[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RRG] Moving forward...



On 6/6/08 1:33 PM, Joel M. Halpern allegedly wrote:
I disagree. I don't object to having a solution that also helps IPv4, but I do not think it is as important as getting something that works for IPv6.

We can keep the IPv4 net working for quite a while.

The point is, you don't have a choice. What year do you think you will be able to stop making services available on IPv4?

> Particularly if we
can provide some reason for folks to believe that v6 will matter.

iiuyc you are saying you are willing to support IPv4 if you can demonstrate to people that IPv6 is better. What if you can't demonstrate that? Are you going to stop supporting IPv4 in order to force people to move?

> If
there is some actual benefit to the users (like better multi-homing) from IPv6, then not only will folks move towards it, but folks will believe that it is likely to matter, which reinforces the effect. This in turn should help keep the pressure on v4 within the manageable range.

It's good to try to offer a better experience on IPv6, otherwise what's the point of IPv6 at all? However, if you think that will remove the state*rate problem we are developing, I believe you are being overly optimistic.

Note that trying to do this without solving the architectural probloems will merely result in v6 continuing to be ignored. Or worse, v6 becoming an even worse case of the problems we have with v4.

If you only support the new architecture with IPv6, then you will be forcing people to move, but they won't be able to do so quickly. And by the time we are able to successfully deploy IPv6 to any degree, there will be a significant expansion of IPv4 routing. You will be inflicting suffering in order to get them to do what you want. Can you tell that doesn't feel right to me?

I am not adamant that IPv4 MUST be supported equally. If a design is clearly superior in a number of ways except that it cannot effectively support IPv4, then the benefits may outweigh the costs. But I see clearly that it SHOULD be supported at least as well as IPv6.

And best of all is a design that allows us to deploy IPv10 a little more easily than the job we'll have with IPv6.

Scott

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg