Search RPD Archives
[rpd] New Draft Policy Proposal - Update of PDP AFPUB-2026-GEN-005-DRAFT01
Benson Muite
benson_muite at emailplus.org
Wed Nov 18 18:59:57 UTC 2026
"jordi.palet--- via RPD" <rpd at afrinic.net> writes:
Hi Jordi,
Thanks for your reply.
> Hi Benson,
>
> Sorry the late response, too busy with IETF matters and regular day job to keep up with the list this week and the previous one.
>
> Some responses below, in-line.
>
> Regards,
> Jordi
>
> @jordipalet
>
>> El 13 nov 2026, a las 19:11, Benson Muite <benson_muite at emailplus.org> escribió:
>>
>> Hytham El-Nakhal <hytham at tra.gov.eg> writes:
>>
>>> Dear PDWG,
>>>
>>>
>>> We have received a new draft policy proposal named "Update of PDP" with ID: AFPUB-2026-GEN-005-DRAFT01.
>>>
>>>
>>> The proposal contents are published at:
>>> https://www.afrinic.net/afpub-2026-gen-005-draft01
>>>
>>>
>>> This proposal is an update to the expired proposal with ID: AFPUB-2021-GEN-002-DRAFT03 https://www.afrinic.net/update-of-pdp-draft-3x . The version history of the current proposal will allow you to consult the past versions for reference purposes.
>>>
>>>
>>> We encourage you to take some time to go through the proposal contents and provide feedback as follows:
>>>
>>>
>>> * a) Do you support or oppose the proposal?
>>
>> The proposal will improve how the policy development process, but may
>> need adjustments.
>>
>>>
>>> * b) If you oppose the proposal, state your reasons.
>>
One part that I am opposed to is the vague identity verification. At
present, should there be a data leak on the part of AfriNIC, only people
in Mauritius would have easy recourse to legal action, or in case of
appeal to higher courts, people in the United Kingdom. Thus collecting
ID information is problematic. Furthermore, some people who use the
internet may not have government issued identity documents, or even
worse, may have multiple government issued identity documents.
I did not present today because AfriNIC asks people presenting to upload
identity documents to their servers. I have no idea who runs these
servers, and who else has access to data stored on these servers. The
IETF does not ask for any identity documents to take part in their
meetings online, though of course if one needed to travel and required a
visa, one might need to provide a passport to get a visa support letter.
It seems that even though one person one vote is not followed in the
PDP, there is the hope of one person one voice. At present, given that
there are 2000 resource members and typical PPM attendance is less than
100, clear and effective communication rather than identity verification
would likely be more important. Public key cryptography does allow one
to tie an email address to an identity. Resource members make payments,
so requiring a public key from a reource member when making a payment is
one way of verifying identity. Other participants could make a small
token payment e.g. $1 or be required to send an SMS as a way of
verifying that there is a person - this could still be gamed, just like
identity documents - but the aim should be to have authentic productive
discussion without preventing interested parties from participating and
without unnecessary data harvesting.
>>
>>>
>>> * c) Is there anything in the proposal that is not clear?
>>
>> The definition of consensus could be improved. Many policies have both
>> technical and social considerations. If a policy proposal is not
>> technologically feasible, then it cannot be implemented. There may
>> however be proposals that are possible technically, but would do much
>> soical harm. It is not clear that the current definition of consensus
>> could accomodate such a situation.
>>
>
> The definition of rought consensus in the RIRs comes from IETF RFC7282. However, the PDP doesn’t explain it. The reason why the definition of consensus is included in the proposal, is so there is a short summary. We aren’t changing the way consensus is determined by the chairs, just making a short summary so everyone is aware of that.
>
> I agree with your point: However, in the end all the decisions have a technical part (for example, we can’t allocate more IPv4 addresses if none are left in the pools, so we need to have a soft landing when time comes, so there is no social harm).
>
>>>
>>> * d) What changes could be made to this proposal to make it more effective?
>>
>> The suggestions for time periods for people to discuss a proposal are
>> helpful. A 5 minute discussion to determine if consensus has been
>> reached either forces proposals to go through two PPMs or requires very
>> fast decision making that maybe liable to errors.
>
> Yes, this is a key point. 3 out of 5 RIRs are already doing that without asking the chairs to rush in decisions in 5 minutes (which also steals precious time from the agenda). Only AFRINIC and APNIC still do that. LACNIC was doing that until a proposal like this one changed it, and it has been successful for 7-8 years already. In IETF and other sister organizations is also not done in a rush.
>
>>
>> Section 3.4.5 could be updated to account for the fact that an online
>> PPM can be held to get community feedback. In most cases, scheduling an
>> online PPM within a month is reasonable. There can be situations that
>> prevent an online PPM, afpub-2026-gen-001 tries to address some of
>> these.
>>
>
> I’m not sure we want a PPM for just community feedback, for that we have the webinars, and of course, they can be scheduled in a shorter period of time, as it has been always the case.
>
If the board is making an important decision that affects the community,
then a presentation at least at one PPM seems to be norm for community
feedback that might not be picked up on the mailing list. This could be
revised, but waiting potentially upto 6 months to rectify an emergency
decision when one can use the internet seems inefficient.
More information about the RPD
mailing list