May introduce a user opinion here? End-users go by the box they (are to) buy. Lead-users go by the box they enhance. - in the first case there should be a clear table of functions by a clear list of box names, indicating which function is included or not in each box name. - in the second case, the number of functions to be supported is limitless, so the name is to be generic and refer to the location of the box in the network. The name IUCG uses is "interbox".
So, an interbox is a generic evolutive physical/logical box that can gather many smart gateway functions that are documented as physical/logical specific named boxes. This makes then sense then that each new function can be specified and tested under a box name (NAT66, NAT64, STAT66, etc.) and/or encapuslated in interboxes. People interboxes are/will be understood as the permanent IP gateway to whatever IPv4/6 network system they may have set-up under IPv4 or IPv6. They will most probably include services such as multilevel addressing and naming management, registry servers, firewall, ONES/OPES shims, DDDS semantic converters, [GR]IDv6 routers, until we have a true network interoperation system (what we call Netix, a concept users understand easily), etc. IMHO if you want people to buy IPv6 you have to proceed as sales do: - first to sell yourselves: make people accept you are credible, - then to make them dream about what they will be able to do with your proposition (be extremly clear about what they can dream of, because they have to get aroused by what they feel they can do more with the next box they get, - then do not "violate" too much: make the transition easy in selling the box to the ISP, users will have to be incitated into the change by the "plus" their ISP will propose them. This is what IUCG calls "Internet PLUS" (parallel layered users systemic) more functions being permitted at each layer due to forgeting past constraints the user is not interested in. Until we extend the whole protocol pile to support the new functions. - finaly make them feel they have made the good choice in accepting your proposition (that means that the transition is seamless and that they get more for their hassle of changing their system, than they expected). Please, do not refer to anything old. People do not know about NATs, they know about their ISP smart boxes. What they need to hear is about a general new generation boxes, i.e. smart boxes becoming smart gateways. Do not talk of "new internet", people need to feel secure and that the "new generation internet" is the old internet they know and their friends use. Please also remember that people are really important in all this because they are the demand and the payers who will oblige providers to move. And the market which will decide if your propositions are good. Users do not care about IPv6. They never cared about IP, v4 or as V6, because they never had a permanent IP address. So, there are only two arguments they can be pleased with: 1. IPv6 means a permanent dedicated network number like the phone (but not yet like the mobile, because "they" did not organised properly the numbering plan yet) [talking of address to users confuses them, talking of number is easier for them to understand and feel "confortable" with - the key word] 2. IPv6 means that they also can have billions of IIDs they can play with by themselves, for example through IPv6 supported "IDv6", or with other with general regions of IDv6 (GRIDv6), starting with plug and play, network translation (as in memory translations), semantic access engines and addresses, etc. What is important for the users is not what your boxes could do, but what they feel confortable they can do themselves if they invest in your boxes. This means that Rémi Desprès is totally right in terms of semantic need, but has an obvious difficulty to propose something that both users (me) and IETF history (Margaret) can identify as the same thing. I hope it helps. jfc 2009/4/1 Margaret Wasserman <[email protected]> > > Hi Remi, > > On Apr 1, 2009, at 11:57 AM, Rémi Després wrote: > >> >> That said, I think Remi has made a good suggestion here. Calling it >>> Stateless Address Translation makes sense, I think. >>> >> >> I do believe it will help if we can make such a change as early as >> possible. >> > > You believe that changing the name will help _what_? > > At this point, we have made a proposal to the IETF for an IPv6-to-IPv6 NAT, > and we've called it NAT66. If an IETF WG takes on this proposal and has > consensus to change its name, that's totally fine with me. Right now, > though, there are a few people who want to change the name (although there > isn't agreement about what we should call it instead) and there are a few > people who want to keep the name the same. So, I don't see any consensus to > change the name, nor have I been offered a compelling reason to do so. > > Perhaps it would be better to focus this discussion on the technical > aspects of this proposal, instead of focusing on its name? > > Margaret > > > > > > _______________________________________________ > nat66 mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/nat66 >
_______________________________________________ nat66 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nat66
