Re: [ncc-services-wg] 2012-07 New Policy Proposal (RIPE NCC Services to Legacy Internet Resource Holders)
On 11/09/2012 13:20, Emilio Madaio wrote:
A new RIPE Policy Proposal has been made and is now available for discussion.
You can find the full proposal at:
Unless I'm mistaken, this is an almost verbatim copy of what Niall posted to this list on Aug 22 18:28:48. Some way down that discussion, Niall noted:
Background work on a new version of the policy proposal has begun. It is premature for me to say more at this stage.
So, given that it appears that the proposers of this policy are interested in a new version, is it appropriate for people to comment on this version? Perhaps a rep of one of the policy proposers could clarify this? thanks, Nick
Hi Nick,
You can find the full proposal at:
Unless I'm mistaken, this is an almost verbatim copy of what Niall posted to this list on Aug 22 18:28:48.
Correct
Some way down that discussion, Niall noted:
Background work on a new version of the policy proposal has begun. It is premature for me to say more at this stage.
So, given that it appears that the proposers of this policy are interested in a new version, is it appropriate for people to comment on this version?
Yes please! One thing we realised when talking about the current text is that it is more than just a policy proposal. It also contains historic stuff and summaries of other documents. We (the proposers) are thinking about cutting away the informational text (probably to put it in a separate informational document) to leave only the real policy related text. So all comments are still greatly appreciated! They will certainly be taken into account for either the final policy text or the informational text.
Perhaps a rep of one of the policy proposers could clarify this?
I hope I have :-) Sander
On 12/09/2012 00:42, Sander Steffann wrote:
One thing we realised when talking about the current text is that it is more than just a policy proposal. It also contains historic stuff and summaries of other documents. We (the proposers) are thinking about cutting away the informational text (probably to put it in a separate informational document) to leave only the real policy related text.
So all comments are still greatly appreciated! They will certainly be taken into account for either the final policy text or the informational text.
ok. my main concerns about it are: 1. it doesn't establish a quid pro quo between the erx holders and the ripe ncc. It looks to me like all the obligations are on the RIPE NCC and that the ERX holders have no obligations whatsoever. This is not - and cannot become - the basis of a functional relationship between the RIPE NCC and the ERX holders because the basis of any functional relationship between two entities must be a quid pro quo. 2. there is a lot of talk about the rights of ERX holders in this document. I'm unclear on why the proposers believe that the RIPE community has the competence to issue statements of rights like this, given that they don't believe that the RIPE community has the competence to create policies concerning this address space. You can't have it both ways. 3. suggesting policy statements which cannot be undone by future policy statements seems...odd. 4. it's unclear to me to what extent the policy document represents the consensus viewpoint of the ERX holder community. I have a lot of other smaller concerns, but this is more than enough to start with. In short, I can't really see how this document could become the basis of an agreement between the RIPE NCC and the ERX holders. Daniel's email to ncc-services-wg of Aug 29 16:37:07 CEST 2012 seems like a much more sensible starting point. Nick
Hi Nick,
ok. my main concerns about it are:
1. it doesn't establish a quid pro quo between the erx holders and the ripe ncc. It looks to me like all the obligations are on the RIPE NCC and that the ERX holders have no obligations whatsoever. This is not - and cannot become - the basis of a functional relationship between the RIPE NCC and the ERX holders because the basis of any functional relationship between two entities must be a quid pro quo.
I don't agree at all with you here. The proposal says that the RIPE NCC has to deliver certain services (registry + reverse DNS) and that the legacy resource holders have to sign contracts (and very probably have to pay some money, like with 2007-01) and maintain their data in the registry. The goal of the recent actions of the RIPE NCC was to make the registry data more accurate, and I think that is a very important goal. The RIPE NCC officially has no 'rights' over the legacy resources as they were given to the holders before the RIPE NCC even existed, so this proposal tries to provide a framework to establish a formal relationship so that the registry can be properly maintained.
2. there is a lot of talk about the rights of ERX holders in this document. I'm unclear on why the proposers believe that the RIPE community has the competence to issue statements of rights like this, given that they don't believe that the RIPE community has the competence to create policies concerning this address space. You can't have it both ways.
I very much dislike your usage of the word 'competence'. I don't think it's appropriate here. The thing is that currently the RIPE NCC and the RIPE community have no rights regarding legacy resources. The legacy resource holders are afraid that signing contracts (directly or indirectly) with the RIPE NCC will make them lose certain rights. The current template contracts distributed by the RIPE NCC (see http://www.ripe.net/lir-services/resource-management/legacy-space/template-l...) contain text like "We are aware that from the moment that the address space mentioned above is registered in our LIR account, it will be considered as address space distributed to us by the RIPE NCC. It will then be subject to relevant RIPE policies and RIPE NCC procedures." and this is not acceptable to many legacy resource holders. The reaction to these texts was to explicitly make sure that rights were not lost in the policy proposal. The problem is that the current situation for legacy resource holders seemed to be "If you want your reverse DNS to work you have to give up all your legacy rights to your address space". More recent communication from the NCC seems much more relaxed, but the contracts still contain that wording...
3. suggesting policy statements which cannot be undone by future policy statements seems...odd.
I fully agree, but I also understand that legacy holders don't want to sign contracts that currently let them keep their rights but in the future might take them away. The reason that this policy proposal is in the NCC Services working group is that legacy resource holders would like certain services like registry updates and reverse DNS to keep working and they don't mind contributing financially to running the RIPE NCC, but they don't want any address policies to affect them since that address space was not given to them by the RIPE NCC in the first place.
4. it's unclear to me to what extent the policy document represents the consensus viewpoint of the ERX holder community.
I'll leave that question for ERX holders to answer :-)
I have a lot of other smaller concerns, but this is more than enough to start with.
In short, I can't really see how this document could become the basis of an agreement between the RIPE NCC and the ERX holders. Daniel's email to ncc-services-wg of Aug 29 16:37:07 CEST 2012 seems like a much more sensible starting point.
I agree. I think we have to focus on "So what we have to decide as a community is: under which policies does the RIPE community allow legacy space holders to register their address space in the RIPE Internet Number registry. Nothing more, nothing less.". The recent actions by the RIPE NCC have caused some fear and frustration amongst the legacy resource holders, and the current policy proposal reflects that. The next version of this policy proposal should remove that and focus on what Daniel said (+ reverse DNS services I think). Thanks, Sander PS: Daniel said in his post that "The RIPE community has carefully avoided to make legacy address space subject to address space *distribution* policies, such as utilisation criteria.". I fully agree with this, but unfortunately this is not reflected in the template contracts as mentioned above.
On 12/09/2012 12:17, Sander Steffann wrote:
I very much dislike your usage of the word 'competence'. I don't think it's appropriate here.
Oops sorry, I meant "competent" in the legal sense. This is a specific legal use of the word which just means "within jurisdiction" in this context. It has no negative connotations whatever. Nick
Hi Nick,
I very much dislike your usage of the word 'competence'. I don't think it's appropriate here.
Oops sorry, I meant "competent" in the legal sense. This is a specific legal use of the word which just means "within jurisdiction" in this context. It has no negative connotations whatever.
Ah, ok :-) I'm not that familiar with legal terms ;-) Thanks! Sander
On Wed, Sep 12, 2012 at 01:17:03PM +0200, Sander Steffann wrote:
I agree. I think we have to focus on "So what we have to decide as a community is: under which policies does the RIPE community allow legacy space holders to register their address space in the RIPE Internet Number registry. Nothing more, nothing less.". The recent actions by the RIPE NCC have caused some fear and frustration amongst the legacy resource holders, and the current policy proposal reflects that. The next version of this policy proposal should remove that and focus on what Daniel said (+ reverse DNS services I think).
I think it's simple: Which is the higher good, a high-quality ripedb or control of the ERX space? IMO, that is for the *community* to decide, and therefore gives the *community* standing to make such policy. If the ripedb quality is the goal, I have no problems with a policy framework that imposes no obligations on the ERX holders vis-a-vis the NCC. Even requiring payment might be detrimental to this goal as some ERX holders might not want to pay and don't care about db entries... I'd support the proposal but for a disagreement with the wording in (ia) s5.5: "In the case that the holder of a legacy Internet resource opts to engage via a sponsoring LIR and wishes to avail only of basic services as defined above, assessment of the billing category of the sponsoring LIR should take account of the legacy resource using a score or rate less than that applicable for a corresponding PI resource" I've always taken the position that resources that are not controlled, merely sponsored, by a LIR should not count into the billing category calculations for that LIR, and thus cannot agree with any proposal that can be construed to cement this practice for the forseeable future. rgds, Sascha Luck
Hi Sasha,
I'd support the proposal but for a disagreement with the wording in (ia) s5.5:
"In the case that the holder of a legacy Internet resource opts to engage via a sponsoring LIR and wishes to avail only of basic services as defined above, assessment of the billing category of the sponsoring LIR should take account of the legacy resource using a score or rate less than that applicable for a corresponding PI resource"
I've always taken the position that resources that are not controlled, merely sponsored, by a LIR should not count into the billing category calculations for that LIR, and thus cannot agree with any proposal that can be construed to cement this practice for the forseeable future.
I agree that text related to billing categories should be left out of the next version of the proposal. Thanks, Sander
Hi Sander,
I don't agree at all with you here. The proposal says that the RIPE NCC has to deliver certain services (registry + reverse DNS) and that the legacy resource holders have to sign contracts (and very probably have to pay some money, like with 2007-01) and maintain their data in the registry.
I read it differently, in that there was no requirement for erx holders to sign anything if they chose not to, and that if they chose not to, the policy proposal required that the RIPE NCC would still provide services indefinitely. Could you confirm whether this is the case?
The current template contracts distributed by the RIPE NCC (see http://www.ripe.net/lir-services/resource-management/legacy-space/template-l...) contain text like "We are aware that from the moment that the address space mentioned above is registered in our LIR account, it will be considered as address space distributed to us by the RIPE NCC. It will then be subject to relevant RIPE policies and RIPE NCC procedures." and this is not acceptable to many legacy resource holders. The reaction to these texts was to explicitly make sure that rights were not lost in the policy proposal.
Yep, I see the issue here, except that these documents are now redundant, as far as I can tell from what Nigel said a couple of weeks ago. Could someone from the RIPE NCC clarify the position here? My understanding - and again I am open to correction on this - is that we are now where we were pre Oct 2011: everything is on the table for discussion, ERX holders are able to update their objects as before and the RIPE NCC are not moving forward with the mandatory ERX agreement that you referred to.
The problem is that the current situation for legacy resource holders seemed to be "If you want your reverse DNS to work you have to give up all your legacy rights to your address space". More recent communication from the NCC seems much more relaxed, but the contracts still contain that wording...
If the contracts no longer reflect the position of the RIPE NCC, this should be made clear by either removing them or else by making a statement to the effect that the URLs + contract are no longer applicable. It would be really helpful if we had a statement from the RIPE NCC on this.
3. suggesting policy statements which cannot be undone by future policy statements seems...odd.
I fully agree, but I also understand that legacy holders don't want to sign contracts that currently let them keep their rights but in the future might take them away.
Can the RIPE Community create a policy that the RIPE Community cannot change? I seem to remember that there are a bunch of somewhat narcissistic philosophical treatises on this particular subject. But at a more practical level, I don't really see that this position is either possible or necessary. It's certainly contentious.
The reason that this policy proposal is in the NCC Services working group is that legacy resource holders would like certain services like registry updates and reverse DNS to keep working and they don't mind contributing financially to running the RIPE NCC, but they don't want any address policies to affect them since that address space was not given to them by the RIPE NCC in the first place.
Completely reasonable.
4. it's unclear to me to what extent the policy document represents the consensus viewpoint of the ERX holder community.
I'll leave that question for ERX holders to answer :-)
Nicely dodged :-)
I agree. I think we have to focus on "So what we have to decide as a community is: under which policies does the RIPE community allow legacy space holders to register their address space in the RIPE Internet Number registry. Nothing more, nothing less.". The recent actions by the RIPE NCC have caused some fear and frustration amongst the legacy resource holders, and the current policy proposal reflects that. The next version of this policy proposal should remove that and focus on what Daniel said (+ reverse DNS services I think).
Ok. In the interim, I think we need some clarity from the RIPE NCC about the status of the content on the web tree with this root: http://www.ripe.net/lir-services/resource-management/legacy-space Nick
Hi Nick,
I don't agree at all with you here. The proposal says that the RIPE NCC has to deliver certain services (registry + reverse DNS) and that the legacy resource holders have to sign contracts (and very probably have to pay some money, like with 2007-01) and maintain their data in the registry.
I read it differently, in that there was no requirement for erx holders to sign anything if they chose not to, and that if they chose not to, the policy proposal required that the RIPE NCC would still provide services indefinitely. Could you confirm whether this is the case?
Well, that is still an ongoing discussion. So far we have been looking at legacy resource holders that want to cooperate with the RIPE NCC. What happens when a legacy resource holder refuses to sign anything should be discussed here I think. My personal opinion is that at least a best effort attempt should be made to always keep the RIPE database as accurate as possible, even without any signed contracts. That is because keeping the RIPE database accurate is important for the whole community, not just for the legacy resource holder.
[...]
Ok. In the interim, I think we need some clarity from the RIPE NCC about the status of the content on the web tree with this root:
http://www.ripe.net/lir-services/resource-management/legacy-space
Yes, I would also appreciate that! Sander
Dear Nick, All, On 9/13/12 5:10 PM, Nick Hilliard wrote:
I don't agree at all with you here. The proposal says that the RIPE NCC has to deliver certain services (registry + reverse DNS) and that the legacy resource holders have to sign contracts (and very probably have to pay some money, like with 2007-01) and maintain their data in the registry. I read it differently, in that there was no requirement for erx holders to sign anything if they chose not to, and that if they chose not to, the policy proposal required that the RIPE NCC would still provide services indefinitely. Could you confirm whether this is the case?
The current template contracts distributed by the RIPE NCC (see http://www.ripe.net/lir-services/resource-management/legacy-space/template-l...) contain text like "We are aware that from the moment that the address space mentioned above is registered in our LIR account, it will be considered as address space distributed to us by the RIPE NCC. It will then be subject to relevant RIPE policies and RIPE NCC procedures." and this is not acceptable to many legacy resource holders. The reaction to these texts was to explicitly make sure that rights were not lost in the policy proposal. Yep, I see the issue here, except that these documents are now redundant, as far as I can tell from what Nigel said a couple of weeks ago. Could someone from the RIPE NCC clarify the position here?
The references to the possible freeze of RIPE Database objects and reverse DNS services have been removed, as per the emails from Nigel and Andrew to this list. The RIPE NCC receives requests from legacy address space holders to register their resources on a weekly basis. When this happens we make them aware that if they do not want to register, their RIPE Database objects will not be frozen and that we will continue to provide rDNS services. Additionally we direct them to this discussion. If they still wish to register their legacy resources, we ask them to fill out and submit the document which is found on our website. For this reason, amongst others, the documents and webpage in question are still valid. Finally, since the start of this discussion we have ceased contacting non-LIR legacy resource holders (phase 3), as we are awaiting the outcome of this policy proposal.
My understanding - and again I am open to correction on this - is that we are now where we were pre Oct 2011: everything is on the table for discussion, ERX holders are able to update their objects as before and the RIPE NCC are not moving forward with the mandatory ERX agreement that you referred to.
Changing the relationship with legacy resource holders that have signed an agreement with the RIPE NCC can be done retroactively, based on the outcome of this policy proposal. I would like to add that during this project the RIPE NCC has not frozen any resources or denied any services.
The problem is that the current situation for legacy resource holders seemed to be "If you want your reverse DNS to work you have to give up all your legacy rights to your address space". More recent communication from the NCC seems much more relaxed, but the contracts still contain that wording... If the contracts no longer reflect the position of the RIPE NCC, this should be made clear by either removing them or else by making a statement to the effect that the URLs + contract are no longer applicable. It would be really helpful if we had a statement from the RIPE NCC on this.
The contract on our website does not contain any references to reverse DNS. Please see: http://www.ripe.net/lir-services/resource-management/legacy-space/template-l... Best regards, Andrea Cima Registration Services RIPE NCC
3. suggesting policy statements which cannot be undone by future policy statements seems...odd. I fully agree, but I also understand that legacy holders don't want to sign contracts that currently let them keep their rights but in the future might take them away. Can the RIPE Community create a policy that the RIPE Community cannot change? I seem to remember that there are a bunch of somewhat narcissistic philosophical treatises on this particular subject.
But at a more practical level, I don't really see that this position is either possible or necessary. It's certainly contentious.
The reason that this policy proposal is in the NCC Services working group is that legacy resource holders would like certain services like registry updates and reverse DNS to keep working and they don't mind contributing financially to running the RIPE NCC, but they don't want any address policies to affect them since that address space was not given to them by the RIPE NCC in the first place. Completely reasonable.
4. it's unclear to me to what extent the policy document represents the consensus viewpoint of the ERX holder community. I'll leave that question for ERX holders to answer :-) Nicely dodged :-)
I agree. I think we have to focus on "So what we have to decide as a community is: under which policies does the RIPE community allow legacy space holders to register their address space in the RIPE Internet Number registry. Nothing more, nothing less.". The recent actions by the RIPE NCC have caused some fear and frustration amongst the legacy resource holders, and the current policy proposal reflects that. The next version of this policy proposal should remove that and focus on what Daniel said (+ reverse DNS services I think). Ok. In the interim, I think we need some clarity from the RIPE NCC about the status of the content on the web tree with this root:
http://www.ripe.net/lir-services/resource-management/legacy-space
Nick
On 11 Sep 2012, at 13:36, Nick Hilliard wrote:
Unless I'm mistaken, this is an almost verbatim copy of what Niall posted to this list on Aug 22 18:28:48. Some way down that discussion, Niall noted:
Background work on a new version of the policy proposal has begun. It is premature for me to say more at this stage.
So, given that it appears that the proposers of this policy are interested in a new version, is it appropriate for people to comment on this version?
Perhaps a rep of one of the policy proposers could clarify this?
Sander already responded, but I'ld like to add a few words. The document which, mea culpa, I inappropriately released to the mailing list on 22 August, contains what was intended by the authors as version 1 of a policy proposal. The document formally released by Emilio as version 1 of policy proposal 2012-07 is indeed essentially identical. My intent, which Emilio took every care to accommodate, was to minimize the potential for confusion that seemed inevitable in case there was a significant difference between the two documents. I acknowledge responsibility for this decision, thank Emilio for his meticulous support, and thank Bijal and Kurtis for their continued understanding. Best regards, Niall O'Reilly
Niall,
The document which, mea culpa, I inappropriately released to the
just as a side note: i do not understand this sentiment. Nowhere does it say that policy ideas and proposals cannot be discussed on WG mailing lists before formal submission.
authors as version 1 of a policy proposal. The document formally released by Emilio as version 1 of policy proposal 2012-07 is indeed essentially identical.
So, I understand this is to ensure the legacy resource holders of the continued, unconstrained and, at least in part, free of charge, access to certain services that are tightly bound to the then registration (well, back then it was an assignment even if practices predate the term). I would like to see some motivation why this subject to the PDP at all rather than, maybe, an issue for the NCC AGM (not taking a position, but asking for clarification) because it in part looks like a service fee waiver. If there was a policy to instantiate, the infinite perpetuation clause in section 6.3 appears to me like a non starter. However, if this is more the expression of an opinion or legal standpoint, then it should be made visible as such. Which leaves me to conclude that most of what is proposed is actually not a proposal but a memorandum of understanding (or simply an acknowledgement) of the "rights" of legacy resource holders with respect to today's governance framework and service environment. It is much more deductive than constructive, so again the question is why the PDP is appropriate (and also, why this can be appropriately addressed by collective action). In any case, I would question the part where the "rights" are attached to the resource rather than the historic assignment act. Looking forward to the edited next version. -Peter
The document which, mea culpa, I inappropriately released to the just as a side note: i do not understand this sentiment. Nowhere does it say that policy ideas and proposals cannot be discussed on WG mailing lists before formal submission.
however it seems to have become a policy proposal whether we like it or not. and as such it seems to have some formal restrictions on being updated, whether we like it or not.
So, I understand this is to ensure the legacy resource holders of the continued, unconstrained and, at least in part, free of charge, access to certain services that are tightly bound to the then registration
i believe you understand incorrectly. the words "unconstrained" and "free of charge" are yours.
I would like to see some motivation why this subject to the PDP at all rather than, maybe, an issue for the NCC AGM (not taking a position, but asking for clarification) because it in part looks like a service fee waiver.
there are far more issues than fees, which i believe are the purview of the AGM. and i believe what is being asked to be considered is in fact to pay for services now received for free. but there are also issues of address policy and similar.
Looking forward to the edited next version.
as are many. i think folk are figuring out how to do that within the rule trap into which it has fallen. randy
On 13 Sep 2012, at 14:43, Randy Bush wrote:
[Peter Koch]
Looking forward to the edited next version.
as are many. i think folk are figuring out how to do that within the rule trap into which it has fallen.
Besides, we have to finish editing it first. 8-) One of the slides I'm drafting, and which I expect will be shown at RIPE 65, currently looks like this: -------- Policy Proposal 2012-07 (version 2) * Shorter * No need to itemize defence of what is not under threat * Focus: * registration and related services * fair access to services for legacy resource holders -------- It may save us all much time and effort if further discussion on this list is focussed along these lines, as indeed some of the discussion to date has been. ATB /Niall
On 13 Sep 2012, at 13:43, Peter Koch wrote:
Niall,
[ceterum censeo] I'm not ignoring you, Peter, but Randy already did a good job of answering. ATB /Niall
On 12 sep 2012, at 18:40, Niall O'Reilly <Niall.oReilly@ucd.ie> wrote:
My intent, which Emilio took every care to accommodate, was to minimize the potential for confusion that seemed inevitable in case there was a significant difference between the two documents. I acknowledge responsibility for this decision, thank Emilio for his meticulous support, and thank Bijal and Kurtis for their continued understanding.
So just to clarify how the chairs sees this working out. We wanted the original version posted so we could track it and save some time. Due to a little misunderstanding and unclarity on my part the posting got delayed. Now at least we have a version posted, and we have both the previous discussions as well as some new comments. The authors have expressed that they are taking this feedback into account and will look at a new release. We will wait for further policy actions until we have a new version posted. Best regards, - kurtis -
participants (8)
-
Andrea Cima -
Lindqvist Kurt Erik -
Niall O'Reilly -
Nick Hilliard -
Peter Koch -
Randy Bush -
Sander Steffann -
Sascha Luck