>Smoot Carl-Mitchell writes:
>
>> >David Lang <[EMAIL PROTECTED]> wrote
>> >
>> >The e-mail system was developed in the days when reliable permanent
>> >connections were the exception rather then the rule. [...]
>>
>> This is not correct. TCP was just as reliable in the early days of the
>> Internet as it is today. Most mail in those days was sent to its final
>> destination in a single client/server session just as it is today.
>> There was not a lot of hop-by-hop mail delivery.
>
>Kim-Minh Kaplan <[EMAIL PROTECTED]> wrote
>Hum... E-Mail existed long before TCP. And in the UUCP days you even
>needed to actually *know* part of the hops your message needed to do
>to arrive to its destination !
UUCP mail was a different story than early Internet mail. Interestingly,
they grew up almost side by side. UUCP was developed by Bell Labs and
appeared in very early version of Unix (V6 I believe). The UUCP mail system
and its hop-by-hop routing was a design decision due to the UUCP network being
an adhoc collection of dialup links.
The mail system used by NCP of the old ARPANET was actually
grafted onto the file transfer protocol. It general delivered mail
directly to the intended recipient. In those days (1970s) email addresses
in the ARPANET were of the form user@host or literally "user at host". So
the mail transport made a connection directly to the receiving host.
RFC821 was an evolution of the old ARPANET mail system.
>
>> The RFC821 specification for SMTP simply did not contemplate SPAM mail. [...
>]
>
>Of course at that time nobody cared about RFC821...
The early Internet community certainly did. 821 is an update to 733 I
believe which came out in the very early 1980s. I forget the exact date.
But there was a great deal of early interest in developing a standard email
system architecture.
Smoot
-
[To unsubscribe, send mail to [EMAIL PROTECTED] with
"unsubscribe firewalls" in the body of the message.]
