Ronda Hauben on 11 Feb 2001 15:26:37 -0000


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

[Nettime-bold] The origins of the Internet - mailing list discussion



There's been an important discussion on the community memory
mailing list about the origins of the ARPANET and Internet,
which started with a post by Les Earnest critiquing the 
history channel (cable tv in the US) account of how
the Net was started by desire to hook together the terminals
to three different time sharing system supported by ARPA's
IPTO.

I thought this discusion would be of interest to those
on Nettime - and so am sending to nettime this post I 
recently sent to the community memory mailing list about
the narrowness of the discussion thus far.

>From ronda@panix.com  Sat Feb 10 19:11:20 2001
>Date: Sat, 10 Feb 2001 19:11:20 -0500 (EST)
>From: <ronda@panix.com>
>To: CYHIST@MAELSTROM.STJOHNS.EDU
>Subject: Re: [CM>] Origins of the ARPANET


vint cerf <vcerf@MCI.NET> wrote:

>I think it is fair to say that the History Channel program used Bob Taylor's
>remarks (repeated in Katie Hafner's book) in an oversimplified way. I think
>Bob actually felt motivated by the belief that the systems he used required
>three distinct terminals (for whatever reason) and that it should have been
>possible to reach all three timeshared machines with one. Rightly or wrongly
>this led to the notion of networking things.

>Plainly the outcome solved a lot more than that "problem"


Yes and such oversimplification leads to myths that impede the public
understanding of the origins of the ARPANET and then the Internet.
And hence can also be harmful in the dissemination of the needed 
public knowledge to support the continue the development of the Internet.

That is why it is good Les made an effort to challenge such 
oversimplifications.

With regard to the origins of the ARPANET, JCR Licklider explains
(I think it was in his interview with the Babbage Institute) that
he had a vision for the need for an Intergalactic computer network, 
but that it was too early to start it when he started at ARPA 
and started the Information Processing Techniques Office in 1962.

However several people including Larry Roberts say that Lick's
vision of the importance of doing networking was an inspiration
to them. And Ivan Sutherland in his interview with the Babbage
Institute describes how he tried to start a networking program
at UCLA during his turn at IPTO, but that that didn't then succeed.
Also I think Steve Crocker in his interview with the Babbage
Institute mentions having known of or having been part of 
this earlier effort to start a networking program by ARPA/IPTO.

Also ,in his Babbage Institute interview, Ivan Sutherland suggests 
that it was important to have researchers who would take on the 
problems of a networking research program, and he suggests that
that was one of the important contributions that made the 
ARPANET research program a success as opposed to the problem
that hadn't been taken on by the earlier UCLA program.
This helps to suggest that it wasn't only a question of those
who would fund a networking research program, but that identifying
and welcoming the contributions of capable researchers was an
important task for ARPA/IPTO to have taken on which contributed
to the success of the ARPANET and then the Internet research
programs at ARPA.


It is good that Bob Taylor and Charles Herzfeld decided to fund
a networking program. But it is important that why the ARPANET
program was able to be started, not be looked at in that narrow
perspective.


Also it is important that the ARPANET not be confused with the 
Internet, as it so often is in the portrayal in books and 
tv etc.

The ARPANET program pioneered packetswitching and many other
important developments, it was one big network under the control
of one entity.

The Internet, via the creation of tcp/ip was able to change the 
paradigm, by making it possible for diverse networks, under
different administrations and with different local characteristics
to communicate. The Internet made it possible to create a metanetwork
of networks, and hence to have an Internet.

This is an important development that needs to be understood
in its contribution.

While connecting different terminals was crucial for the development
of the ARPANET (for example this was as I understand it a special
challenge that was taken on at the 1972 ARPANET demonstration
at the ICCC'72 in Washington), it was not yet taking on the problem
that had to be solved to create an Internet.

That problem had to do with making it possible for different networks
to communicate. France for example by the 1972-1973 period was 
creating Cyclades under Louis Pouzin and England the NPL under
Donald Davies and others. 

It would not be possible to expect these countries to subordinate
their networks to the ARPANET, and to ask them to become a part
of the ARPANET.

By 1973 the need was recognized for the creation of an means of making 
possible communication across the boundaries of dissimilar networks.

The problem was taken on in the paper describing TCP/IP (then 
called TCP) which was presented at an international network working 
group meeting in September 1973 in Great Britain and published by
Vint Cerf and Bob Kahn in May 1974 in the  IEEE Transactions on 
Communications, vol Com-22, N. 5, May 1974, pg 637.

It seems important to recognize the fact that it is the creation
of the TCP/IP protocol and the ability to have dissimilar networks
communicate that it makes possible that is at the essence
of what has made the Internet possible.

I have been working on some draft papers about these developments
and welcome comments on them.

They are online at 

http://www.columbia.edu/~rh120/other/

Two particularly relevant papers to this discussion include

 
part IV http://www.columbia.edu/~rh120/other/computer-communications.txt
part V http://www.columbia.,edu/~rh120/other/birth_internet.txt


Ronda
ronda@panix.com
http://www.columbia.edu/~hauben/netbook/









_______________________________________________
Nettime-bold mailing list
Nettime-bold@nettime.org
http://www.nettime.org/cgi-bin/mailman/listinfo/nettime-bold