The technical development of internet email

IEEE Annals of the History of Computing

Volumn 30, Issue 2, 2008, Pages 3-29

  Partridge, Craig ^a  

^a BBN TECHNOLOGIES   (United States)

Author keywords

Electronic mail; Internet; Multimedia communication; Protocols

Indexed keywords

EVOLUTION (CO); TECHNICAL DEVELOPMENTS;

INTERNET; INTERNET PROTOCOLS; LAWS AND LEGISLATION; STANDARDIZATION; STANDARDS;

ELECTRONIC MAIL;

EID: 46649088930     PISSN: 10586180     EISSN: None     Source Type: Journal    
DOI: 10.1109/MAHC.2008.32     Document Type: Article

References (150)

1. J.S. Quarterman, The Matrix: Computer Networks and Conferencing Systems Worldwide, Digital Press. 1990;
2. P.H. Salus, Casting the Net: From ARPANET to Internet and Beyond, Addison-Wesley, 1995;
3. K. Hafner and M. Lyon, Where Wizards Stay up Late, Simon & Schuster, 1996;
4. I.R. Hardy, "The Evolution of ARPANET Email," Univ. California, Berkeley, master's thesis, 1996;
5. J. Abbate, Inventing the Internet, MIT Press, 1999.
6. Quarterman's book sought to document the state of networking in the world in 1990 and is a tremendously valuable testament to a time just before the Internet took over. The works of Salus, Hafner and Lyon, and Abbate are general histories in which email plays a modest part - this article is, in some sense, the fine-grained version of email's history. Hardy's thesis seeks to understand the social dynamics of the community in which email developed and is a useful complement to this work. T. Haigh, "The Web's Missing Links: Search Engines and Portals," The Internet and American Business, W. Aspray and P. Ceruzzi, eds., MIT Press, 2008, also has a useful perspective.
7. J.K. Reynolds, Post Office Protocol, Internet Request for Comments No. 918, Oct. 1984;
8. J. Myers and M. Rose, Post Office Protocol - Version 3, Internet Request for Comments No. 1939, May 1996.
9. (For information on retrieving RFCs online, see Ref. 5).
10. M.R. Crispin, Interactive Mail Access Protocol: Version 2, Internet Request for Comments No. 1064, July 1988.
11. T. Van Vleck, "The History of Electronic Mail," http:// www.multicians.org/thvv/mail-history.html.
12. The Internet Request for Comments series is maintained online. To retrieve a particular RFC, use http://www.rfc-editor.org/rfc/rfc#.txt, where # is replaced with the one-, two-, three-, or four-digit RFC number.
13. R.W. Watson, A Mail Box Protocol, Internet Request for Comments No. 196, 20 July 1971.
14. R. Tomlinson, personal communication, 13 Apr. 2006.
15. A similar line of thinking had led to email in Multics. Louis Pouzin wanted a way to send a message to an operator and that idea morphed into sending messages between users. J. Klensin, personal communication, 10 June 2006.
16. D.G. Bobrow et al., "TENEX, a Paged Timesharing System for the PDP-10," Comm. ACM, vol. 15, no. 3, 1972.
17. SNDMSG'S origins are uncertain. Tomlinson ported SNDMSG to TENEX from another operating system. He believed the operating system was Berkeley's SDS-940 system, but the SDS-940 veterans report it did not have an email program.
18. The choice of did cause some controversy. It turned out that was a reserved character in Multics that caused all input to that point on a line to be deleted.
19. A. Bhushan, File Transfer Protocol, Internet Request for Comments No. 114, 10 Apr. 1971;
20. A. Bhushan et al., The File Transfer Protocol, Internet Request for Comments No. 265, 17 Nov. 1971.
21. A side note: The problem of developing a general distributed file system (which was the goal of the initial FTP work) turned out to be excruciatingly hard and was not solved until the early 1980s and required the development of the concept of remote procedure call (see B.J. Nelson, "Remote Procedure Call," doctoral dissertation, Carnegie Mellon Univ., 1981)
22. and distributed transactions (W.E. Weihl, "Transaction Processing Techniques," Distributed Systems, 2nd ed., S. Mullender, ed., Addison-Wesley, 1993).
23. A.K. Bhushan, Data and File Transfer Workshop Notes, Internet Request for Comments No. 327, 27 Apr. 1972.
24. A.K. Bhushan, File Transfer Protocol, Internet Request for Comments No. 354, 8 July 1972.
25. A.K. Bhushan, Comments on the File Transfer Protocol, Internet Request for Comments No. 385, 18 Aug. 1972. Despite the title, it is the document that defined MLFL and MAIL and appears to have been the standard reference for the next eight years.
26. D. Crocker, personal communication, 26 Apr. 2006.
27. The Arpanet community eventually developed a term for this kind of problem: The n X m rule. The n X m rule, paraphrased, says that one should abhor designing systems where the consequence of adding a new system is that every other system needs to learn how the new system works (i.e., where the new system places users' mailboxes).
28. A.K. Bhushan, RFC-385;
29. M. Padlipsky, personal communication, 14 Apr. 2006.
30. M. Padlipsky, "And They Argued All Night...," note at http:// www.lafn.org/%7Eba213/allnight.html.
31. S. Lukasik, oral history interview by J.E. O'Neill, 17 Oct. 1991, Redondo Beach, Calif., OH 232, Charles Babbage Inst. Lukasik guessed he was using email on Arpanet by 1971 (p. 11). That is too early, but 1972 is plausible.
32. The original name was Tape Editor and COrrector, but by 1973, the acronym had evolved. Thanks to Dan Murphy (TECO's author) and John Vittal for tracking down when the name changed.
33. Lukasik (Ref. 21) says Roberts produced RD overnight. Roberts dates the invention of RD to July 1972 in his Internet chronology (http:// www.packet.cc/internet.html).
34. M. Yonke, personal communication, 26 Apr. 2006. There was an intermediate program between NRD and BANANARD, called WRD (for Wessler's RD). Yonke recalls it was only around briefly and was largely Wessler's code with bug fixes, but otherwise unmodified.
35. M. Yonke, personal communication, 26 Apr. 2006;
36. S. Crocker, personal communication, 31 May 2006;
37. J. Vittal, personal communication, 5 June 2006;
38. Yonke remembers the index size as 5,000 while Crocker remembers it as a much smaller number (a few hundred). Vittal remembers hitting the limit.
39. B.D. Wessler, personal communication, 17 Apr. 2006.
40. R. Tomlinson, personal communication, 13 Apr. 2006;
41. J. Vittal, personal communication, 5 June 2006.
42. The meeting announcement (A. McKenzie, File Transfer Protocol-Meeting Announcement and a New Proposed Document, Internet Request for Comments No. 454, 16 Feb. 1973) contains a draft new specification and includes suggested improvements to MAIL and MLFL.
43. J. Day, personal communications, 14 and 16 Apr. 2006.
44. N. Neigus, File Transfer Protocol, Internet Request for Comments No. 542, 12 Aug. 1973.
45. The common set of attendees was Abhay Bhushan (from MIT), Bob Braden (UCLA), Alex McKenzie (BBN), Jon Postel (UCLA), and Jim White (SRI).
46. That there were FTP meetings at roughly the same time of year in both 1972 and 1973 has caused some confusion decades later. People are sometimes confused about which meeting they attended.
47. J. Postel, "Mail Protocol," DDN NIC memo 29588, 18 Feb. 1976 in ARPANET Protocol Handbook, DDN Network Information Center, Jan. 1978.
48. Allowing multiple addresses in a MLFL/MAIL command was proposed in A. Bhushan, File Transfer Protocol (FTP) Status and Further Comments, Internet Request for Comments No. 414, 29 Nov. 1972, and again (now using new commands) in K. Harrenstein, FTP Extension: XRSQ/XRCP, Internet Request for Comments No. 743, 30 Dec. 1977. NIC 29588 suggests that some sites used the RFC-414 scheme but that it was not universally accepted.
49. There were some proposals for an email protocol (the preface to RFC-724 mentions "Several versions of such a protocol have been proposed ..."). However, none seem to have gotten serious attention; only one seems to have been issued as an RFC (J.E. White, Proposed Mail Protocol, Internet Request for Comments No. 524, June 1973), and there's a strong impression that the community paid very little attention to the issue.
50. J. Vittal, "MSG - A Simple Message System," Computer Messaging Systems, R.P. Uhlig, ed., North Holland, 1981.
51. BBN collected Arpanet traffic measurements monthly during this time, but I have been unable to find a copy of them and thus could not verify this claim.
52. S. Walker, "Message Group Status," email to MsgGroup of 7 June 1975.
53. The Navy was a pioneer in the use of email for operational needs. In 1973, the Navy had inaugurated the "Navy Communications Processing and Routing System (NAVCOMPARS)," an internal email system to distribute orders to shore bases and to ships (messages to ships were relayed via shortwave radio using human operators). NAVCOMPARS remained a key part of the Navy's infrastructure until it was turned off in 2002. B.M. Hintz, "The Naval Communications Processing and Routing System: Analysis of an Automated System," master's thesis, Naval Postgraduate School, Mar. 1976. A good description of the system as it was operating in 1984 can be found in S. Blumenthal et al., "NAVCAMS LAN Engineering Plan," BBN Report 5907, Mar. 1985.
54. S. Walker, personal communication, 8 June 2006.
55. D.P. Deutsch and D.W. Dodds, Hermes System Overview, BBN Report 4115, May 1979.
56. D.H. Crocker, Framework and Functions of the "MS" Personal Message System, tech. report R-2134-ARPA, RAND Corp., Dec. 1977.
57. A copy of the original memo can be found at http:// rand-mh.sourceforge.net/book/overall/hiofmh.html.
58. D.M. Ritchie and K. Tompson, "The UNIX Time-Sharing System," The Bell System Technical J, vol. 57, no. 6, July-Aug. 1978, part 2, pp. 1905-1930.
59. Dave Crocker observes the interesting counterpoint that attempts to "enhance" MH, such as xmh and mhe, have sought to move the MH commands back into a monolithic program.
60. Dave Crocker reports that the MS manual, to his surprise, does not describe features for searching. D. Crocker, personal communication, June 2006.
61. I interacted with Rose and Jacobson on MH support in the 1980s. Other names come from J. Peek, MH & xmh: Email for Users & Programmers, O'Reilly and Associates, 3rd ed., 1995.
62. A. Bhushan et al.,Standardizing Network Mail Headers, Internet Request for Comments No. 561, 5 Sept. 1973.
63. T.H. Myer and D.A. Henderson, Message Transmission Protocol, Internet Request for Comments No. 680, Apr. 1975.
64. See E. Stefferud, "MSGGROUP Situation Report #1," email to Header-People of 2 Dec. 1975.
65. In particular, Mooers wrote emails on behalf of the BBN team explaining details of RFC-680. Years later, Mooers was CSnet's "postmistress" and internationally known for her expertise in solving email problems.
66. J. Haverty, "Re: [ih] NIC 7104 (ARPANET Protocol Handbook)" email to Internet-History mailing list of 28 Apr. 2006.
67. See RFC-724 preface.
68. Interview with A. Vezza on 3 May 2006. He believes his (now lost) memo, "Message Services Committee Minority Report," Jan. 1975, expressed the view that headers should be machine readable. See also the preface to RFC-724, which reflects the continued debate.
69. The initial membership of the new committee was Walker, John Seely-Brown, David Farber, Ken Pogran, and John Vittal. J. Vittal, personal communication, 5 June 2006.
70. For key notes in the discussion, see the note from J. Haverty (JFH MIT) on 30 Sept. 1976, "Re: Your message to MSGGROUP at from and sender ... " and J. Vittal, "Some comments (RFC-724, etc.) ..." of 9 Nov. 1976, both emails to MsgGroup. Confusing the discussion is that an early draft of RFC-724 was apparently distributed, with its assigned RFC number, in 1976 (well before its official publication date). K. Pogran et al., Proposed Official Standard for the Format of ARPA Network Messages, Internet Request for Comments No. 724, 12 May 1977.
71. J. Vittal, "Comments on the state of the world," email to Header-People mailing list of 29 Oct. 1977.
72. See the numerous emails between 4 and 11 Oct. 1977 in Header-People.
73. Mailing lists had an odd history in the message format standardization process. No later than 1975, there were mailing lists as we know them today, in which email to, say, MsgGroupΙSI was delivered to host ISI. Host ISI then redistributed the message to the members of the list. Yet there seem to have been a number of different practices intended to show that the message was to a mailing list. So, at one time, ISI would rewrite the outbound TO field of MsgGroup messages to read "[ISI]MSGGROUP:".
74. D. Crocker, Standard for the Format of ARPA Internet Text Messages Internet Request for Comments No. 822, Aug. 1982;
75. P. Resnick, Internet Message Format, Internet Request for Comments No. 2822, Apr. 2001.
76. CSnet supported RFC-733 format from the start. The UUCP network took somewhat longer. The driving forces were sendmail (used on many UCP systems) and netnews B, which intentionally used 733 format for bulletin boards. Both software systems (plus a desire to easily gateway to the Internet) pushed the community to informally standardize on 733 for email. (S. Bellovin, personal communication, 30 May 2006;
77. M. Horton, personal communication, 6 June 2006;
78. see also, M. Horton, UUCP Mail Interchange Format Standard, Internet Request for Comments No. 976, Feb. 1986.) Bitnet started using a custom VM email format but soon shifted to 733 (J. Klensin, personal communication, 26 May 2006).
79. A. Bhushan, File Transfer Protocol (FTP) Status and Further Comments Internet Request for Comments No. 414, 29 Nov. 1972, lists the implementation status of various FTP implementations and observes that Clements has implemented email retransmission.
80. See Deutsch and Dobbs, Hermes System Overview, p. 21.
81. Compare with B. Reid, "Let's hear it for uniform standards," email to Header-People of 10 Feb. 1978.
82. This decision to interconnect is, in retrospect, somewhat surprising. The different networks did view themselves as competing with each other. However, they also viewed themselves as competing with the postal services (derisively dubbed "snail mail") and prided themselves on getting email where it belonged faster and more effectively than paper-based mail. Credit should also be given to Larry Landweber, a member of both CSnet's and Bitnet's boards, and a vigorous advocate of interconnecting networks.
83. D.A. Nowitz and M.E. Lesk, "A Dial-Up Network of UN IXTM Systems," 18 Apr. 1978, Unix Programmer's Manual, 7th ed., Bell Telephone Laboratories, 1979.
84. J.S. Quarterman, The Matrix, p. 251 and 278.
85. P. Honeyman and S. Bellovin, "PATHALIAS or the Care and Feeding of Relative Addresses," Proc. 1987 Summer Usenix Conf., 1986, Usenix Assoc., pp. 126-141.
86. D. Comer, "The Computer Science Research Network CSNET: A History and Status Report," Comm. ACM, vol. 26, no. 10, 1983, pp. 747-753.
87. Peter Denning, one of the CSnet principals, remembers that, to make the CSnet proposal "researchy" enough to be acceptable to the National Science Board, the proposal emphasized "resource sharing" rather than email, but everyone, including the junior NSF staffers, understood this was a fig leaf for email. (P. Denning, personal communication, 5 June 2006.)
88. D.E. Comer and J.T. Korb, "CSNET Protocol Software: The IP-to-X.25 Interface," Proc. ACM SIGCOMM '83, ACM Press, 1983, pp. 154-159.
89. L. Lanzillo and C. Partridge, "Implementation of Dial-up IP for UNIX Systems," Proc. 1989 Winter Usenix Conf., Usenix Assoc., pp. 201-208.
90. Surviving source code (version 2.7 from 1981) is about 6,800 lines of C code. There seem to have been no technical papers describing delivermail. The discussion here comes primarily from reading the source code and its Unix manual pages.
91. D.H. Crocker, E.S. Szurkowski,, and D.J. Farber, "An Internetwork Memo Distribution Capability-MMDF," Proc. 6th ACM/IEEE Data Comm. Symp. ACM Press, 1979, pp. 18-25.
92. This list is a subset of the list of differences in E. Allman, "Mail Systems and Addressing in 4.2bsd," Proc. 1983 Winter Usenix Conf. Usenix Assoc., 1983, pp. 53-62.
93. E. Allman and M. Amos, "Sendmail Revisited," Proc. 1985 Summer Usenix Conf., Usenix Assoc., 1985, pp. 547-555.
94. See J. Postel, "Internet Meeting notes - 4, 5, & 6 February 1980," Internet Engineering Notes, no. 134, 29 Feb. 1980, which notes that ISI was working on "Internet Mail, which includes the development of mechanisms for delivery of mail in an internet and provision for multi-media data in the mail."
95. J. Postel, "Internet Meeting Notes - 14 & 15 May 1980," Internet Engineering Notes, no. 145, 25 May 1980.
96. V. Cerf and J. Postel, Mail Transition Plan, Internet Request for Comments No. 771, Sept. 1980.
97. S. Sluizer and J. Postel, Mail Transfer Protocol, Internet Request for Comments No. 772, Sept. 1980. Some people have conflated MTP and MP (the Mail Protocol - see the Early Multimedia section) and incorrectly believe that MTP supported multimedia.
98. RFC-771 does not explain why it was written. But RFC-772 makes clear that MTP is intended solely for use for gateways.
99. Suzanne Sluizer recalls Jon "Postel saying people thought that MTP was too complicated." C.J. Bennett, "A Simple NIFTP-Based Mail System," Internet Engineering Notes, no. 169, 23 Jan. 1981, lists issues with MTP on pp. 4 and 5.
100. J. Postel, "Internet Meeting Notes - 28-29-30 January 1981," Internet Engineering Notes, no. 175, 13 Mar. 1981, lists four implementations: MIT, ISI, DCEC, and COMSAT.
101. S. Sluizer and J. Postel, Mail Transfer Protocol: ISI TOPS-20 Implementation, Internet Request for Comments No. 784, 1 July 1981; S. Sluizer and J. Postel, Mail Transfer Protocol: ISI TOPS-20 File Definitions, Internet Request for Comments No. 785, 1 July 1981.
102. E. Stefferud, "Subdivision of Messages," email to MsgGroup of 11 July 1975.
103. The first use of envelope to mean metadata appears to be by D. Crocker, E. Szurkowski, and D. Farber, "An Internetwork Memo Distribution Capability - MMDF," Proc. 6th ACM/IEEE Data Comm. Symp., ACM Press, 1979, pp. 18-25.
104. The idea for pipelining originated with Phil Karn around 1990, when he told it to me, and I in turn, repeated the idea to Van Jacobson. Jacobson thought it was a wonderful idea, and put it into sendmail only to discover that many SMTP implementations failed if pipelining was turned on. Some of the painful experience is described in P. Kam, email to IETF mailing list of 8 Sept. 1993. Eventually, the IETF approved a pipelining extension to SMTP to make it official: N. Freed, SMTP Service Extension for Command Pipelining, Internet Request for Comments No. 2920, Sept. 2000.
105. C. Partridge, Duplicate Messages and SMTP, Internet Request for Comments No. 1047, 1 Feb. 1988.
106. Unfortunately, SMTP is no longer quite this simple. Some commands now have additional parameters (a consequence of the 8-bit enhancements; J. Klensin, Simple Mail Transfer Protocol, Internet Request for Comments No. 28 21, Apr. 2001), and there's pressure to standardize the error codes to avoid dependence on the error message, which may be a local language (J. Klensin, personal communication, 11 June 2006).
107. E. Feinler et al., DoD Internet Host Table Specification, Internet Request for Comments No. 810, 1 Mar. 1982.
108. Z. Su and J. Postel, Domain Naming Convention for Internet User Applications, Internet Request for Comments No. 819, 1 Aug. 1982.
109. See A.D. Birrell et al., "Grapevine: An Exercise in Distributed Computing," Comm. ACM, vol. 25, no. 4, 1982, pp. 260-274.
110. P. Mockapetris, Domain Names: Concepts and Facilities, Internet Request for Comments No. 882, 1 Nov. 1983;
111. P. Mockapetris, Domain Names: Implementation Specification, Internet Request for Comments No. 883, 1 Nov. 1983.
112. P. Mockapetris and K.J. Dunlap, "Development of the Domain Name System," Proc. ACM SIGCOMM '88, ACM Press, 1988, pp. 123-133.
113. C. Partridge, "MF in domain database," message to NameDroppers mailing list of 29 Oct. 1985.
114. C. Partridge, "MID and MF for one host," message to NameDroppers mailing list of 11 Nov. 1985.
115. The description of the issues is now lost, but it seems useful to reconstruct it from memory. If a name could have both MID and MF records associated with it, we needed a set of rules for delivery in the presence of both records. The obvious answer was that a host that was neither an MID nor an MF for the name could deliver email to either the MID or the MF; a host that was an MF could only deliver to an MD; and a host that was an MD could do a DNS lookup, but, once it realized it was an MD, had to look in other databases to figure out how to deliver the message. Now consider the problem of host H, trying to deliver a message to domain name D. H looks up D in the DNS and gets back a set of email resource records. H must then examine all the records to see if H is listed as either an MD or an MF for D to behave in accordance with the delivery rules. Herein lay a problem. It was possible in the DNS to deliver an incomplete list of MDs and MFs. In particular, the DNS's caching mechanism (combined with the fact that one could query for MDs and MFs either individually or using an aggregate MAILA query) allowed for look up responses to contain either the MDs for D or the MF's for D, or both. And if H did not get both MDs and MFs, H could make an incorrect decision.
116. These discussions involved both private and public messages. The private messages have been lost. The key public messages are P. Milazzo, "Re: MD and MF for one host," message to NameDroppers on 11 Nov. 1985 in response to the message cited in Ref. 95;
117. C. Partridge, "Mailers use MD and MF," message to NameDroppers on 12 Nov. 1985;
118. and P. Mockapetris, "MD, MF and larger issues," message to NameDroppers on 15 Nov. 1985. It is my recollection that the Mockapetris note came after a private email exchange among Postel, Mockapetris, and Partridge. Rudy Nedved (then of CMU) and Jon Crowcroft (then of University College London) also made important contributions, especially in thinking how MX RRs interacted with sites that gatewayed email.
119. P. Mockapetris, Domain Systems Changes and Observations, Internet Request for Comments No. 973, Jan. 1986;
120. C. Partridge, Mail Routing and the Domain System, Internet Request for Comments No. 974, Jan. 1986. While the RFCs were issued in late January, I recall that the work on RFC-974 was largely done by Thanksgiving of 1985 (which is remarkable, given the issues with MD and MF surfaced on 11 Nov.) and the delay until January was to give Mockapetris time to think about how to address other DNS issues so that RFC-973 could be a comprehensive update.
121. C. Partridge, "Mail Routing Using Domain Names: An Informal Tour," Proc. 1986 Summer Usenix Conf, Usenix. Assoc., 1986, pp. 366-376.
122. No notes of this meeting survive. The attendance list is crafted from my recollections and those of Mary Ann Horton. There's uncertainty about whether Steve Kille was there (he's not sure) but it is more likely than not. Kille would have provided an international perspective (he was at University College London) and an X.400 perspective. The account of the meeting is my recollection. Horton's recollections place somewhat more emphasis on finalizing the list of top-level domains.
123. The issue surrounding net was that SRI (operator of the DDN NIC) and BBN (operator of the CSnet CIC) competed for network operations contracts and differed in their strategies. BBN's approach was to build the brand of the entity for which BBN operated the network (so, for instance, BBNers on the CSnet project had CSnet business cards) on the theory that BBN's dedication to building the brand made BBN more attractive to the customer. SRI sought to strongly link the entity to SRI, on the theory the customer would be more reluctant to change operators. So CSnet wanted to see a net top-level domain so that NIC's name would be, say, nic.inter.net. SRI wanted the NIC's name to be nic.sri.com. In the event, .net was created, but the NIC became nic.ddn.mil.
124. For Postel's views on .arpa and .uucp and the like, see his email "re: Naming and routing" to the NameDroppers mailing list on 8 Feb. 1995.
125. R.E. Millstein, "The National Software Works: A Distributed Processing System," Proc. ACM'77 Conf., ACM Press, pp. 44-52.
126. J. Postel, Internet Message Protocol, Internet Request for Comments No. 753, Mar. 1979.
127. The uuencode format was also adopted by several early email tools, notably Microsoft Mail and Lotus cc:MAil, for packaging attachments. M.A. Horton, personal communication, 6 June 2006.
128. J. Postel, Internet Message Protocol, Internet Request for Comments No. 759, Aug. 1980.
129. J. Postel, A Structured Format for Transmission of Multi-Media Documents, Internet Request for Comments No. 767, Aug. 1980.
130. V. Cerf, Comments on NCP/TCP Mail Service Transition Strategy, Internet Request for Comments No. 773, 1 Oct. 1980.
131. For a brief overview of the projects, see J. Postel, Multimedia Mail Meeting Notes, Internet Request for Comments No. 807, 9 Feb. 1982.
132. R.H. Thomas et al., "Diamond: A Multimedia Message System Built on a Distributed Architecture," Computer, Dec. 1985, pp. 65-78.
133. Concurrently a similar project, Project Athena, was under way at MIT.
134. The first paper on the Andrew Messaging System cites none of the prior Internet-based work. See J. Rosenberg, C.F. Everhart, and N.S. Borenstein, "An Overview of the Andrew Message System: A Portable, Distributed System for Multi-media Electronic Communication," Proc. ACM SIGCOMM '87, ACM Press, pp. 99-108.
135. For a general overview of work in data encoding for the period, see C. Partridge and M. Rose, "A Comparison of External Data Formats," Message Handling Systems and Distributed Applications (Proc. IFIP Workshop on Message Handling), E. Stefferud and O. Jacobsen, eds., North Holland, 1989.
136. There were some pioneers. The earliest idea I have seen for an external data format is J. Haverty, MSDTP-Message Services Data Transmission Protocol, Internet Request for Comments No. 713, 6 Apr. 1976. It contains a remarkably thorough understanding of the problem of creating an external data format. Haverty was at BBN when Deutsch and Vittal were doing their work.
137. D. Deutsch, R. Resnick,, and J. Vittal, Specification of a Draft Message Format Standard, BBN Report 4486, Sept. 1980.
138. D. Deutsch et al., Specification for Message Format for Computer Based Message Systems (Revised), BBN Report 4765R, 23 Apr. 1982. Online literature generally gives all credit for ASNA to Jim White. As explained to me (some years ago), White gets credit for ASN.1's language for expressing types, but the actual on-the-wire encoding (the Basic Encoding Rules) is the creation of Deutsch's group. The dates of these two BBN reports are consistent with that story (they define what clearly became the encoding rules) and explain why the encoding rules are completely unlike Courier, which was White's invention and the Xerox external data format of the time.
139. There is considerable debate, even today, about how viable X.400 would have been as the Internet's email system. Several readers of the paper felt the characterization of X.400 both in terms of the quality of its technology and its chances for success is far too generous. Some others felt this was about right.
140. Proc. Nineteenth IETF Meeting, IETF, Dec 1990, pp. 72-76.
141. Proc. Twentieth IETF Meeting, IETF, 11-15 Mar. 1991, M. Davies and G. Vaudreuil, eds., pp. 75-84.
142. There are some hints in the meeting notes that the two groups initially may have been confused about whether they were complementary or competing. Participants' memories vary. But it is hard to believe that there was much competition, given they had the same chairman.
143. The decision not to support uuencode is noted on p. 62 of Proc. Twenty-Second IETF Meeting, IETF, 18-22 Nov. 1991, M. Davies, C. Clark, and D. Legare, eds.
144. Until Nov. 1991, both the SMTP and RFC-822 extensions groups were chaired by Greg Vaudreuil. Vaudreuil was a good group leader, but also new to the field and quite young and thus, unlike Klensin, in no position to dictate a solution to a fractious (and senior) group of techies. Vaudreuil continued as chair of the 822-extensions group and brought it to a successful conclusion. Klensin remembers Phill Gross, IETF chair, "dragged me, I'm tempted to say kicking and screaming" into taking on the SMTP problems (J. Klensin, personal communication, 11 June 2006).
145. M. Davies, ed., Proc. Twenty-Third IETF Meeting, IETF, 15-20 Mar. 1992.
146. Recollections differ slightly about how this change of course came about. (J. Klensin, personal communication, 11 June 2006;
147. D. Crocker, personal communication, 31 May 2006). Rose pushed for the simplification. What is unclear is whether EHLO was Rose's idea, or the reworking of an earlier idea by Klensin that the working group had discarded. I cannot find documentation pointing to one or the other answer.
148. J. Klensin et al., SMTP Service Extension for 8bit-MIME Transport, Internet Request for Comments No. 1426, Feb. 1993.
149. Of the Dec. 1990 group, only Bob Braden of ISI had written an VITA or participated in crafting an email standards document.
150. Borenstein, Freed, Crocker, and Klensin's background is described earlier in the article. Stefferud ran the Msg-People mailing list in the 1970s and coined the term "envelope." Fair was widely respected as an expert at keeping email systems running and, at the time, managed Apple Computer's email systems. Huitema was a pioneer in networking in France.