[SIGCIS-Members] Was email really already 75% of ARPANET traffic by 1973?

[thomas.haigh at gmail.com]

Great! So that gets us closer to an actual source for the claim, and also moves the year.

Thanks to Dave Walden I've received some off-list input from Dave Crocker, Vint Cerf and Dan Lynch. They think that email would have been a majority of bytes sent over the network by the end of 1973 (though possible not a majority of packets as telnet would produce a lot of largely empty packets). But they don't recall an actual study.

The figure is also in Wikipedia, without a real source, so I'm acutely aware that it's one of those factoids that historians and journalists will find somewhere like Wikipedia or the Hobbes Timeline and then copy, thus creating a "reliable source" that Wikipedia can then cite to support the information. It's like a time paradox.

Tom

-----Original Message-----
From: Win Treese <treese at acm.org> 
Sent: Wednesday, July 29, 2020 3:02 PM
To: thomas.haigh at gmail.com
Cc: members <members at sigcis.org>
Subject: Re: [SIGCIS-Members] Was email really already 75% of ARPANET traffic by 1973?

HI, Tom. Stephen Lukasik’s retrospective "Why the Arpanet Was Built” (IEEE Annals of the History of Computing, July-September 2011, pp. 4-21, vol. 33 https://www.computer.org/csdl/magazine/an/2011/03/man2011030004/13rRUxly9fL) says "A 1974 Mitre study of Arpanet usage showed that about three-quarters of the traffic was email”. 

It seems odd that Licklider and Vezza said a lot about email but didn’t include that fact in their 1978 “Applications of Information Networks” paper (Proceedings of the IEEE, Vol. 66, NO. 11, November 1978).  They wrote (among other statements:

  "By the fall of 1973, the great effectiveness and convenience of 
   such fast, informed messages services... had been discovered by 
   almost everyone who had worked on the development of the ARPANET -- 
   and especially by the then Director of ARPA, S.J. Lukasik, who soon 
   had most of his office directors and program managers communicating 
   with him and with their colleagues and their contractors via the 
   network. Thereafter, both the number of (intercommunicating) 
   electronic mail systems and the number of users of them on the 
   ARPANET increased rapidly."

A Gizmodo article from 2016 (https://paleofuture.gizmodo.com/the-defense-department-got-mad-at-darpa-for-creating-em-1763274070) has:

BEGIN QUOTE
The explosion of email was swift. In 1974, ARPA asked MITRE to study how the network was being used. They 	were shocked to find out that roughly 75 percent of the net packets were for email.

I reached out to Steve Lukasik, former director of ARPA during the late 1960s and early 1970s, who told me about the bureaucratic hurdles that the agency faced once they had cracked email’s technical problems. History books often ignore, or don’t fully appreciate, the bureaucratic hurdles that must be jumped to accomplish major technological feats. Al Gore didn’t invent the internet, for example, but without him the bureaucratic barriers wouldn’t have been overcome to privatize it.

Email’s use of 75 percent of network traffic in 1974 “had enormous bureaucratic implications that were initially worrisome,” Lukasik told me. “DoD auditors slapped our wrist for violating DoD procedures. They said we had constructed a communication system, but that was the responsibility of the Defense Communication Agency.”
END QUOTE

I couldn’t find any actual details on the "MITRE study” beyond that. It’s fragmentary, but perhaps helpful.

Best,

Win

Win Treese
treese at acm.org


> On Jul 29, 2020, at 1:36 AM, thomas.haigh at gmail.com wrote:
> 
> Hello SIGCIS,
>  
> Our of the last unsourced footnotes for the Revised History of Modern 
> Computing holds a note to myself concerning a possibly exaggerated 
> factoid from the “Hobbes’ Internet Timeline.” 
> https://www.zakon.org/robert/internet/timeline/
>  
> According to the timeline entry for 1973: “ARPA study shows email composing 75% of all ARPANET traffic.” Keep in mind that Tomlinson sent the first network mail in 1971 and mail technologies were rather immature for the first few years.
>  
> If that is true it’s certainly a fact worth including in the book to demonstrate the very rapid spread of email on the ARPANET. But “ARPANET study” is not something I can use to confirm the original source.
>  
> I haven’t been able to find anything so specific in Janet Abbate’s book Inventing the Internet though she features email prominently and agree that its rise was both rapid and unexpected.  Ian Hardy’s undergraduate thesis, an early historical look at Internet email, does not include this particular figure. https://www.livinginternet.com/References/Ian%20Hardy%20Email%20Thesis.txt Craig Partridge’s IEEE Annals article “Technical Development of Internet Email” didn’t, on a recent skim, seem to say anything on this topic either.
>  
> Does anyone know where this number might be coming from? Or have a well-sourced alternative for slightly later year like 1975 or 76?
>  
> Best wishes,
>  
> Tom
>  
>  
>  
>  
> _______________________________________________
> This email is relayed from members at sigcis.org, the email discussion 
> list of SHOT SIGCIS. Opinions expressed here are those of the member 
> posting and are not reviewed, edited, or endorsed by SIGCIS. The list 
> archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ 
> and you can change your subscription options at 
> http://lists.sigcis.org/listinfo.cgi/members-sigcis.org