Fw: [address-policy-wg] 2005-08 New Draft Documentis Published (Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy)
Hi, Both of these proposals change the onus on an LIR from having "... a plan for making at least 200 /48 assignments to other organisations within two years." to having "... a plan for making at least 200 assignments to other organisations within two years." While this increases flexibility it still has an arbritrary feel to it. Why 200 assignments? I can see that this is meant to prevent the numberspace becoming too fractured, so reducing the size of the routing table. However, I worry that it may have the side effect of imposing a hierarchy of ISPs within IPv6 that does not meet the requirements of smaller organisations. It seems there is little interest at present within the larger UK-based ISPs in providing IPv6, (with the exception of NTT-Verio), and this change won't affect that. Is there a better way to encourage the uptake of IPv6 without preventing exponential routing table growth? Ian ----- Forwarded by Ian Meikle/Nominet on 27/07/06 15:53 ----- Filiz Yilmaz <filiz@ripe.net> Sent by: address-policy-wg-admin@ripe.net 27/07/06 12:42 Please respond to filiz@ripe.net To policy-announce@ripe.net cc Hans Petter Holen <hph@oslo.net>, Kurtis Lindqvist <kurtis@kurtis.pp.se>, Geoff Huston <gih@apnic.net>, address-policy-wg@ripe.net Subject [address-policy-wg] 2005-08 New Draft Documentis Published (Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy) PDP Number: 2005-08 Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy Dear Colleagues As you might remember, it was decided during RIPE 51 that the proposal described in 2005-08, "Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy" should be split into two parts. Accordingly, we have published two draft documents. You can find the draft documents at: http://www.ripe.net/ripe/draft-documents/2005-08-56s.html defining allocation efficiency measurement unit as /56 and http://www.ripe.net/ripe/draft-documents/2005-08-hd-ratio.html changing the HD ratio value to 0.94 You can find the full proposal at: http://ripe.net/ripe/policies/proposals/2005-08.html We encourage you to read the draft document text and send any comments to address-policy-wg@ripe.net before 24 August 2006. Kind regards, Filiz Yilmaz RIPE NCC Policy Development Officer
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi. Don't think 200 /48 is totally unrealistic. There is a point in keeping the global ipv6 routing table small. Things however depends on how rapidly the industry will adapt ipv6 into their equipment. Also In the near future we will without a doubt see more connected household appliances. At the moment It's really, really hard to predict growth rate. I would say 200 /48 within 2 years.. Maybe, maybe not :-) Best regards. - --Dennis Lundström GippNET AB (AS34537) Ian.Meikle@nominet.org.uk wrote:
Hi,
Both of these proposals change the onus on an LIR from having "... a plan for making at least 200 /48 assignments to other organisations within two years." to having "... a plan for making at least 200 assignments to other organisations within two years."
While this increases flexibility it still has an arbritrary feel to it. Why 200 assignments?
I can see that this is meant to prevent the numberspace becoming too fractured, so reducing the size of the routing table. However, I worry that it may have the side effect of imposing a hierarchy of ISPs within IPv6 that does not meet the requirements of smaller organisations. It seems there is little interest at present within the larger UK-based ISPs in providing IPv6, (with the exception of NTT-Verio), and this change won't affect that. Is there a better way to encourage the uptake of IPv6 without preventing exponential routing table growth?
Ian
----- Forwarded by Ian Meikle/Nominet on 27/07/06 15:53 -----
Filiz Yilmaz <filiz@ripe.net> Sent by: address-policy-wg-admin@ripe.net 27/07/06 12:42 Please respond to filiz@ripe.net
To policy-announce@ripe.net cc Hans Petter Holen <hph@oslo.net>, Kurtis Lindqvist <kurtis@kurtis.pp.se>, Geoff Huston <gih@apnic.net>, address-policy-wg@ripe.net Subject [address-policy-wg] 2005-08 New Draft Documentis Published (Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy)
PDP Number: 2005-08 Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy
Dear Colleagues
As you might remember, it was decided during RIPE 51 that the proposal described in 2005-08, "Proposal to Amend the IPv6 Assignment and Utilisation Requirement Policy" should be split into two parts.
Accordingly, we have published two draft documents.
You can find the draft documents at:
http://www.ripe.net/ripe/draft-documents/2005-08-56s.html defining allocation efficiency measurement unit as /56
and
http://www.ripe.net/ripe/draft-documents/2005-08-hd-ratio.html changing the HD ratio value to 0.94
You can find the full proposal at:
http://ripe.net/ripe/policies/proposals/2005-08.html
We encourage you to read the draft document text and send any comments to address-policy-wg@ripe.net before 24 August 2006.
Kind regards,
Filiz Yilmaz RIPE NCC Policy Development Officer
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD4DBQFEye3EsqJZaeZjsn8RAlyUAJdVueQfvepLitYfGUaND84k/Ov9AJ4m13Yn tp49EmYUzMbea4NeCLwuvg== =pHiK -----END PGP SIGNATURE-----
address-policy-wg-admin@ripe.net wrote on 28/07/2006 11:58:12:
Hi. Don't think 200 /48 is totally unrealistic. There is a point in keeping the global ipv6 routing table small. Things however depends on how rapidly the industry will adapt ipv6 into their equipment. Also In the near future we will without a doubt see more connected household appliances. At the moment It's really, really hard to predict growth rate. I would say 200 /48 within 2 years.. Maybe, maybe not :-)
Best regards.
- --Dennis Lundström GippNET AB (AS34537)
Any figure appears arbitrary, and risks imposing a business model on those who want to deploy IPv6 that does not fit their organisation. On the other hand, the lack of a metric makes things harder for RIRs, and I can see why one has been specified, besides the comments already made regarding the size of the global routing table. IPv6 is a hard sell. There is no application that requires it, and nobody worries enough about resource exhaustion until it's too late. This is as true for IPv4 as it is for oil. If we want to encourage adoption then we shouldn't put unreasonable barriers in place. Ian
While this increases flexibility it still has an arbritrary feel to it. Why 200 assignments?
Please see: http://www.ripe.net/ripe/policies/proposals/2006-02.html The intention is to replace the completely arbitrary "200 /48's" with "a reasonable number of /48's". In fact, they mean the same thing, because any LIR can have a *plan* to assign 200 x /48s, regardless of whether this plan is ever going to be implemented or not. But this is not particularly relevant to proposal 2005-08. Nick
Hi Nick, address-policy-wg-admin@ripe.net wrote on 29/07/2006 09:43:17:
While this increases flexibility it still has an arbritrary feel to it. Why 200 assignments?
Please see:
http://www.ripe.net/ripe/policies/proposals/2006-02.html
The intention is to replace the completely arbitrary "200 /48's" with "a
reasonable number of /48's". In fact, they mean the same thing, because
any LIR can have a *plan* to assign 200 x /48s, regardless of whether this plan is ever going to be implemented or not.
I agree with the change in wording in 2006-02, but I don't think they amount to the same thing. The 200 x /48s acts as a psychological barrier preventing many people from asking for IPv6, and none of us wants to lie to the hostmasters, do we?
But this is not particularly relevant to proposal 2005-08.
True, and I felt I should point out that I didn't support that part of 2005-08. Ian
Hi Ian! Ian.Meikle@nominet.org.uk wrote:
I agree with the change in wording in 2006-02, but I don't think they amount to the same thing. The 200 x /48s acts as a psychological barrier preventing many people from asking for IPv6, and none of us wants to lie to the hostmasters, do we?
If we are taking actions to pervent people asking (so - using) IPv6 when it is only a funny toy now - why and how it will be in real life production in the future? P.S. Does RIPE/RIPE NCC have "PR-department" or so to make policies, adversitement, press-releases, propagation, etc for all of the new things and technologies invented by RIPE? If not, why? If yes - why I don't hear anything about it? ;) Invent a cool thing is only a half of a deal, the second half is make world beleave in it! -- WBR, Maxim V. Tulyev (MT6561-RIPE, 2:463/253@FIDO)
Hi Ian! Ian.Meikle@nominet.org.uk wrote:
I agree with the change in wording in 2006-02, but I don't think they amount to the same thing. The 200 x /48s acts as a psychological barrier preventing many people from asking for IPv6, and none of us wants to lie to the hostmasters, do we?
If we are taking actions to pervent people asking (so - using) IPv6 when it is only a funny toy now - why and how it will be in real life production in the future? P.S. Does RIPE/RIPE NCC have "PR-department" or so to make policies, adversitement, press-releases, propagation, etc for all of the new things and technologies invented by RIPE? If not, why? If yes - why I don't hear anything about it? ;) Invent a cool thing is only a half of a deal, the second half is make world believe in it! -- WBR, Maxim V. Tulyev (MT6561-RIPE, 2:463/253@FIDO)
participants (4)
-
Dennis Lundström -
Ian.Meikle＠nominet.org.uk -
Max Tulyev -
Nick Hilliard