There were more early connections than you might think. The electronic computer was largely built out of off-the-shelf parts from the then-100-year old telecom industry: relays, plugboards, vacuum tubes, etc. Leaving aside the question of their direct influence on the very first computers, both Alan Turing and Claude Shannon worked on voice encoding for telecommunications, and Shannon’s initial work on information theory was largely at Bell Labs and in that context. The transistor itself was invented in the late 1940s at Bell Labs. The modem dates back to 1949. There are lots of other examples.
So while early computers may have been largely standalone, they partly grew out of the telecom industry. They even borrowed paper tape readers from telegraph equipment, not to mention the teleprinter to be used as a computer terminal. Most undersea data cables today still follow the routes laid down in the 19th century for telegraphy.
In fact, part of the reason most early computers were standalone is that communication in general was already well-automated and thus cheap. By contrast calculation and data processing (the latter even with punched card equipment) remained very expensive, and was an attractive market for early computer companies.
Computers communicated with their own peripherals from early on, but most early companies saw little market value in making them connect easily with each other. Only specialized users like militaries bothered to try until the later 1960s, and the rest is networking history.
Many histories tend to focus on either computing or telecom. For a reference that looks specifically at the intertwined history of computing and telecom/datacom, I suggest Jim Pelkey’s
History of Computer Communications, which we host online at CHM along with transcripts of his 80+ original interviews (
https://historyofcomputercommunications.info/). Jim and Andrew Russell are doing an update of the book with ACM Press.
Best, Marc
p.s. In terms of air gaps, note that even standalone early personal computers got viruses! They simply arrived via floppy disk – sneakernet – rather than actual networks.
Curatorial Director, Internet History Program
Computer History Museum, 1401 N Shoreline Blvd., Mountain View CA 94043
Dear SIGCIS members,
I'm a freelance writer/journalist who's published in multiple
news outlets on hacktivism and who's lurked on this email list for
some time. The past several months, I've been reading a great
dealing about Alan Turing and the math behind Computable
Numbers (fundamental theorem of arithmetic, Gödel
encoding, etc). A fairly straightforward question occurred to me,
one I hope this list can help answer:
As is well known, Turing's 1936 paper Computable Numbers
invented the concept of a universal machine, which includes what
today would be called an airgapped computer. For quite a while,
all computers (universal machines) were airgapped devices. The
historical casual chain is clear: first the idea documented in Computable
Numbers came into existence, and only later are physical
computers actually built, initially as standalone, airgapped
devices.
But how did plugging computers into one another with wires/cables
begin? Did a thinker first conceive of a profound idea
underpinning wired/cabled networking, and then only later,
engineers implemented that concept in the physical realm? Or, did
people first begin hooking computers up to one another, perhaps
experimentally, and then a theorist subsequently created an idea
to describe/frame what was happening (maybe a mathematical graph
theory or something)?
To put it another way, in terms of a simple standardized
test-like verbal analogy, Computable Numbers is to
airgapped computers as ??? is to wired/cabled networking of
computers.
I omit wireless connections (e.g., Bluetooth) for the time being.
Thanks much,
Doug Lucas
_______________________________________________
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