jf@ap.co.umist.ac.uk (John Forrest) (11/16/90)
Neil Rickert has very kindly done some mods to the IDA cf file for me. The aim of this was to be able to conditionally reverse incomming mail addresses if required [remember, traditionally in the UK we do this backwards]. As an example, our relay might send us local names looking like "jf@uk.ac.umist.co.ap" which we reverse to give "jf@ap.co.umist.ac.uk". I suggested two specifications to him (depending on the required functionality locally). Spec 1 - Cope with incomming mail with reversed names. It after looking up the name in the various forms, the last element of the address is not a "top domain" (in CT?), and the first element of the address (after @) is, reverse and try again. Spec 2 - Allow local users to type mail reversed addresses. If after looking up the name in the various forms, the last element is not a "top domain", reverse and try again. He has implemented Spec 1, explaining that Spec 2 requires non-trivial modifications. We have so far not used the mods in anger, but they have worked in testing (both in -bt mode, and sample outgoing mail). However, we have suddenly realised that the function described is flawed, for the reasons several people here are finding out. Basically, consider: uk.ac.ucl.cs If we had this as an incomming mail, we would interpret this as being somewhere in Czechoslovakia, not as Computer Science at University College, London. Note we don't have direct access to the Internet, and our local name server contains (or will soon contain): *.cs. MX 0 ...... so we just assume anything in "cs." is to be relayed. Has anyone got a better algorithm that the above - really to detect whether the name is valid. The only things I can think of are: 1) Hardwire all the problem cases in one of the databases. 2) Let's give up sending reversed addresses around the UK. (please?) John Forrest Dept of Computation UMIST.
iiitsh@cybaswan.UUCP (Steve Hosgood) (11/19/90)
>so we just assume anything in "cs." is to be relayed. Has anyone got a better >algorithm that the above - really to detect whether the name is valid. The only >things I can think of are: > > 1) Hardwire all the problem cases in one of the databases. > 2) Let's give up sending reversed addresses around the UK. (please?) > As my old computer science tutor used to say: Option 1 (above) is rather like walking along the road and finding that because you've got one foot in the gutter and the other on the walkway, that you're going up and down with every step. You examine the situation and decide to strap a block of wood under the foot that's in the gutter so that you can walk level again. This stinks! Why don't we just scrub the stupid system? The amount of time I spend hot-wiring addresses to get them through, and explaining the system to others. Support option 2 (above) now! Surely no-one would object if we (JANET) did this? Steve | WALES: "Land of Song iiitsh@pyr.swan.ac.uk | and Rugby^H^H^H^H^H ..or in Britain, where we drive on the other side:| Ice Hockey" iiitsh@uk.ac.swan.pyr |
brian@ucsd.Edu (Brian Kantor) (11/19/90)
I have this neurotic compulsion to keep beating on a dead horse: If the people running the UK e-mail and news gateways were to obtain the top-level internet domain .GB and simply encapsulate their internal hostnames such as uk.ac.ucl.cs within that domain when the mail crossed the gateway, they'd not only solve the ordering problem, but also bring their domain in conformance with the ISO country codes. Thus jon@uk.ac.ucl.cs inside the UK would cross the mail gateway and appear everywhere else in the world as jon@uk.ac.ucl.cs.gb and when it was replied to, the mail would simply to those mail gateways MX'd for *.gb, and they'd strip off the .gb, and there you have it. Is this doomed to failure because it's too simple and easy? Brian Kantor UCSD Network Operations brian@ucsd.edu BRIAN@UCSD ucsd!brian
ronald@robobar.co.uk (Ronald S H Khoo) (11/19/90)
In article <23165@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes: [ method for solving uk.* deleted ] > Is this doomed to failure because it's too simple and easy? What about poor old UUCP sites running smail 2.5 ? And what would I put in my netnews headers ? The same as now ? Reversed but with .gb tacked on the end ? How do I reply to other UK sites ? There are *lots* of people under .UK who don't have the capability of understanding reversed names. Lots of people don't have time to diddle with the vendor-supplied sendmail.cf, and there are good reasons not to go with uk-sendmail, like perhaps not liking reversed addresses ? After all, remember, only the campus mail gateway which talks to JANET needs to do domain reversal. I know *academic* sites which operate internally ONLY in the correct domain order, and lets the mail gateway reverse it *just* before it hits JANET. Of course, if the other end did exactly the same, no one would notice that there were any domain reversals going on at all. This is a *much* better solution. Sure, it affects all the users in a very visible way, but at least they'll be warned, instead of just puzzled when their top secret defense-related mail to cs.ucl.ac.uk ends up behind the iron curtain :-) Domain reversed names at a user-visible level is a *bug waiting to happen*. Not all academic sites allow them. The correct fix is for ALL of them to do likewise. No change in JANET software is needed -- that bit only takes reversed addresses. Yes, the user level code for non-UNIX systems will have to change, but that can be gradual. You'll be surprised how easy to teach new users that "that machine over there uses reversed names, and this one here uses sane ones". Old users, well, that's the sort of problem that needs "transition relief" in the form of extra support staff to take the strain of being shouted at :-) If one JANET site can do it (I know at least one does) why can't they all? -- ronald@robobar.co.uk +44 81 991 1142 (O) +44 71 229 7741 (H)
jf@ap.co.umist.ac.uk (John Forrest) (11/19/90)
In article <23165@ucsd.Edu>, brian@ucsd.Edu (Brian Kantor) writes: |> I have this neurotic compulsion to keep beating on a dead horse: |> |> If the people running the UK e-mail and news gateways were to obtain |> the top-level internet domain |> .GB |> and simply encapsulate their internal hostnames such as |> uk.ac.ucl.cs |> within that domain when the mail crossed the gateway, they'd not only |> solve the ordering problem, but also bring their domain in conformance |> with the ISO country codes. |> |> Thus |> jon@uk.ac.ucl.cs |> inside the UK would cross the mail gateway and appear everywhere else in |> the world as |> jon@uk.ac.ucl.cs.gb |> and when it was replied to, the mail would simply to those mail gateways |> MX'd for *.gb, and they'd strip off the .gb, and there you have it. |> |> Is this doomed to failure because it's too simple and easy? |> |> Brian Kantor |> UCSD Network Operations |> brian@ucsd.edu BRIAN@UCSD ucsd!brian I'll actually post up on this, although I was tempted to just ignore the silly comment. As far as everyone here is concerned UK is our top domain, I don't know where the ISO GB comes in - just safe to say that no-one uses it. It is worth noting that Great Britain is not the same as the United Kingdom - although I don't really want to get into a great argument on the politics of this (mainly Irish related), we DO use the term United Kingdon here at lot, just as United States of America is abbreviated to United States and US. Thus UK is a more appropriate term that GB, and since it is already used, we should continue to do so. As for the comment about putting .gb on the end of grey-book, this breaks both domain conventions and doesn't solve the problem! Basically, our mailers have evolved to a situation where they try to be soft on which order the address is given. This has worked up to now, but it is ceasing to work unambiguously. We, in the UK, thus need to solve our little problem. There are many, I for one, who seem to prefer abandoning the reverse notation, but while there are mailers who send out the stuff, we will be stuck with this problem. It isn't a problem of the rest of the world taking to us, so much as problems around the UK. Talking of country codes, why isn't it brian@ucsd.edu.US ? John Forrest Dept of Computation UMIST.