Robert.Ullmann@en-c06.x400.prime.COM (07/20/89)
[long; gets more interesting as it goes ... :-] Hi, [Mark Horton is ">"] > I don't believe there is a GB Internet domain. To quote Steve Kille, from a recent note: "For example, in the UK, under domain GB, we have (currently) defined: GOLD-400.GB <-> /ADMD=Gold 400/C=GB/ TMAILUK.GB <-> /ADMD=TMAILUK/C=GB/ AC.UK <-> /PRMD=UK.AC/ADMD=Gold 400/C=GB/" It certainly looks like the UK people intend to register one! > I don't believe there are ADMD's in the US called MA and the other > state names. Of course not, those are internet [UUCP zone] domains, the second level label does not have the semantic meaning of "ADMD" > I don't believe that all countries have set up their Internet domain > name spaces to look like OU.Org.PRMD.ADMD.Co . (In fact, I am not > aware of any country that has done so.) Right! They are _internet_ domains, not X.400 domains. The binding of the semantic meaning of "PRMD" to the 3rd level label is peculiar to X.400, and only meaningful inside an X.400 domain. In general (not specifically internet or X.400) the point of having heirarchical domains is to hide the structure and semantic meaning of the subdomain labels. In "Company.#.DIALCOM.US", an X.400 domain, "DIALCOM" is unquestionably an ADMD, and "Company" an "Organization". But in "Prime.COM.XA", it makes less sense to try to assign the meaning PRMD to the label "Prime", since it is not (within) an X.400 domain. The X.400 idea that the 3rd level label is a PRMD is directly comparable to the idea that "51" in the IP address [10.0.0.51] is an IMP/PSN number. It is, but only because net 10 is(was) an 1822 type network. To say that the "6" in [129.122.1.6] is an IMP is nonsense, net 129.122 is a different type (X.25 and ethernet). > And I don't believe that this gatewaying method has already been > tested for years and found to work well in practice on the existing > networks as a pragmatic solution. No, it hasn't. It has been prototyped, and does work so far as it has been tested. > I don't believe the X.400 world will honor requests from the XA > country to be routed to the Internet. Pete Lynch (then Prime UK), asked British Telecom about this. Their answer was that they weren't going to forward traffic for the internet, no matter what you called it; specifically they would not consider /C=US/A=COM/ [this is hearsay, and I can't confirm; Pete is now working for a start-up, and no longer with Prime] So we might as well call it something that won't conflict with any of the ISO-3166 X.400 domains used by countries, but still is technically within the protocol specification. so it won't confuse private MDs trying to handle both internet and X.400 domains. > Change /C=XA/ to /C=US/. The rest of the world has long told us that > they view the big six (COM, EDU, GOV, MIL, ORG, NET) domains as if they > had a .US on the end of them, anyway. With the exception of a limited > number of Canadian sites (which I think are moving to .CA) the only people > to register in these domains have been in the US anyway. This leaves the rest out in the cold. If we are going to ask the X.400 world to recognize the internet, we might as well solve the *whole* problem, and give ourselves a place to put other nets that might be out there. (i.e. things like JUNET, if it hasn't been totally transferred to .JP yet) In COM, there are lots of places outside the US; Prime.COM MTAs are located in 28 countries (22 of them connected, the others just named so when they are nothing must change). We would like to be one domain, and do not want that domain associated with the US (in particular, after 1992 in the EEC, we want to be seen as an international, or even European company, not as a U.S. company). Others like DEC, IBM, XEROX, HP are even more serious about this. ORG and NET are not by any definition U.S.: AMPR.ORG, BIT.NET, FIDONET.ORG. And the idea of being eu.net.us would surely surprise them! > Personally, I don't believe either buy-in option is very likely. Agreed, for the PTTs where they have a monopoly. In the U.S. it is very different: any service provider that doesn't play will be put out of business by their competition. (An interesting case-study is to compare the Sprint long distance service, before and after the breakup of the ATT monopoly.) The advantage of XA is that the private domains can use it to route to the internet. They won't (or where there is a monopoly can't) go through the service provider(s) anyway; as you point out: > agreements with some other vendors.", and "If we get incoming email from > the Internet, whom do we bill for it? We won't pass mail we can't bill for." So, for example, Prime computer sends traffic for the X.400 ADMDs to the appropriate ADMD (through our local office in that country) with an RFC987 magic-table-map if needed to get it accepted; we send mail to other PRMDs direct where we have connections; we send mail to the internet domains through our internet connection. Users within Prime, on X.400 or SMTP or uucp or (proprietary) OAS, see a simple, uniform address space that they can send to and reach *anywhere*. The reason for XA, rather than ad hoc (MIL, GOV, COM, EDU in US, NET and ORG somewhere, and what do you do with INT? saying it should include NATO.INT.US *is* silly!) is that it is one rule. If you can get an X.400 service provider to do an internet route, it is easier to get them to do C=XA, and handle all the others inside it, then it is to ask for six in US, and then the other 7 (8, 9 ...) in somewhere else. But much more important: it is USER-VISIBLE! It is possible to tell users on X.400 MUAs that when they see an address that has more than 2 letters in the last component they should add XA. (my business card, for example, says Ariel@Relay.Prime.COM) It is also possible to tell internet MUA users to leave off the .xa if there is one. It is even easier to have the MUA apply the rule. (If you like. Your MUA is your business!) But if .uucp is handled by one rule, .bitnet by another, "big six" by another, INT by ... (ad nauseum) it isn't practical any more. [Stef: "#"] # their PRMD name. We do not see the Commercial Community choosing to # operate PRMD=COM, since many of them will want their own PRMD anyway. # We expect that PRIME will want to do this, along with DEC, IBM, XEROX, # et al... In any case, the internet wizards cannot choose for them! Yes, we do want to be a PRMD, and it is convenient that Prime.COM.XA puts "Prime" at the level of PRMD (whether it has that semantic meaning depends on whether a given MTA is X.400 or not ...) Note that if I send mail to Kean College (for example), which is Kean.EDU.XA, Prime is expected to forward it to Kean, not through any intervening MD; this looks like a PRMD-PRMD connection to me. (likewise, this note went from Prime.COM.XA to LLNL.GOV.XA) In general, the level corresponding to the *role* PRMD is uniformly at the 3rd level using the XA mapping: Prime.COM.XA Kean.EDU.XA Army.MIL.XA NASA.GOV.XA cs.net.XA BITNIC.BITNET.XA NATO.INT.XA # But, the fact is that the decisions cannot be imposed by the internet # for purposes of simplifying the mappings because the Internet does not # have the required authority to impose any such decision. No it doesn't. But it certainly has the right to _propose_ such a mapping, and let others make their own decisions. # Yes, it would be nice if we could simply dictate this sort of thing, but # then we would probably not like the choices made by the appointed # dictator. I think it is safe to expect that we would not be named to # this exalted position. Quite right; but we can agree among ourselves on something, and propose it to everyone else; if we have a good method/answer/solution, they will probably listen. This is the way the entire internet was built. (yes, except MILNET; but that was done by dictating standards that were developed by the informal process). With one exception, every single person who has commented to me has said something along the lines of: "You can't do that, there is no Authority to Authorize it!" I find this a very European view. For myself, as an [expletive deleted] Yank, I put exactly the same fact this way: "There is no Authority. Therefore there isn't anybody with the authority to tell us we can't do it!" So what you see as a fatal flaw, I see as the most important reason that it can be made to work. Robert Ullmann work: Ariel@Relay.Prime.COM.XA home: semantic!xa!ariel@ssgp32.prime.com.xa