, it wasn't obvious whether this was a joke or yet another nasty remark. But it was off-topic in any case, so it doesn't particularily matter.
And try to apply the same standard to RIR supporters as to critics, ok?
I do. If discussions stray too far from the topic at hand (which is "IPv4 maintenance policy", not "global policy forum" or "the RIRs are good/evil/..." whatever), I'll call to order.
Now, if you have something to contribute, I'm all ears...
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 (89) 32356-444 USt-IdNr.: DE813185279
On 24 Apr 2012, at 14:48, Milton L Mueller wrote:
Gert: Let me get you corrected on the facts. You seem to have lost track of them and are confusing your animosity toward certain ideas/people with off-topic discussion.
My goal is to encourage a more robust and substantive discussion of the reasons why RIPE (or any other RIR) finds it necessary to sacrifice registry accuracy in order to pressure legacy holders into contracts.
Milton, what you've just stated is an opinion, not fact. Andrea has already explained why much of the ERX space that was thrown over the wall to the NCC is tagged as "EU". Perhaps you didn't see this before you sent your email. In any case, a childish game of "oh yes it is, oh no it's not" or an existential debate over whether your opinion (or mine) is fact or not is unhelpful and inappropriate for this list. So if you want to carry on with that futile discussion, please do so somewhere else. Meanwhile, you could also improve the signal to noise ratio by actually discussing address policy or even submitting a proposal to this list. So instead of claiming that the NCC or RIRs in general are behaving in ways that are anti-competitive (in your opinion) please put forward proposals which address these issues or at least clearly identify the perceived problem/issue. Suggest solutions. If you can submit a constructive proposal that's technically sound, it will be warmly welcomed.
There is nothing in the messages under this heading about a "global policy forum."
This is grossly misleading and you know that. You have been posting on this list (admittedly in a different thread) about your IGF workshop and stating there will be a global policy proposal. I presume you mean this IGF workshop will or could turn into that forum.
-----Original Message----- From: Jim Reid [mailto:jim@rfc1035.com]
My goal is to encourage a more robust and substantive discussion of the reasons why RIPE (or any other RIR) finds it necessary to sacrifice registry accuracy in order to pressure legacy holders into contracts.
Milton, what you've just stated is an opinion, not fact. Andrea has already explained why much of the ERX space that was thrown over the wall to the NCC is tagged as "EU". Perhaps you didn't see this before you sent your email.
I did see it. I actually already knew a bit about the ERX, & thought Andrea's explanation was helpful. But it was not germane to the point I was making. My point was about the problems associated with trying to force legacy holders to sign RIR contracts before they can update their records. Perhaps you didn't see my original note. If you think my "opinion" that RIPE is willing to sacrifice registry accuracy in order to pressure legacy holders into contracts is incorrect, please support your counter-opinion with rational discussion and not with expressions of personal hostility. I note that another member of this list stated: "It's clear that there is a problem here which needs to be fixed, and which balances both the concerns of the ERX holders and the RIPE community (+ by extension, the RIPE NCC)." While it would be nice to have a specific proposal on that, until we discuss it and get a better grasp of the pros and cons of different approaches, that is obviously premature.
In any case, a childish game of "oh yes it is, oh no it's not" or an existential debate over whether your opinion (or mine) is fact or not is unhelpful and inappropriate for this list. So if you want to carry on with that futile discussion, please do so somewhere else.
Jim, you are in no position to tell me or anyone else on this list what is helpful and appropriate. Please understand: I am not going to be silenced by unfair charges that my points are childish bickering simply because you and Gert have chosen to engage in childish bickering. I know the game you're playing very well, and it won't work. I will remain focused on substantive issues and I will not allow an important discussion to be diverted or suppressed by these kinds of tactics.
Meanwhile, you could also improve the signal to noise ratio by actually discussing address policy or even submitting a proposal to
Again, this is an obviously false charge. I am discussing a specific proposal (see the header) and my comments referenced a specific section of it. Several other members of this list are involved in the discussion. You could improve the signal to noise ratio by actually discussing the policy issue instead of wasting our time on a futile attempt to drive me away with repeated expressions of personal hostility.
There is nothing in the messages under this heading about a "global policy forum."
This is grossly misleading and you know that. You have been posting on this list (admittedly in a different thread) about your IGF workshop and stating there will be a global policy proposal. I presume you mean this IGF workshop will or could turn into that forum.
Yes, I do. That is an entirely appropriate topic for this forum. Indeed, there is a proposal on the table in RIPE for an inter-RIR transfer policy, and in that context an IGF workshop focused on a global transfer policy is a highly relevant topic. It has been greeted with interest by many people, including in ARIN, APNIC and AfriNIC, as well as operators. It is interesting to me that you are oblivious to the contradiction between telling me to develop proposals on the one hand and this attempt to suppress any discussion of an attempt to develop a proposal. --MM
On 25/04/2012 14:21, Milton L Mueller wrote:
I note that another member of this list stated: "It's clear that there is a problem here which needs to be fixed, and which balances both the concerns of the ERX holders and the RIPE community (+ by extension, the RIPE NCC)." While it would be nice to have a specific proposal on that, until we discuss it and get a better grasp of the pros and cons of different approaches, that is obviously premature.
It is in progress - some of the ERX holders are discussing this on a mailing list. In the interim, there's not much point in creating proposals until the ERX holders have come up with some concrete ideas about what they're looking for. Nick
On 25 Apr 2012, at 14:21, Milton L Mueller wrote:
It is interesting to me that you are oblivious to the contradiction between telling me to develop proposals on the one hand and this attempt to suppress any discussion of an attempt to develop a proposal.
Milton, there is clearly no point discussing this further and I deeply resent your repeated efforts to personalise matters and make false allegations. Whatever contradiction you perceive exists in your head. Nobody is suppressing anything here: just asking for the discussion to be relevant and on-topic. Ad-hominen attacks -- you falsely accuse me of being repeatedly personally hostile to you -- are not appropriate for this list, so take them elsewhere. Once again I encourage you to submit a constructive proposal that's technically sound.
Dear Hank, On 4/20/12 12:42 PM, Hank Nussbacher wrote:
At 02:42 PM 4/19/2012 +0000, Milton L Mueller wrote:
This aspect of the policy regarding legacy holders needs clarification:
"Leave data as it is in the RIPE Registry. The Legacy Resource Holder will not be able to add to or alter their data and will not have access to any RIPE NCC services such as reverse delegation and certification."
It is likely that in response to this policy legacy holders will choose to use an alternate registrar for the services you are precluding them from using (e.g., reverse delegation and certification). In that case RIPE NCC will need to negotiate an interoperability or interconnection agreement with these service alternate providers to ensure that a globally applicable unique registration occurs.
If RIPE NCC is not willing to do that, it appears to be attempting to leverage its monopoly to force legacy holders into purchase and use of their services, something that raises obvious competition policy issues. I wouldn't advise you to do down that path.
The problem is not so much the whois database but rather the delegated file which is the more "official" data for all RIRs: ftp://ftp.ripe.net/pub/stats/ripencc/delegated-ripencc-latest
When ARIN and RIPE did their ERX thing in 2003, ARIN mistakenly transferred 4600+ IP blocks to RIPE with country=EU rather than country=xx.
The ERX project was a database project. Its aim was ensuring that ranges were listed in the correct RIR database. IP blocks were transferred from the ARIN database to the RIPE Database with the country code listed in the inetnum objects at time of the transfer. All legacy resources transferred to the RIPE NCC, that are not associated with any LIR and part of the RIPE Registry, are registered under a placeholder 'eu.zz-transfer'. EU is the country code used when there is no specific country associated with the resources in the RIPE Registry. This results in the EU country code in the delegated stats file (not in the RIPE Database).
There are numerous software packages that use the delegated data to do geolocation. Examples: http://code.google.com/p/ci-geoip/ http://16bytes.com/geo_locating-html/ https://metacpan.org/module/IP::Country::DB_File::Builder http://www.codecodex.com/wiki/IP_Address_to_Country So your IP block would not say your country code but rather EU. There is one commercial firewall that uses the delegated data incorrectly as well and hides the details behind something more powerful than a firewall - a lawyerwall.
The delegated stats file was never intended to be used for geo-location information: ftp://ftp.ripe.net/pub/stats/ripencc/_README However as you mentioned, different organisations are using it for this purpose.
RIPE NCC is unwilling to fix this, unless one submits the legacy IP space to be listed under a LIR.
You can register the legacy resources you are the legitimate holder of, to your LIR. These resources would then get the country code of that LIR. This is however by no means enforced... Creating fake LIRs and/or temporarily move resources to an LIR without a contract in place are no options. These are quick "fixes" that will just do harm on the medium term. For this reason we are currently working on improving the delegated and the delegated extended stats files: they will use a different data-set and show the country codes relative to the single resource. The country code will be taken form the RIPE Database. This will be released in the near future. Best regards, Andrea Cima RIPE NCC
-Hank
participants (8)
-
Andrea Cima
-
Gert Doering
-
Hank Nussbacher
-
Jim Reid
-
McTim
-
Milton L Mueller
-
Nick Hilliard
-
Randy Bush