Path: utzoo!attcan!uunet!mcsun!ukc!axion!masalla.fulcrum.bt.co.uk!igb From: I.G.Batten@fulcrum.bt.co.uk (Ian G Batten) Newsgroups: comp.mail.uucp Subject: Re: British DNS (was Re: Paths and Precedence...) Message-ID: Date: 12 Jul 90 10:33:26 GMT References: <++J4QGC@xds13.ferranti.com> <1141@vision.UUCP> Sender: root@fulcrum.bt.co.uk (Root on Masalla) Organization: BT Fulcrum, Birmingham Lines: 25 Disclaimer: Organisation given for identification purposes only chris@vision.UUCP (Chris Davies) writes: > ... the NRS (Name Registration Scheme) in the UK > decided that names need not be fully qualified, ie you can address email to > any of (eg) site.CO.UK, site.CO, site (!!) and the local s/w is supposed to > know what to append for you :-( Thus we would not be able to address email > to a top-level domain .AC or .CO... No, Grey Book says that the ``JNT Header'' --- the transport recipient that forms the first block of the Blue Book transfer --- is fully qualified. It'll normally work if it isn't, be if it doesn't there's little recourse. The wildcarding you suggest I always rationalised by the fact that you should add parts of your OWN address to the recipient. I should first try to add fulcrum.bt.co.uk (so I can mail user@balti, user@masalla), then bt.co.uk (so I can mail user@axion, user@planet). I don't try to add co.uk or uk, becaise I don't hold tables that complete. This strikes me as reasonable. Since the completion is only affixed if I have EVIDENCE that the resulting compound exists, if I typed ``user@foo.co'' it would be forwarded as is to my relay. I only do domain expansion if I know the compound is valid. ian