Search RPD Archives
[rpd] Ratified Policy Proposal - AFPUB-2020-GEN-006-DRAFT03 AFRINIC Number Resource Policy Transfer
jordi.palet at consulintel.es
jordi.palet at consulintel.es
Sun May 24 07:56:13 UTC 2026
Hi Paul,
I don’t recall now all the details of the discussion, which it seems you reviewed in detail and I agree with your summary.
What I’m sure is that I always discussed the lack of complete reprocity with other RIRs, which impacts on the volume of transfers that can come in to AFRINIC, and also discussed that changes in a policy proposal can’t be done after consensus has been reached.
In the last call we can do an editorial change, but what is the border line about what is an editorial change and what not?
Some times can be subjective. For me is clear: anything that corrects grammar or orthography or improves the reading of the text, only if is clear that it doesn’t change the meaning of what was written when the consensus was achieved.
To solve this problem, in LACNIC I submitted a proposal to modify the PDP in several aspects including this one, it reached consensus and was ratified in mid-2023, and was implemented in June 2024 with this text:
"To publish a four-week last call for comments period for any proposal that reaches consensus. In the case of editorial changes, a new version of the proposal must be published and the last call for comments period must be restarted."
This way, the chairs have the chance to re-evaluate that the editorial changes aren’t altering the view of the PDWG.
Regarding the policy compliance dashboard, it was not ratified by the board:
https://lists.afrinic.net/pipermail/rpd/2026/014703.html
As a consequence, the authors had a call with the staff and then submitted several weeks ago a new version. We are waiting for possible inputs from the staff before publication.
The major problem we are having is what I asked the staff to confirm very urgently in this email:
https://lists.afrinic.net/pipermail/rpd/2026/014777.html
Without a clear statement on that, any proposals that we submit, either, we need to reinforce the “supervision” by AFRINIC and possible recovery of resources in case of lack of compliance, or AFRINIC has a general mandata to verify compliance of all the CPM and taking actions.
Note that in all the RIRs this is very clear, and we even have in some cases, defined a very clear procedure by means of a policy (https://www.lacnic.net/687/2/lacnic/) so it is crystal clear and fair to all the resource holders. This is despite the RSAs being similar and not contradictory in this aspect. So either we have a problem with Mauritius legislation or the legal advisor of AFRINIC and the Board should reconsider if the CPM and possible new policies are in fact “Encroaching” AFRINIC instead of trying to set fair procedures (please read the staff assessment https://afrinic.net/policy/proposals/2021-gen-003-d2#impact).
Regards,
Jordi
@jordipalet
> El 24 may 2026, a las 3:55, Paul Hjul <hjul.paul at gmail.com> escribió:
>
> I shared Mark Elkins curiosity. Although I was perhaps a little less optimistic. Then I saw Owen's post and really hoped he was wrong. Being less optimistic has meant going over the record and putting together this email. It being a lazy Saturday has meant that this email comes after a couple of additions to the discussion.
>
> Owen has pointed out that the AS WRITTEN policy imposes on the recipient compliance with AFRINIC policies and to sign an RSA regardless of whether the recipient is an AFRINIC member or not. I really really really hoped he was wrong but unfortunately he is not.
>
> There is a difference in the presentation of Draft 3 (which is what has been ratified) and Draft 2. Draft 2 (https://afrinic.net/policy/proposals/2020-gen-006-d2#proposal) appears as follows:
>
> <image.png>
>
> The bullets suggest (but an ambiguity arises) that you have 1,2 then 3 under one set of circumstances followed by 4 if 3 was not applicable. Followed by a general statement that lends itself to being an alternative to the whole section.
>
> Draft 3 (https://afrinic.net/policy/proposals/2020-gen-006-d3#proposal) presents the conditions as:
>
> <image.png>
>
> Draft 3 contains no indication that the section 3.6 was amended:
> <image.png>
>
>
>
> The PDWG co-chairs in an email provide a different formatting of the text:
>
> <image.png>
>
>
> Even with this formatting (which formatting was not circulated in any last call. The conditions on the recipient flow as: 1] Will be subject to AFRINIC policies, then 2] Must sign RSA, 3] IF not holding resources; 4] ELSE, 5] IF
> 1 and 2 are not indicated as conditional and nothing in 5 gives any indication of suppressing the application of 1 - 4.
>
> Therefore while "the full text provides necessary context" it doesn't provide clarity. To get clarity you have to dig further and when we dig further the clarity that comes up is not good for the PDP.
>
> I understand that Gregoire is attempting to convey that Owen is wrong and that the intent of the final point is to act as an alternative to the earlier points but this is not the most logical understanding of the text. The fact that the as written text produces an absurd policy that creates a massive problem for the organization doesn't mean that the text can be magically changed. At best it means the policy will be ignored or that a judicial authority will step in (and there are good reasons to assume that they won't).
>
> The phrase "the counterpart's RIR transfer policy" does not make sense to me. The term arises undoubtedly from the idea that other RIRs are counterparts to AFRINIC but there is nothing contractual (or at least there shouldn't be ...) involving AFRINIC the recipient, the source and another RIR in the transfer process. But putting that to one side its clear that both AFRINIC and another RIR transfer policy will give rise to conditions. This creates a mess.
>
> The real problem is the presumption that the recipient as a resource holder is under an RIR and that resource holders are in one region in the divided up world.. RIR service regions aren't a domicile or jurisdiction. I think the M word will start getting thrown around several gaskets blown when talk of carving up the market into regions with an enterprise controlling the market in their agreed upon region.
>
> It is difficult to say that a policy has "rough consensus" if there isn't agreement as to what a policy means or if the meaning is vague or ambiguous. As written draft 3 imposes on the recipient (whoever they may be) the
>
> It is worth noting that "Must sign RSA" doesn't specify which RSA, A major opportunity for problems arises if a host of conflicting contractual instruments are produced and I am quite sure that the A in RSA is for agreement. Does the policy act to mandate entering into an agreement by which supply of services are collusively controlled? Lets hope not, and lets really hope that it can't be argued persuasively that that is what is happening.
>
> It is also worth noting that as written the policy causes the resources to be "marked". Presumably once marked the resources will be the subject of further unstated policy prescriptions as may be promulgated. Nothing in the policy assures continued grandfathering in respect of "legacy".
> I don't understand why anybody who has "legacy" would "transfer" those resources into AFRINIC under this policy. The policy causes the resource ... I suspect that the authors of the policy (and probably AFRINIC staff) are harbouring the assumption that an AFRINIC member cannot also hold legacy resources or be a resource holder with resources within the administrative ambit of different RIRs. The staff assessment correctly identifies this problem:
> "It is important to highlight that, as a matter of law, legacy resource holders existing within the AFRINIC’s service region are not contractually bound by AFRINIC’s adopted policies such that these policies have no direct effect on legacy resource holders, and it is up to those legacy-holders to adhere to AFRINIC’s policies. Thus, the authors should bear in mind that obligations impacting legacy resource holders may not necessarily achieve the intended results if the legacy resource holders refuse to opt for voluntary registration of the transfer with AFRINIC."
>
> Looking at it from the perspective of a legacy resource holder the idea of this sort of divide up and marking would raise eyebrows. I suspect that an attempt to mute objections to the policy from legacy resource holders while avoiding a proper grandfathering in is what is attempted but the overall "solution" is probably not the outcome anybody was banking on.
>
> Now the best case scenario is to assume that the staff assessment's proposition of law is correct and that legacy resource holders aren't given an RSA to sign but the policy does not address the issue and you have legacy resource holders who do enter into an RSA for other resources. This policy demands the signing of an RSA. It is a little more than a bit absurd.
>
> I don't know if the ratification by the Board proves anything about the Board at all (so at least Owen is in my view wrong about one thing, unfortunately its a minor one and not what I was hoping he was wrong about when I took a dive into this). Ratification is on the advise of the PDWG and staff and if the Board refuses to ratify when given advise that the policy is order and after it has been vetted for implementation by staff it would be quite a tall order for the Board to refuse and with the documentation which would have been placed in front of the Board I don't think they or the Receiver had any basis not to ratify. This is particularly because of the language of the CPM (as at version 1.3) at 3.4.4 "The recommendation shall include a report of the discussions of the draft policy and feedback from the Last Call. The draft policy shall be ratified by the AFRINIC Board of Directors". I do not see real scope for reconsideration by the Board on their own volition. For the Board to act beyond being a rubber stamp requires reliance on the bylaws 11.4 as amended in 2020 [and that is assuming that 11.4 survives scrutiny] although in light of what will follow it might be necessary for the Board to consider an emergency measure.
>
> Now for the reasons I have set out the Board probably should revoke the ratification (although to be honest I am not entirely sure that such a power exists) or at least should not oppose the legal challenge which appears (https://afrinic.net/court-cases) to have been brought by Skyconnect in SC/COM/PWS/132/2026 [plaint challenging the Board's ratification of the transfer policy] to the ratification. It will be most problematic if AFRINIC expends resources on a policy that is dead on arrival or worse introduces problems for the organization as written. If the authors want the document defended then they can instruct solicitors in Mauritius.
>
> The process manual is clear that "No change can be made to a draft policy within one week of the meeting. This is so that a stable version of the draft policy can be considered at the meeting" . At PPM 34 (https://afrinic.net/policy/development-working-group/ppm-afrinic-34) draft 2 was under discussion. The discussion was dominated by defensive posture of the authors and editorial changes desired by staff. Rough consensus was declared on a commitment to editorial changes.
>
> I don't see how you can have a version change of the document submitted after the PPM presumed to find rough consensus is held. This point was raised in the discussion by Jordi and I cautioned as to handling the situation in light of discussions of other conflicting policies that were ongoing. Be that as it may substance rather than form should be looked at. You could have a situation where a substantive and informed discussion on a 3rd draft produces actual consensus that is more coherent than the rough consensus in the 2nd draft.
>
> However what clearly happened is that version 3 was not circulated and discussed upon during last call and only an email (https://lists.afrinic.net/pipermail/rpd/2021/014054.html) containing a link to version 3 was sent as "Summary of Proposal". I am not sure that such a summary can really be considered as an announcement but for present purposes it doesn't matter CPM 3.4.2 requires a distinct announcement of last call because even if it is accepted that the PDWG was in last call between 8th December 2021 and 5th January 2022 the document presented to PDWG gave no indication of the change to 3.6. More importantly the entirety of the discussion in 2022 (https://lists.afrinic.net/pipermail/rpd/2022/date.html) (and the discussion from the date of the PPM) was in respect of version 2. There is not a single email outside of the co-chairs touching on version 3. There is also concurrent discussion on the withdrawn competing policy proposal and that discussion has not be absorbed into the consensus assessment of this policy proposal.
>
>
> If the AFRINIC staff responsible for enforcement of the code of conduct were committed to fair application that code they would take steps to address the clear and egregious misconduct of the authors of this policy proposal who have acted in bad faith and contrary to the effort to find solutions to problems. Owen's statement that "board remains capable of absurdity" is not one I agree with but its well within the bounds of civic discourse. The policy as ratified is absurd. The false accusation by one of the authors of the policy should result in the organisation doing the homework of walking through the adoption process.
>
> As I have confidence in the integrity of Dr Vincent Ngundi and Darwin Da Costa (the co-chairs) I expect that on being presented with clear evidence that the presumed rough consensus that they will take appropriate and prudent steps to guard the PDP from being brought into further disrepute. I hope I am not wrong on this point.
>
> In my view this particular policy proposal was a good faith effort by the PDWG co-chairs to get manifestly incompatible efforts to dictate policy into a single policy document but that unfortunately in the exercise of doing so a gremlin slipped through the cracks. What Owen has pointed out - and Mark sitting in anticipation as to "how this becomes implemented" - requires the PDWG co-chairs to do something. I am acutely aware of there existence of pressure to get an inter-RIR transfer policy and the ideological hunting grounds which this policy has landed in. The worry though is some years down the line and the need for a defensible and sound policy has only grown.This policy proposal was moved forward with amendments as other conflicting policy proposals timed out. The absence of an inter-RIR transfer policy was raising issues surrounding whether AFRINIC or the RIRs more broadly are operating such that the M word should be thrown about. And the as ratified proposal probably should see the M word being thrown about but it really should be getting some thought about related questions of collusive agreements and the like.
>
> Mike Silber strenuously disagreed with the contention that RIRs without an intra-RIR transfer policy would be a de facto monopoly and asserted that people were arguing beyond their scope of knowledge. When it comes to the field of competition law and policy everybody invariably steps a little bit out of their lane because of the interplay with economics, People involved in the Internet and telco space seem to really get quite religious about the M word. Of course competition considerations are broader than whether something is a monopoly. Now I've publicly taken the stand that in terms of design the RIRs, particularly AFRINIC does not fall within the general remit of Mauritius competition law or of any other competent jurisdiction's competition law (and by competence here I am referring of course to whether the jurisdictions law can legitimately find application rather than a comment on the jurisdiction or its judicial officers, lest somebody mischievously steps in on the word "competence", although there are conceivable circumstances where a possible cause outside Mauritius could arise) but both that there are circumstances in which a competition issue could arise and that the broad concept of promoting a competitive ecosystem should be a factor in decision making. Moreover, Mauritius is not in the European Union (or notwithstanding Brexit the United Kingdom - now through Statutory Instruments and using its own CMA) and so the Pavlov doctrine is not applicable and so arguing that AFRINICis an "undertaking" or more importantly an "enterprise" is a hard ask. I do think that if AFRINIC was in a country subject to the Treaty on the Functioning of the European Union a whole host of complexity would arise but ultimately that one of the 101(3) based block exemptions would apply [I don't feel like diving into my course and study notes on block exemptions beyond the TTBER as the nightmares for preparing for the exam Competition law and technology transfer haven't yet abated]. I actually followed up with the Mauritian authorities (and published to the Community Mailing List) on the matter and the response affirmed what I had said and contradicted the attempted spin from AFRINIC staff. I am of the view that the definition of "enterprise" (in Mauritius) which requires "engaged in commercial activities for gain or reward" is dispositive so long as AFRINIC is able to demonstrate that it is properly operating as a non-commercial entity that is not-for-profit. However a given RSA could be held to be within the scope of Section 41 or Section 44 (and thereby Section 41) of the Mauritius Competition Act of 2007 (as amended) does not depend on an enterprise being identified. if it can be shown that the RSA is an "agreement" which "significantly prevents, restricts or distorts competition" that agreement can be nullified by statute. Therefore an intentional policy that obliges a commercial entity to enter into an agreement which has the object or effect of restricting the acquisition of services by any person and significantly distorts competition puts you in a dangerous space. It is honestly a significant failing that within the staff assessment the legal assessment contains no indication of consideration of the Competition Act. You are welcome to argue (wrongly) that there are no concerns touching of the Competition Act but in light of what was discussed in the mailing lists the staff assessment is incomplete with this wrong position not being stated. In my view that this point alone necessitates a review of the action of the applicable staff who advised the Board to ratify. Also Mike was quite ready to point out that "ICP-2 no longer has any status" (https://lists.afrinic.net/pipermail/rpd/2022/014296.html) and well that has definitely changed - exactly what the status of ICP-2 with the revision is is a lot more complicated. Moreover ICANN are now a party in winding up proceedings (Crystal Web's stance on the winding up of AFRINIC is on record) and I think they are very much taking on the role of supervising compliance to assure the Supreme Court that it is not in the interests of justice to wind up AFRINIC. So even without invoking the mighty M word there are clearly reasons to consider the competition dynamics and aspects of the registry system and the role the RIRs are playing.
>
> As written the policy holds that a "recipient" is subject both to AFRINIC policies (including any policy which is restrictive or materially hinders competition) and that they "must sign" a resource services agreement (which agreement would be void if it is held to be a collusive horizontal agreement) coupled with a sharing of commercial plans. Moreover the recipient is agreeing to adhere to further conditions imposed by a "counterpart" that being another RIR. It frankly does not matter whether somebody wants the policy to mean that the conditions of the recipient are different (rather than cumulative) if the recipient "is in another region".
> Moreover the "in another region" language presents its own major problem. A multinational resource holder would be in Africa and Europe, if such entity transfers resources to their European operation is the presence of the entity in Africa of relevance. Equally if an Australian company is the recipient of resources which it uses in African operations the phrase used ties those resources to two sets of policies... What of an African company (incorporated in an African state) that has significant ultimate beneficial shareholders in the "other regions"?
>
>
> If we consider what Ernest Byaruhanga and his henchmen were (and quite likely are) trying to do it may be the case of cockup before conspiracy but isn't that half of the idea and this point? Noah's email I think speaks to the fact that at least one of the authors intends the policy to make the impositions which Owen has raised as an absurdity. If it is indeed the case that the co-chairs presumed the agreed to amendments meant one thing and one of the authors was quite intentionally seeking to slip through something else well then you've got a problem.
>
> And here is another procedural problem, policy version 3's change from policy version 2 on the point at hand was not communicated publicly by the chairs and according to https://www.afrinic.net/policy/appeal-committee#members there was no appeal committee in effect at the time the chair's called consensus.
>
> <image.png>
>
> Moreover there isn't an PDAC at present. So clearly you have a situation in which any disregarding of valid technical objections by the co-chairs would fall through the cracks. Add to this the fact that the policy proposal was amended to address input from the staff and that there was a body of discussion on the version 2.
>
> I don't know why the Board in 2022 (while still quorate) did not ratify the policy but what is clear is that once a Board took office this policy was put before the Board as if the policy had simply been waiting ratification.
>
>
> I don't know what the co-chairs of the PDWG should do here. What I do know is that there is a problem with what was ratified and the authors of the policy document clearly want to drag the PDP into further disrepute. What I also know is that they owe Owen an apology. If the PDWG is to be a "professional and respectful environment is essential to achieving productive outcomes for the community" then that starts with the co-chairs. As I've shown it is incorrect to attribute to the Board but it is equally wrong to claim that an absurdity being ratified is outside the scope of the PDP when the absurdity is from the PDP.
>
> Considering that same co-chairs had previously muzzled Owen (https://lists.afrinic.net/pipermail/rpd/2021/013969.html) the situation is a little worse. The stated grounds for the muzzling was improper and are a misuse of moderating power. Moreover it is clear that the complaint against Owen (which was spurious and an abuse of office) by the then CEO of AFRINIC and it is reasonable to assume that the PDWG was instructed to take steps against Owen due to the fact that he drew attention to the known corruption. As he correctly pointed out it is quite difficult to assert that a statement that there are crooks within an organization is defamatory especially if the speaker can point to a factual basis.
> If Owen was a vindictive character ... Fortunately for the individuals who are the co-chairs I don't think that he is.
> For present purposes though what is clear is that during the critical period when a discussion on the particulars of the 3rd draft was needed Owen was excluded from participation this not only means . It further gives a reasonable apprehension that the co-chairs exhibited bias in their handling of objections to the draft.
>
> I hope that if indeed the imposition of moderation was communicated by but not decided by the co-chairs that same will inform the PDWG of the same but failing that being communicated I am afraid they are responsible for the misconduct at hand. Even if a person wishes to gloss over the issue the consequence that the declared consensus is unsustainable cannot be ignored. What I hope is that the co-chairs reflect on their misconduct, apologise to both Owen and the PDWG and build on the experience. Then again I might be wrong and the co-chairs could decide to double down and back the authors of the impugned policy in which case the problems in the process do unfortunately become a concern for the Board. Either way I am afraid that trying to build consensus for the next PPM is going to be a bigger challenge than it really should be.
>
> On another note from PPM-34 then Proposal 5 "Policy Compliance Dashboard - Policy Proposal" reached consensus but does not appear to have moved on. I was supportive of this policy with an amendment and I really don't know why it hasn't moved further into planned implementation. If anything that policy with amendments was more ready for ratification than the transfer policy.
>
> Paul
>
> _______________________________________________
> RPD mailing list
> RPD at afrinic.net
> https://lists.afrinic.net/mailman/listinfo/rpd
**********************************************
IPv4 is over
Are you ready for the new Internet ?
http://www.theipv6company.com
The IPv6 Company
This electronic message contains information which may be privileged or confidential. The information is intended to be for the exclusive use of the individual(s) named above and further non-explicilty authorized disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited and will be considered a criminal offense. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, even if partially, including attached files, is strictly prohibited, will be considered a criminal offense, so you must reply to the original sender to inform about this communication and delete it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20260524/73f04a42/attachment-0001.html>
More information about the RPD
mailing list