[ih] Converging on MX use

[Greg Skinner]

On Feb 27, 2025, at 1:34 PM, Craig Partridge via Internet-history <internet-history at elists.isoc.org> wrote:
> 
> OK, back story.  And yes, USENIX paper is the documentation.
> 
> RFC 974 was completed in November '85.  It was the product of some public
> and some private email discussions after I discovered a bug in MD/MF RRs
> (basically, Jon said "you found the bug, you fix it -- and I'll ask a few
> folks to help you").  It was NOT a joint product of the various email
> networks.
> 
> Jon delayed RFC 974's publication so that Paul Mockapetris could write up
> RFC 973, which was a collection of updates to the DNS, including defining
> the MX RR type and such.  I suspect Jon also wanted to see if the SRI
> meeting changed anything in either Paul's or my draft RFCs.
> 
> The meeting in January was hosted by Jake Feinler and revolved around two
> issues: (1) what should the top level domain names be?; and (2) what level
> of interoperability could we get between the major email networks
> (Internet, BITNET, CSNET, UUCP).  On email: UUCP was eager to reduce the
> dependence on relative routes using ! -- and they had a routing tool
> (written by Bellovin or Honeyman? can't recall who) that would enable
> domain name usage.  CSNET was already committed to domain names.  BITNET
> was interested but unsure how.  But we agreed to all make it work.  And RFC
> 976 by Horton, a month later (Feb 1986) starts to sketch how it would work
> for UUCP.

I believe ‘pathalias’ is the tool you’re thinking of.

PATHALIAS or The Care and Feeding of Relative Addresses
Honeyman, Peter; Bellovin, Steven M.
1986-06

https://deepblue.lib.umich.edu/bitstream/handle/2027.42/107874/citi-tr-86-2.pdf?sequence=1&isAllowed=y

--gregbo