On Tue, Sep 11, 2012 at 10:07 PM, Joe Touch <[email protected]> wrote: > > > On Sep 11, 2012, at 11:31 PM, Pars Mutaf <[email protected]> wrote: > > Hi all, > > Instead of IP translator, I use the new term "IP payload copier". It > copies payload between different IP versions. TCP does not work on this > because it assumes same IP version for source and destination. > > We probably need a new host identifier, > > > And that is IPv6 right now. > > Call us when you fix that. and maybe also look at realm-specific IP. > > and change TCP to use this identifier but the IP version transition > problem is solved forever. Everybody use the IP version of their > preference. This is the end-to-end princple, we do not care about the core > Internet, we change the end-hosts. > > See details in the paper: > > http://www.scribd.com/doc/105448105/Discrete-IP > > > There are no details. > > E.g., let's say I'm ipv4 and want to talk to you who are ipv6 right now. > So I go to the DNS and ask for you. First, you have to assume a global > namespace above IP - the DNS name - the identifies endpoints. Second, what > will you give me? I don't understand an ipv6 address. So you hand me the > ipv4 of the translator? Then if I send a packet there, it won't know who to > send it onward to. >
All these are explained in the paper Joe. I tunnel the first packet to the copier. (I don't use the term translator to avoid confusion) Why do you refuse to *read it?* > > You have a paper, but it is lacking in exactly the solutions everyone has > been trying to find for years, and some have shown don't or can't exist. > > So I agree with some others that if you insist a solution can exist, you > need to demonstrate that with code..that is how the IETF works when the > community is skeptical of your approach, which is the case here. > This is just loss of time. There are many IETF protocols that are designed first, implemented later. Ex: IPv6, MIPv6 etc. So... In addition, as I said earlier, having running code doesn't mean that it is a good idea. The correct approach is: 1. Decide what we want. 2. See if it is feasible. If it is not, we know that what we want is not feasible. (I would be surprised because anything is feasible) In my opinion Discrete IP is feasible. The paper (you did not read before entering this discussion) gives enough evidence. But if you look for unnecessary details like running code you cannot see this. > > Joe >
_______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
