funk@osiris.cso.uiuc.edu (10/07/89)
In our implementation of email (over DDN), we often see sites that use addresses in the form user%subsite@host.subdomain.domain. The use of the % notation has been variously described as being part of RFC 822, RFC 733, or JNT. I've read 822/733, and while they will accept that as part of the user portion of an address, I don't see that they define anything about that. Have not had a chance to get documentation on JNT, but suspect it will show the same thing. Where does the % notation come from? Please email. ------------------------------------------------------------------------------- | Bruce Funk INTERNET: funk@osiris.cso.uiuc.edu | |C/ADP Branch MILNET: bruce@acseh.kaiserslau-emh1.army.mil | |ACSEH, 21st TAACOM __________________________________________________| |Kaiserslautern, W. Germany | Any resemblance between me and reality | |(guesting on osiris) | is strictly coincidental | -------------------------------------------------------------------------------
lear@NET.BIO.NET (Eliot Lear) (10/08/89)
When you see user%subsite@host.subdomain.domain, you should treat everything including before the @ as user data, and things will work. Use of % in addresses (be they recipients or senders) is discouraged in the Internet. For more information, see RFC 1123 sections 5.2.6 and 5.2.16. -- Eliot Lear [lear@net.bio.net]
pcg@aber-cs.UUCP (Piercarlo Grandi) (10/12/89)
In article <10800001@osiris.cso.uiuc.edu> funk@osiris.cso.uiuc.edu writes:
In our implementation of email (over DDN), we often see sites that use
addresses in the form user%subsite@host.subdomain.domain. The use of the
% notation has been variously described as being part of RFC 822, RFC 733,
or JNT. I've read 822/733, and while they will accept that as part of the
user portion of an address, I don't see that they define anything about that.
I remember having read some posting from an Internet guy saying that in
the future the @ form for "routing" is going to be deprecated, and that
the % form will become the official one. Apparently this is because the
Internet powers have realized that % is very common, while @a:u@b is very
rarely used, and so a bit of realpolitik comes about... (also, the %
form is slightly easier to parse by machine).
In particular, end users find it much easier to write "routes" (and please
do not object that domainized routes are never necessary, at least at the
user level -- this is wishful thinking in the presence of sysadms that do
not update tables, networks that are not registered, etc...) with % rather
than @. Let me please repeat the old whine that the UUCP ! form could have
been adopted throughout, and we would all have been spared a lot of misery.
As to the origin of %, I suspect it is either Berkeley or CSnet; surely the
latter did a lot to make it common, especially when ARPA was the Internet
and everybody else had to pass thru the relay at udel.
--
Piercarlo "Peter" Grandi | ARPA: pcg%cs.aber.ac.uk@nsfnet-relay.ac.uk
Dept of CS, UCW Aberystwyth | UUCP: ...!mcvax!ukc!aber-cs!pcg
Penglais, Aberystwyth SY23 3BZ, UK | INET: pcg@cs.aber.ac.uk