[ih] Invention of The Internet - circa 1920

[John Day]

In the late 19thC at least in Europe, telegraph was visual, not auditory. (Yea, I was surprised too.)  The reason I know is a very good book I highly recommend:
Basil Mahon, The Forgotten Genius of Oliver Heaviside.
Heaviside had scarlet fever as a child and was badly hard of hearing. But because telegraph was visual, he became a telegraph operator for one of the first undersea cable companies from the UK to Norway. Later becoming chief operator, worked out the theory of transmission lines (I learned why I had a semester course for my undergrad EE), and translated Maxwell into vector calculus and much else. Heaviside was entirely self-taught.  A pretty amazing guy.  (I am guessing that the telegraph scorched marks or holes on a paper tape and that punched paper tape came later, but maybe not.)

What Karl is describing is really message switching. Packet switching broke up the messages and allowed interleaving different pieces of messages, so that short messages didn’t have to wait behind long ones. (Still delayed, but shorter completion time. Think OS scheduling.) 

While Baran’s interest in inventing packet switching was survivability and the military applications, Davies wasn’t doing work for the UK military and resiliency was distinctly secondary. His interest was greater efficient use of the lines. Think of it as messages switching was FCFS batch processing, while packet switching was multiprocigramming. So virtual circuit is multiprogramming with contiguous allocation, and datagrams were a tool for the next step. The big surprise was that at the time, nothing else was needed, except congestion control, which Davies recognized in his 1966 paper, as did CYCLADES when they adopted it in 1972 and funded Waterloo to work on it.

That’s the short version.

BTW, before you read the Heaviside book.
Read, Nancy Forbes and Basil Mahon, Faraday, Maxwell, and the Electromagnetic Field on how they figured out E&M. The establishment kept trying to make it Newtonian and it wasn’t of course. It took 3 outsiders to figure it out.  It is a fascinating story and sets the stage for Heaviside.

Take care,
John

> On Nov 28, 2023, at 05:31, Vint Cerf via Internet-history <internet-history at elists.isoc.org> wrote:
> 
> Karl,
> I think you have a reasonable point regarding store/forward and telegraph.
> The early days of telegraphy involved headset, handset, pencil and paper.
> Teletypes used paper tape and the messages were punched onto tape, fed into
> a teletype and fed down the line. The operators would tear off the tape,
> hang it on a peg waiting to be forwarded to the next telegraph station.
> This was called 'torn tape" and literally was store (on the peg) and
> forward (feed to next teletype connect by dedicated circuit to the next
> hop).  AUTODIN was a 1960s store and forward messaging system that emulated
> this but all electronically.
> 
> v
> 
> 
> On Tue, Nov 28, 2023 at 3:18 AM Karl Auerbach via Internet-history <
> internet-history at elists.isoc.org> wrote:
> 
>> In a couple of pieces that I've written/recorded I tried to nail the
>> start of the net sometime in the 1830s with the invention of the
>> electric telegraph.  (But, truly it is an exercise as fruitful as trying
>> to nail Jello to a ceiling.)
>> 
>> The reason that I picked that was that the electric telegraph was an
>> electronic store-and-forward packet switching system.  That is if one
>> equates telegrams with packets.  The store-and-forward part came from
>> the manual writing-down and then transmitting on the appropriate
>> outgoing link at relay locations along the path from the source of the
>> telegram to the destination.  And whether said in jest as a pun or being
>> serious it is the case that the signalling on the early telegraph
>> network was quite "digital", being driven by finger -digit - action.
>> 
>> I tend to not give much credit to the voice telephone system as a
>> progenitor of the net as it was largely end-to-end circuit switched and
>> analog.  (At a later stage I think that the telephone systems' work on
>> imposing modulated signals onto various media was a significant, even
>> major, contribution, but a contribution to a design already established
>> by the telegraph system.)
>> 
>>     --karl--
>> 
>> 
>> On 11/27/23 11:14 PM, Jared E. Richo via Internet-history wrote:
>>> 
>>> The Marconi Wireless Telegraph, invented circa 1902/1903 [1], set the
>>> foundation for a LOT of modern technology. It's where I begin in my ~
>>> 120 years of Vulnerability History talk.
>>> 
>>> So in this example, just under 20 years later, but before we saw
>>> wireless used for transferring encrypted/encoded comms, which led to
>>> another 'fun' chapter in that history (WW2)?
>>> 
>>> It tracks =) Hard to say if they did any research, but the arbitrary
>>> (?)  timeline is believable, especially if there were no wars,
>>> corporate espionage, or whatever else looming at the time.
>>> 
>>> .b
>>> 
>>> [1] While that date is more arguably established, the relevance to
>>> where I begin my talk is a tad more murky. The demo from Marconi and
>>> his assistants happened at a given time, yes! But the six+ lead-up
>>> that led to that event happened before the public articles I have seen
>>> give any attribution to. So I am speaking at "technology inception" vs
>>> "technology demonstration" vs "technology hacked" vs "omg why was it
>>> hacked on the day it was 'unveiled'?!". It's a bit nuanced, especially
>>> via the lens of modern vulnerability disclosure timelines. To this
>>> day, it is perhaps the most valuable use-case for why it matters.
>>> 
>>> On 11/27/2023 11:47 PM, Jack Haverty via Internet-history wrote:
>>> 
>>>> Yes, it's fiction, but I just saw an interesting episode of Murdoch
>>>> Mysteries, in which the Internet is invented, over a century ago,
>>>> with lots of its advantages and foibles revealed.   If you get a
>>>> chance to se it, it's an interesting alternative view of Internet
>>>> History, and commentary on the real Internet of today.
>>>> 
>>>> https://www.imdb.com/title/tt18602066/
>>>> 
>>>> The Inventor, in the TV show, also wears a 3-piece suit.
>>>> 
>>>> Jack Haverty
>> --
>> Internet-history mailing list
>> Internet-history at elists.isoc.org
>> https://elists.isoc.org/mailman/listinfo/internet-history
>> 
> 
> 
> -- 
> Please send any postal/overnight deliveries to:
> Vint Cerf
> Google, LLC
> 1900 Reston Metro Plaza, 16th Floor
> Reston, VA 20190
> +1 (571) 213 1346
> 
> 
> until further notice
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history