[ih] W3C origins [was: Re: J.C.R. Licklider]
Jack Haverty
jack at 3kitty.org
Tue Nov 3 12:45:42 PST 2026
Hi John!
It's been a long time since the days in Tech Square. Thanks for
mentioning the Cambridge Project.
In all the time I was in Lick's group in the 1970s, I don't recall ever
hearing any mention all that time of the Cambridge Project, even its
name. Lick did disappear from Dynamic Modeling (DM) for days at a time
but we all just assumed he was off at meetings or such stuff. Lick did
tell us when he left for a year or so to return to ARPA, and explained
that he had to go back "to fix some things." While I was there, I never
knew about his second job as PI of the Cambridge Project.
However, now knowing a bit about the Cambridge Project, he did bring
some of that psychology and sociology influence into the DM group. One
of the things he got us involved in was "electronic mail", more
generically known as "computer assisted human communications." This fit
his vision of people using computers to do what they do, and especially
the notion that everyone would have their own "personal machine", which
would interact with others' personal machines using the whimsical
intergalactic network, instantiated at that time as the ARPANET and
accounts on some timeshared computer.
That led to the notion of "mail servers" always active and working to
move human communications around even when their human was absent. My
task in DM was to create the mail server; others in the DM group had the
task to make various UI interfaces between humans and the server.
One of the psychology/sociology principles that Lick introduced was the
notion that human communications needed some kind of inherent "back
pressure" to discourage overuse. If communications was free, sociology
and psychology indicate that it will get overused, and everyone will be
swamped with all sorts of unwanted input. Sound familiar...?
As engineers this made a lot of sense to us programmers. Electrical
engineers are taught that negative feedback is important to design of
amplifiers. Mechanical engineers similarly learn that engines need
negative feedback, e.g., governor on 19th century steam engines.
So in electronic mail, we tried to introduce some mechanism for such
"negative feedback". We tried to introduce the notion of "stamps" for
electronic mail. Such feedback didn't have to be blatantly financial;
even some limits such as a cap on number or size of messages sent per
day for an account would provide some negative feedback.
That idea was soundly rejected by the email community. Some thought
email should be always free. Some thought it was too much work and
more important to keep email simple (hence SMTP - the Simple Mail
Transport Protocol). In any event it never happened.
Lick's vision for "human communications" was quite broad, encompassing
not only what we now know as email but also functions such as "workflow"
to process messages through a chain of approvals in an office, or
services such as "escrow" to provide a way for trusted third parties to
verify that some particular communication had been delivered at a
certain date and time. The server I built even knew how to archive
emails to the Datacomputer by using the ARPANET to get it there or later
retrieve it, as well as make it accessible to other users on the
distribution list. In Lick's vision, a "conversation" might start as a
chat session, progress to a lengthy multi-media interchange, and become
popular enough to establish a "forum" for continued discussion. We
have all of these now, but it's very difficult to tie them together.
In Lick's vision, email was handled by computer programs talking among
themselves and passing data structures around, to do what humans had
requested. We published several RFCs (e.g., 713, 722) to get that
started but, like "stamps" it was deemed too complex. People wanted
human-readable interactions and flexibility to use their own nuances in
what became mail headers.
If Lick's vision had been followed, perhaps spam and forged email
wouldn't be as much of a problem as it has been over the decades. I've
always been disappointed that we never got back to implementing the
various sociology and psychology principles that Lick thought were
important for the intergalactic network. As more and more agents
appear on the Internet, maybe it's time for the sociology and psychology
issues to be included.
/Jack Haverty
On 11/3/26 08:45, John Klensin wrote:
> Brian,
> Apologies for the delay, but let me add three small bits to this for
> anyone who doesn't know and is curious.
>
> * The definitive biography of Lick, going back to the origins of, and
> visions behind, all of this (not just the Internet), is M. Mitchell
> Waldrop's, _The Dream Machine_, Viking Penguin, 2001. IMO, but I am
> very biased, the treatment in Katie Hafner's book falls short of
> "fully credited". Even the Waldrop book falls short in at least one
> area, something that he and I discussed a bit after the book came out
> (which was the first time I heard of the effort). In particular...
>
> * During the late 1960s and through part of the 1970s, Lick was
> Principal Investigator on two separate research projects at MIT. One
> was the Dynamic Modeling effort within Project MAC in the School of
> Engineering, where Jack was involved. The other, far less known these
> days even around MIT, was something called the Cambridge Project, an
> effort in computing in the social and behavioral sciences that was
> sufficiently interdisciplinary and actually inter-university to report
> directly to the Provost. I had something of a leadership role in the
> latter. Lick decided, in around 1972, that it was time to start
> developing and demonstrating ways that the emerging network could and
> would improve collaboration and results well beyond computer science,
> networking, and even the physical sciences, and that the Cambridge
> Project was the place to start that effort. We brought in efforts
> from around the US and, in a few cases, around the world to do that
> and some of the effects are still obvious although rarely credited.
>
> * Lick's interests and impact extended well beyond anything we would
> have thought of in terms of those two project areas at the time.
> For example, his 1965 __Libraries of the Future_ book laid many of the
> foundations for digital libraries and, some have claimed, for
> widespread use of interactive computing.._
> _
> _
> _Towering figure, with a very wide range of interests, and a really
> nice person._
> _
> _
> _ john_
> _
> _
>
>
>
>
>
> On Sun, Nov 1, 2026 at 1:07 AM Brian E Carpenter via Internet-history
> <internet-history at elists.isoc.org> wrote:
>
> On 01-Nov-26 15:50, Jack Haverty via Internet-history wrote:
>
> > Speaking of Internet History, the ARPANET et al --- There's another
> > player from the early days of The Internet that may not be well
> known to
> > Internet historians - J.C.R. Licklider
>
> He's fully credited in any serious history of the Internet, I think,
> e.g. Katie Hafner's book.
>
> > - Lick's vision was continued after he retired by his "chief of
> staff"
> > at MIT, Al Vezza, who convinced Tim Berners-Lee to come to MIT
> and lead
> > the W3C from there.
>
> I would add that Michael Dertouzos was *at least* as influential as Al
> in attracting TimBL to MIT. Probably more, because Dertouzos's old pal
> George Metakides (an MIT alum) was the senior official in the European
> Commissionin charge of funding the European end of the W3C.
>
> My understanding at the time was that Dertouzos and Metakides (they
> were both Greek) had decided that the main part of the W3C would be at
> MIT, with European Commission (EC) funding, but with a small European
> "pole". Metakides thought it would be a fine idea to place the
> European
> unit at CERN, but CERN refused to be part of an exercise whose main
> result would be to send European funds to the USA.
>
> Al Vezza was very much in the room for these discussions. I
> wasn't, but
> I knew all the CERN participants, and in fact met them in the
> departure
> lounge at Brussels airport on their way back from the critical
> meeting.
> (I was in Brussels the same day for a meeting related to growing
> Internet
> capacity in Europe.) The CERN team in that discussion, from
> memory, was
> Mike Sendall, Tim B-L, Robert Cailliau, and Francois Fluckiger.
> Maybe David
> Williams, our Division Leader, was there too. All of them except
> Tim were
> flabbergasted by the idea that CERN (a non-EC institution) should
> be used
> to route money from the EC to MIT.
>
> I heard within a few days that the CERN management definitely refused
> this. It seemed politically dangerous and was it even legal? Also
> it seemed
> outside CERN's mission as a physics laboratory. And nobody missed that
> Metakides and Dertouzos were buddies. So in the end, INRIA in France
> acted as the European pole.
>
> Regards/Ngā mihi
> Brian Carpenter
>
> --
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
> -
> Unsubscribe:
> https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature.asc
Type: application/pgp-signature
Size: 665 bytes
Desc: OpenPGP digital signature
URL: <http://elists.isoc.org/pipermail/internet-history/attachments/20261103/bae9e2c3/attachment.asc>
More information about the Internet-history
mailing list