Fw: how 200 /48's fails the job [Re: [address-policy-wg] Policy proposal: #gamma IPv6 Initial Allocation Criteria]
There is one point I don't understand in the whole discussion: If every RIPE member get's an IPv6 prefix, which is true for IPv4, we are talking about plus 10K prefixes in the table.
If RIPE really and truly believes that IPv6 will become the future core protocol of the public Internet, then RIPE should allocate an IPv6 /32 to every RIPE member who has PI addresses. But that is seperate from the question of new entrants who may, or may not, be following an ISP business model.
Do people *really* think this approach works and do they really think that such an anti-competitive 200 customer policy - does neither hit IPv6 and the idea behind it - does not hit community - will not be forced down by EU commission authorities ?!?
If you want to discuss anti-competitive rules, then how about the 80% threshold which fails to recognize the technical realities of IPv4 subnetting which means that large networks can never be as efficient as smaller ones. But that is being addressed by the HD-Ratio proposal. --Michael Dillon
* Michael Dillon:
If RIPE really and truly believes that IPv6 will become the future core protocol of the public Internet, then RIPE should allocate an IPv6 /32 to every RIPE member who has PI addresses.
No, if everyone believed that IPv6 is the future, policies would not matter much, and there would be little fighting. Everyone would jump through almost any hoop to get what they think they need. 8-) But this is not the case. I don't follow the v6 wars closely, but it appears that several promised improvements over v4 won't be delivered (look at the A6/bitlabel/DNAME deprecation, or even the protocol design optimized for forwarding implementations which now are being phased out).
On Wed, 2005-04-06 at 21:10 +0200, Florian Weimer wrote:
* Michael Dillon:
If RIPE really and truly believes that IPv6 will become the future core protocol of the public Internet, then RIPE should allocate an IPv6 /32 to every RIPE member who has PI addresses.
No, if everyone believed that IPv6 is the future, policies would not matter much, and there would be little fighting. Everyone would jump through almost any hoop to get what they think they need. 8-)
But this is not the case. I don't follow the v6 wars closely, but it appears that several promised improvements over v4 won't be delivered (look at the A6/bitlabel/DNAME deprecation, or even the protocol design optimized for forwarding implementations which now are being phased out).
A6 could cause a long chain of servers to be asked where 1 lookup is enough for normal reverse queries. Also things as signing would become a large Bitlabel is just a different way of writing down reverses, which btw would be incompatible with A6. DNAME exists and is being used. It is sort of a Domain CNAME :) How else did you think I aliased ip6.int to ip6.arpa for silly slow people who do not upgrade their DNS resolvers. The "protocol design optimized for forwarding implementations". You can be referring to a couple of things here, though I can tell you that the header structure is aligned and there is no checksumming anymore, as the hardware layer usually already does that. This speeds IPv6 up already by a couple of factors compared to IPv4, even though one has to look at a 128bit address instead of a IPv4 one. Anything else? Greets, Jeroen
participants (3)
-
Florian Weimer -
Jeroen Massar -
Michael.Dillon＠radianz.com