Arnaud,

This has NOTHING to do with if we trust AfriNIC or not, it comes down to if 
commercial companies trust each other.

NOTHING in your proposal stops organisation X who does not like organisation Y, 
going to AfriNIC and under the policy demanding that AfriNIC audit organisation 
Y.

That drains AfriNIC resources and puts IMMENSE pressure on organisation Y 
because audits take time, effort and resources, all of which cost both AfriNIC 
and organisation Y in time, which equates to money.

Has *ANYONE* proposing this policy actually been involved in a detailed audit 
of IP resources across a large organisation and understands what it entails?  
*Laugh* I have – and if you do it properly, you’re looking at tying up 
resources in organisation Y potentially for *weeks* if they want to verify 
every IP address and its usage.

Why are the proposers not willing to make the policy safer and give everyone in 
this community the assurances that they have recourse should the policy be used 
maliciously by forcing disclosure of who requested the audit?  Or are there 
other motives at play here which aren’t being stated?

You are right – proxies aren’t allowed at PDP – except in this case, they are 
not proxies.  I am the direct administrative contact for those organisations 
and hence I speak for each of them, one organisation at a time.

Andrew


From: Arnaud AMELINA [mailto:[email protected]]
Sent: 01 November 2016 20:29
To: Andrew Alston <[email protected]>
Cc: ALAIN AINA <[email protected]>; General Discussions of AFRINIC 
<[email protected]>
Subject: Re: [Community-Discuss] IPv4 depletion in AFRINIC will speed up IPv6 
adoption - myth or fact?



2016-11-01 16:39 GMT+00:00 Andrew Alston 
<[email protected]<mailto:[email protected]>>:
Arnaud,

Let me be very clear –

I never expressed support for this policy – I asked a question.

I am COMPLETELY opposed to this policy, and will be until such time as the 
policy explicitly states that in the event of a complaint against a member, the 
member is informed when he is audited who complained and the grounds stated for 
the complaint.  That is in line with internationally accepted practice that the 
accused may have the right to confront his accuser, and protects against 
malicious accusations for the sake of causing on going work for an organisation.

It opens the door for legal action against an accuser by the accused, if the 
accused feels the accuser is lodging complaints that are malicious and 
unsubstantiated in nature, and it means that those who are complaining and 
demanding audits are doing so *backed by evidence* rather than because they 
feel they can.

I also argue that any person who issues a complaint should be willing to 
subject themselves to automatic audit, which again, will cut down on unfounded 
complaints that are not backed by evidence.

In addition to this, complaints should be submitted by *members* alone, they 
are the true stake holders, and I will oppose any policy that allows a random 
member of the community to submit arbitrary complaints without proof, evidence 
or proper substantiation.


Andrew,
The relationship between the RIR and its members is based on trust.  I feel you 
miss the intent of this proposal and imagine too many negative things.   A 
network truly using resources should not care.  I also feel you confuse role of 
policy and Afrinic staff who have responsibility to propose a process in 
accordance with the RSA.

Further to this, 3.4 makes reference to unauthorised transfers.  I can issue 
with auditing against something that is unauthorised when there is no process 
to authorise it.  That needs re-wording until such time as there is a transfer 
policy.   I also take issue with 3.4 that it is declared as a non-exhaustive 
list.  That is an open door for abuse, define the reasons for audit and stick 
to them.

I also take *HUGE* issue with 3.6 – if you want 3.6 and you want to publish the 
names of companies that have been audited, firstly, you have to publish the 
names as well of the requesting party, and secondly, that would be a direct 
violation of the confidentiality agreements agreed to in the RSA which you 
yourself refer to in the policy.

Those are my thoughts – but until those issues are rectified – I do not support 
this policy, nor do any of the 15 organisations which I represent directly 
support this policy.

Oooh! I am not aware that we can have proxy for PDP discussions too.

Once those issues are rectified, I will *consider* support for the policy, but 
the above points are deal breakers for me – after that, consideration can be 
given.


--Arnaud

Andrew



From: Arnaud AMELINA [mailto:[email protected]<mailto:[email protected]>]
Sent: 01 November 2016 19:10
To: ALAIN AINA <[email protected]<mailto:[email protected]>>
Cc: General Discussions of AFRINIC 
<[email protected]<mailto:[email protected]>>
Subject: Re: [Community-Discuss] IPv4 depletion in AFRINIC will speed up IPv6 
adoption - myth or fact?

My Contribution through the lines of this message

2016-11-01 15:31 GMT+00:00 ALAIN AINA 
<[email protected]<mailto:[email protected]>>:
Hello,

On Oct 29, 2016, at 7:18 PM, Andrew Alston 
<[email protected]<mailto:[email protected]>> wrote:



>  Yes. the usual story. You only know. Others are either clueless or naive..

Not at all, I’m sure there are plenty of people who may know better than me.  
Unfortunately, as of yet, none of them have bothered to provide realistic ways 
of doing this that contain *detailed* proposals of how this would be 
accomplished and what the end results are.

There are plenty of people here who know far more than me and you , but resolve 
not to speak as you have made this a low floor, repetitive and non  productive 
discussions.

The policy proposal was introduced  on the 18 May 2016, and we have had  these 
discussions. RPD  and the  AFRINIC-24 policy archives are available.

+1 @Alain

Yes Andrew, we had good discussions  and a new version of the proposal  which  
incorporated  the changes  was posted  Tuesday, 9 August 2016, the link below :

http://www.afrinic.net/en/community/policy-development/policy-proposals/1827-internet-number-resources-review-by-afrinic

And I understood trough your e-mail posted on 14 October 2016, which is cuts 
below, you finally express support to this Policy proposal.
====
"Just a question about the audit policy….

Is it agreed that if we have such a policy, we should also audit the v6 
assignments people are holding that should be announced under the needs based 
policy rules?

Thanks"

===
regards


Now  that you have asked again, see below...


See Alain, the difference here, I ask for hard facts and data – and when I’m 
asked for such I provide it – but I will not accept vague positions and 
unsubstantiated nonsense as the grounds for implementing a policy.


a.)    No one has yet proposed how these audits are meant to be realistically 
done beyond looking at the routing tables

Policy proposals do not  dictate implementation, but describe principles to 
action.

The policy  proposal aims to seek compliance to RSA which all members sign 
before applying for the number ressources. Section 4 of the RSA is very clear 
on parties responsibilities. http://www.afrinic.net/en/services/rs/rsa

The policy proposal  says :

===

3.4 In case of non-compliance and if evidence has been established in 
accordance with the non-exhaustive list below:

  *   Unjustified lack of visibility of the resource on the global routing 
table.
  *   Breach of AFRINIC policies.
  *   Breach of the provisions of the registration service agreement or other 
legal agreements between the organization holding the resource and AFRINIC.
  *   Evidence that an organisation is no more operating and its blocks have 
not been transferred.
  *   Unauthorized transfers of resources.
===

Looking at the global  Internet routing table at a given time is an option. 
Visibility or not in the global Internet Routing Table gives an indication of 
how to reach the destination, but does not tell about utilisation. A prefix can 
easily be seen in the Global Internet Table without  being used.
 Utilisation in compliance with RSA and policies is what is sought here.  What 
AFRINIC will be trying to establish is utilisation based on justified  needs  
and compliance with RSA. In doing so, Members are  bound to collaborate with 
AFRINIC as said in section 4.(4) below.


===

(b) Cooperation:

(i) An applicant receiving service under an agreement is at all times bound to 
provide to AFRINIC such information, assistance and cooperation as may be 
reasonably required by the latter in the provision of the service.

(ii) Such request for information may also be made where AFRINIC is 
investigating (reviewing) the applicant's utilisation of the numbering 
resources already assigned to it.

(iii) Failure by the applicant, to comply with a request made at above may:

  1.  entail revocation or withholding of the service supplied by AFRINIC;
  2.  be taken into account by AFRINIC in its evaluation for further and future 
assignment or allocation of numbering resources;
  3.  lead to the closure of an LIR and termination of the agreement by AFRINIC.

===

Investigated members to provide  information and data to convince AFRINIC which 
may not need to do  much.

“Say what you do, do what you say and prove it."



b.)    No one has proposed where the resources to do these audits are meant to 
come from


AFRINIC as RIR is already committed to do this review as prescribed in the RSA. 
This proposal is just  guidelines on how to implement it.  But if there is a 
need for extra ressources, it is up to  AFRINIC staff to say so. The PDP has 
provision for staff analysis on Policy proposal. Shall co-chairs request one ?



c.)    No one has addressed the MASSIVE potential for abuse of this policy


I remembered  the discussions on possible abuse on  the “reported” class of the 
policy proposal.

===
3.3.3 Reported: Here, members are reviewed either because:

  1.  They have requested the review themselves or
  2.  there has been a community complaint made against them that warrants 
investigation.
===

This has been addressed as we all trust AFRINIC to do the right job and as a 
community, we stand to revive policies and implementations in case of known 
abuses.

People have been complaining  about flaws in current policies, practices which 
have been abused by some members.


d.)    No one has addressed the fact that if an audit is needed – then the 
original documentation is in question – and at that point you are by the very 
nature of requesting the audit saying the hostmasters didn’t do their jobs when 
verifying the application in the first place – and if you want to make 
implications like that – you need evidence.


The application approved by Hostmasters at the first place serves as basis to 
all reviews. Questioning  applications approved by Hostmasters is a different 
matter.


I am sick and tired of vague statements, vague insinuations, vague claims that 
everyone is stealing the space and taking it elsewhere,

Ah bon ? I have not heard that.  But in case this exist, I would expect this 
policy proposal to help clear the point.

vague claims that presentations that report on one thing some how are actually 
reporting on other issues they don’t ever reference.

There are data in there, plus some conclusions. Nothing prevents further 
analysis and readings of the data.

—Alain


Come with real data – real facts – real figures – and then let’s have this 
discussion.

Andrew



_______________________________________________
Community-Discuss mailing list
[email protected]<mailto:[email protected]>
https://lists.afrinic.net/mailman/listinfo/community-discuss


_______________________________________________
Community-Discuss mailing list
[email protected]
https://lists.afrinic.net/mailman/listinfo/community-discuss

Reply via email to