[ih] Date of RFC 791 for celebration

Noel Chiappa jnc at mercury.lcs.mit.edu
Tue Mar 28 18:27:22 PST 2006


Michiel Leenaars (Michiel at staff.isoc.nl) is trying to find the exact
birth-date of what we now call IPv4

My sense is that if we could find out the exact date of the Internet Working
Group (INWG) meeting at which they hashed out what looked basically like the
eventual IPv4 (I think it was IP 3.1, if I have the version number right),
that would be "the" birthdate.


This, alas, predates me slightly, and I have the history somewhatl muddled in
my head. Also, even more problematically, many of the relevant IEN's are not
available online, which makes it difficult to research... Anyway, there were
two major predecessor steps shortly before IP firmed up:

- TCP and IP were split
- Variable length addresses were removed, leaving fixed 4-byte addresses

I am not 100% positive which happened first (or if they happened at the same
time), although I have become fairly sure that the split happened first, with
the removal of variable-length addresses later...

And then there's the issue of version numbers: there are 2, 2.5, 3, 3.1 and
4. What I hear is that 2.5 was a "implementation split", in which the unified
single header remained, but the code was split into two. 3 was the first
version that had the headers fully split (and included "protocol numbers" to
identify which transport protocol was being used - see list below), but my
guess is that it included variable-length addresses.  I seem to recall that
3.1 had the variable-length addresses removed, and 4 was an editorial cleanup
of 3.1

So:

- Is that the right order for the split, and variable-length address
	removal?
- Can we identify the version numbers (2.5, 3, 3.1, etc) which go with
	each version? 
- Can anyone identify the INWG meeting at which the latter happened?


----


Additional data:

To help refresh memories, here's a list, from the minutes of the 15 August,
'77 Internet Meeting Notes, of planned future INWG meetings. Ones which
eventually seemed to happen (as evidenced by minutes in the IEN series) are
marked with a '*', and the relevant IEN number (although I have no idea if
they happened at the place listed):

       15 Aug 77 - Internet meeting at ISI
    13-14 Oct 77 - TCP      meeting at SRI* [66]
        3 Nov 77 - Internet meeting at BBN
    30-31 Jan 78 - TCP      meeting at ISI* [67]
        3 Feb 78 - Internet meeting at UCLA* [22 - 1 Feb]
    20-21 Apr 78 - TCP      meeting at BBN
     1- 2 May 78 - Internet meeting at UCL* [33]
    13-14 Jul 78 - TCP      meeting at PARC
     2- 3 Aug 78 - Internet meeting at LL* [53]
    12-13 Oct 78 - TCP      meeting at LCS
     2- 3 Nov 78 - Internet meeting at SRI

and the following meetings for which IEN minutes (numbers in []'s) exist, but
aren't in the above list, also occurred:

    15-16 Jun 78 - TCP [68]
    18-19 Sep 78 - TCP [69]
    30-31 Oct 78 - Internet [63]
    4 Dec 78 - TCP [70]

Alas, none of these minutes are online. To further help jog memories, here are
the listings for all the seemingly relevant early IEN's:

   5	Cerf	Mar-77	    TCP Version 2 Specification
  21	Cerf	Jan-78	    TCP 3 Specification
  26	Cerf	14-Feb-78   A Proposed New Internet Header Format
  27	Cerf	14-Feb-78   A Proposal for TCP Version 3.1 Header Format
  28	Postel	Feb-78	    Draft Internetwork Protocol
  40	Postel	Jun-78      Specification of Internetwork Transmission Control Protocol
					  - Version 4
  41	Postel	Jun-78	    Internetwork Protocol Specification - Version 4

and finally, some other stuff: RFC-750 contains the following numbers for
protocol type (some extraneous ones deleted):

      Decimal   Octal      Format                             References
      -------   -----      ------                             ----------
          0      0         Reserved
          1      1         raw internet                             [42]
          2      2         TCP-3                                    [36]
          5      5         TCP-3.1                                  [45]
          6      6         TCP-4                                    [46]

And there's also a table of IP header version numbers:

      Decimal   Octal      Version                            References
      -------   -----      -------                            ----------
          0      0         March 1977 version                       [35]
          1      1         January 1978 version                     [36]
          2      2         February 1978 version A                  [42]
          3      3         February 1978 version B                  [43]
          4      4         September 1978 version 4                 [44]

   [35]   Cerf, V. "Specification of Internet Transmission Control
          Program -- TCP (version 2)," March 1977.
   [36]   Cerf, V. and J. Postel, "Specification of Internetwork
          Transmission Control Program -- TCP Version 3,"
          USC-Information Sciences Institute, January 1978.

   [42]   Postel, J. "Draft Internetwork Protocol Specification --
          Version 2," USC-Information Sciences Institute, February 1978.
   [43]   Cerf, V. "A Proposed New Internet Header Format," Advanced
          Research Projects Agency, IEN 26, 14 February 1978.
   [44]   Postel, J. "Internetwork Protocol Specification -- Version 4,"
          IEN-54, USC-Information Sciences Institute, September 1978.
   [45]   Cerf, V. "A Proposal for TCP Version 3.1 Header Format,"
          Advanced Research Projects Agency, IEN 26, 14 February 1978.
   [46]   Postel, J. "Specification of Internetwork Transmission Control
          Protocol -- Version 4," IEN-55, USC-Information Sciences
          Institute, September 1978.

Note that both ref 43 and 45 claim to be IEN 26! The second should probably be
IEN 27. ref 42 might be IEN 28.


(I'm CC'ing this to the Internet-History list so that any responses will be
archived for future historical use; apologies to anyone who gets two copies
as a result.)

	Noel



More information about the Internet-history mailing list