IP: Another view on ArpaNet history --

[Dave Farber <farber () cis upenn edu>]

To: Dave Farber <farber () cis upenn edu>
Cc: Danny Cohen <Cohen () myri com>
Subject: Re[1]: IP: from cyhst digest via rand Alumni mailing -- some arpanet history
Date: Sun, 11 Apr 99 16:27:56 -0700
From: cohen () myri com

Dave,

On Friday, March 26 1999 you sent to IP (ip-sub-1 () admin listbox com)
a message [attached] from RAND's Willis Ware with the subject "Arpanet
history".  It deals with (1) the history of Distributed Communications,
and (2) DDN.

I'd like to set the record straighter, responding to the former part
of that message.  I am sure that its latter part is most accurate.

It seems that there are three schools of thought about the origin of the
ARPAnet in particular, and of Distributed Communications in general: the
East Coast school, the West Coast school, and the UK school.

According to the East Coasters, distributed computing ("networking")
idea was first proposed by MIT's late JCR Licklider, and the distributed
communication technology (packet switching, distributed routing, etc.)
was invented by MIT's Lenny Kleinrock.  MIT's Larry Roberts, who worked
closely with Kleinrock at MIT Lincoln lab (as classmates and as
officemates) went to ARPA then to implement Lick's idea using
Kleinrock's technology.

According to the West Coasters, it was all invented by RAND's Paul Baran
who published in 1964 a series of 12 RAND Research Memoranda about how
the USAF's command-control communications would survive a major nuclear
attack using distributed communication.

According to the Brits, it all started at NPL (UK's National Physical
Laboratory) by Donald Davies and Roger Scantlebury of NPL.

The facts, as shown below, support the first school.

By the way, I worked with/for Larry Roberts before and when he moved
from Lincoln Lab to ARPA, in 1966.  I connected his DEC-338 graphic
computer at the ARPA office in the Pentagon (3D-169) with Lincoln's
TX-2, over phone lines.  It sure was the first ARPAnetworking.  :-)

The way I remember the ARPAnet history is as described in "A Brief
History of the Internet" by Barry Leiner et al., Communications of
the ACM, Vol. 40, No. 2, pp. 102-108, February 1997.
        (http://www.unizar.es/isocara/historia/History.html)

The following is quoted from there.

+--------------------------------------------------------------------------+
|
|                       Origins of the Internet                            
|                                                                          
| The first recorded description of the social interactions that could     
| be enabled through networking was a series of memos written by J.C.R.    
| Licklider of MIT in August 1962 discussing his "Galactic Network"        
| concept. He envisioned a globally interconnected set of computers        
| through which everyone could quickly access data and programs from any   
| site. In spirit, the concept was very much like the Internet of today.   
| Licklider was the first head of the computer research program at DARPA,  
| starting in October 1962. While at DARPA he convinced his successors at  
| DARPA, Ivan Sutherland, Bob Taylor, and MIT researcher Lawrence G.       
| Roberts, of the importance of this networking concept.                   
|                                                                          
| Leonard Kleinrock at MIT published the first paper on packet switching   
| theory in July 1961. Kleinrock convinced Roberts of the theoretical      
| feasibility of communications using packets rather than circuits, which  
| was a major step along the path towards computer networking. The other   
| key step was to make the computers talk together. To explore this, in    
| 1965 working with Thomas Merrill, Roberts connected the TX-2 computer in 
| Mass. to the Q-32 in California with a low speed dial-up telephone line  
| creating the first (however small) wide-area computer network ever       
| built. The result of this experiment was the realization that the        
| time-shared computers could work well together, running programs and     
| retrieving data as necessary on the remote machine, but that the circuit 
| switched telephone system was totally inadequate for the job.            
| Kleinrock's conviction of the need for packet switching was confirmed.   
|                                                                          
| In late 1966 Roberts went to DARPA to develop the computer network       
| concept and quickly put together his plan for the "ARPANET", publishing  
| it in 1967. At the conference where he presented the paper, there was    
| also a paper on a packet network concept from the UK by Donald Davies    
| and Roger Scantlebury of NPL. Scantlebury told Roberts about the NPL     
| work as well as that of Paul Baran and others at RAND. The RAND group    
| had written a paper on packet switching networks for secure voice in the 
| military in 1964. It happened that the work at MIT (1961-1967), at RAND  
| (1962-1965), and at NPL (1964-1967) had all proceeded in parallel        
| without any of the researchers knowing about the other work. The word    
| "packet" was adopted from the work at NPL and the proposed line speed to 
| be used in the ARPANET design was upgraded from 2.4 kbps to 50 kbps.     
|                                                                          
| In August 1968, after Roberts and the DARPA funded community had refined 
| the overall structure and specifications for the ARPANET, an RFQ was     
| released by DARPA for the development of one of the key components, the  
| packet switches called Interface Message Processors (IMP's). The RFQ was 
| won in December 1968 by a group headed by Frank Heart at Bolt Beranek    
| and Newman (BBN). As the BBN team worked on the IMP's with Bob Kahn      
| playing a major role in the overall ARPANET architectural design, the    
| network topology and economics were designed and optimized by Roberts    
| working with Howard Frank and his team at Network Analysis Corporation,  
| and the network measurement system was prepared by Kleinrock's team at   
| UCLA.                                                                    
|                                                                          
| Due to Kleinrock's early development of packet switching theory and his  
| focus on analysis, design and measurement, his Network Measurement       
| Center at UCLA was selected to be the first node on the ARPANET...       
|                                                                          
+--------------------------------------------------------------------------+

The best way to resolve the extent of Kleinrock's vs. Baran's influence
on Roberts' thinking is to read what Larry Roberts himself says about it
in (http://www.ziplink.net/~lroberts/InternetChronology.html), his own
"Internet Chronology".  The following is quoted from there.

..............................................................................
:
: * Jul-61                                                                   
:      First Paper on Packet Switching Theory, Leonard Kleinrock,            
:      "Information Flow in Large Communication Nets.", RLE Quarterly        
:      Progress Report. This was the theoretical work that convinced Roberts 
:      that packets could be used for the Internet.                          
:                                                                            
: * Aug-62                                                                   
:      First Paper on Internet Concept by J.C.R. Licklider & Welden Clark,   
:      "On-Line Man Computer Communication".                                 
:                                                                            
: * Oct-62                                                                   
:      ARPA Computer Program Begins, J.C.R. Licklider becomes first          
:      ARPA IPTO Director. Writes internal papers on Galactic Network.       
:      Lick leaves in 64. It was Licklider's concept, which spurred          
:      Roberts to build the Internet.                                        
:............................................................................

The bottom line is that Larry Roberts was influenced by Licklider's and
Kleinrock's work at MIT, not by Baran's work at Rand, and that Davies'
work at NPL was not only done much later than Kleinrock's, but by the
time Larry heard about it, he had already formed his ideas about packet
switching based on Kleinrock's work.

In summary: Licklider deserves the credit for proposing computer
networking, and Kleinrock deserves the recognition for being the first
to establish and publish the underlying principles of distributed
communications (aka Packet Switching), and to conceive and publish the
concept of distributed routing.  In addition, he also influenced Larry
Roberts in the development of the Arpanet.

None of this is meant to challenge the assertion that Baran developed
his ideas independently of Kleinrock.  It is just that Kleinrock was 
there first.

Quoting Willis Ware's message: "It is not clear how this [Baran's] work
got coupled into DOD/ARPA interests."

On the other hand, it is very clear (according to Larry Roberts) how
Licklider's and Kleinrock's work got coupled into the ARPAnet.

                                                                Danny

Willis Ware's message
..............................................................................
: Date:    Thu, 25 Mar 1999 14:33:36 -0500                                   
: From:    Willis H. Ware  <willis () RAND ORG>                                 
: Subject: Arpanet history                                                   
:                                                                            
:      --------------------------------------------------------------        
:      Community Memory: Discussion List on the History of Cyberspace        
:      --------------------------------------------------------------        
:                                                                            
: I want to comment on ARPANET, MILNET, et al history.  I had direct         
: experience with all of those events because of my position at RAND, my     
: involvement with DoD advisory committees, and the ongoing personal         
: interaction with the individuals in question. The following discussion     
: comes from my direct knowledge, plus conversations with the parties        
: concerned.                                                                 
:                                                                            
: 1.  Distributed Communications -- it was the name that Paul Baran coined   
: and used before Donald Davies (of the UK) later introduced the term        
: "packet switching".  At the time, Paul was in the Computer Science         
: Department of RAND; the department contained both a computer-science R&D   
: program and the corporate computer-services organization.  He worked       
: directly for Keith Uncapher [well known for his founding of the USC/ISI,   
: his long term support of (D)ARPA, contributions to the professional        
: societies of our field] and in turn, I was the department head.  The       
: genesis of Paul's work was a concern expressed by the USAF to RAND's       
: president Frank Collbohm that command-control communications would not     
: survive a major nuclear attack.  Frank's thought was to use the AM radio   
: stations as back-up, largely because of their wide spread presence         
: throughout the country.                                                    
:                                                                            
: Frank passed the project to Paul and eventually the Distributed            
: Communications concept arose.  The work is documented in a series of 12    
: Research Memoranda which are online at:                                    
:                                                                            
:       http://www.rand.org/publications/classics                            
:                                                                            
: They discuss routing strategy (one of which was called the "hot potato     
: algorithm" at the time), implementation, network survival under attack,    
: survival as a function of network topology and redundancy, costs... in     
: short a complete and thorough analysis of the construct.                   
:                                                                            
: It is not clear how this work got coupled into DOD/ARPA interests.  The    
: reports were widely distributed and briefed; Uncapher was already at       
: that time in contact with ARPA.  Paul had many conversations with the      
: industry, notably AT&T and Bell Labs, and some interaction with ARPA.      
: There was also networking interests at MIT and Lincoln Labs although the   
: motivation there was not driven by military interests but by computer      
: system networking.  Roberts, Lickleider, Kleinrock, Kahn and others were   
: colleagues; several of them later moved to ARPA.                           
:                                                                            
: In any event, eventually Larry Roberts, Lick Lickleider, Bob Kahn and      
: others at ARPA set out to build (what by then was called) a packet         
: network to inter-connect computer systems.  The clearly expressed belief   
: at the time was that such an interconnect would lead to sharing of         
: computer capabilities across the network, and in particular, make          
: available specialized resources to users nationwide. Some of the           
: motivation was expressed in terms of making super-computers available to   
: the scientific R&D community without having to own one directly.  The      
: history of the early days of ARPANET is well documented; no need to go     
: into it here.                                                              
:                                                                            
: 2. DDN.  The shortfalls of the DOD major message-communication network,    
: AUTODIN, had become well known and there were various proposals for        
: modernizing it.  One was to simply replace the aging computers; another    
: was to replace the entire system with modern technology, notably packet    
: switching.  The final proposal and project for AUTODIN-2 was (as I         
: recall) a mixture of packet-switching and circuit switching; the           
: contractor, I believe, was Western Union.  ARPANET was clearly an          
: operational entity by that time; and hence, the technical contest was      
: (in essence) a fully packetized network vs. a partially packetized one.    
:                                                                            
: In any event, a review of the AUTODIN-2 proposal vs. piggybacking on       
: ARPANET technology was organized by Steve Walker (the founder later of     
: Trusted Information Systems) who was in OSD at the time.  I was a member   
: of his Defense Science Board committee, and one of the issues before us    
: was the contest between X.25 and TCP/IP protocols.  The DOD had, via the   
: ARPANET, adopted TCP/IP; the commercial world was signaling that it        
: intended to adopt X.25.  The National Research Council did a study of      
: the matter and recommended that DOD systems support both protocols.        
:                                                                            
: The committee position and report was to cancel the AUTODIN-2 project,     
: instead to sequester that part of the ARPANET that even then had           
: military sites on it, and eventually to use it as the foundation for the   
: DDN, the Defense Data Network.                                             
:                                                                            
: I recall the name Heidi Heiden (who was, contrary to the name, a male      
: Army colonel) as the action officer on the DOD side. I could possibly      
: find a copy of the report in my historical holdings.                       
:                                                                            
: Steve Walker, now active as a venture capitalist, would know more          
: precisely the details of the AUTODIN-2 caper, the politics of its          
: cancellation, etc.  Perhaps he can be solicited to add his views.          
:                                                                            
: Willis H. Ware                                                             
: RAND Santa Monica, CA                                                      
:............................................................................

uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu