How Ma Bell Shelved the Future for 60 Years [telecom]

What would the world be like if fiber optic and mobile phones had been available in the 1930's? Would the decade be known as the start of the Information Revolution rather than the Great Depression?

The Great Bell Labs

In early 1934, Clarence Hickman, a Bell Labs engineer, had a secret machine, about six feet tall, standing in his office. It was a device without equal in the world, decades ahead of its time. If you called and there was no answer on the phone line to which Hickman's invention was connected, the machine would beep and a recording device would come on allowing the caller to leave a message.

The genius at the heart of Hickman's secret proto-answering machine was not so much the concept - perceptive of social change as that was - but rather the technical principle that made it work and that would, eventually, transform the world: magnetic recording tape. Recall that before magnetic storage there was no way to store sound other than by pressing a record or making a piano roll. The new technology would not only usher in audio cassettes and videotapes, but when used with the silicon chip, make computer storage a reality. Indeed, from the 1980s onward, firms from Microsoft to Google, and by implication the whole world, would become utterly dependent on magnetic storage, otherwise known as the hard drive.

If any entity could have come up with advanced recording technology by the early 1930s it was Bell Labs. Founded in 1925 for the express purpose of improving telephony, they made good on their mission (saving AT&T billions with inventions as simple as plastic insulation for telephone wires) and then some: by the 1930s the laboratories had effectively developed a mind of their own, carrying their work beyond better telephones and into basic research to become the world's preeminent corporate-sponsored scientific body. It was a scientific Valhalla, hiring the best men (and later women) they could find and leaving them more or less free to pursue what interested them.

When scientists are given such freedom, they can do amazing things, and soon Bell's were doing cutting-edge work in fields as diverse as quantum physics and information theory. It was a Bell Labs employee named Clinton Davisson who would win a Nobel Prize in 1937 for demonstrating the wave nature of matter, an insight more typically credited to Einstein than to a telephone company employee. In total, Bell would collect seven Nobel Prizes, more than any other corporate laboratory, including one awarded in 1956 for its most famous invention, the transistor, which made the computer possible. Other, more obscure Bell creations are nevertheless dear to geeks, including Unix and the C programming language.

formatting link
or:

formatting link

Reply to
Joseph Singer
Loading thread data ...

The statement is grossly inaccurate in terms of the time line. Magnetic tape for information processing was developed in the early

1950s, long before silicon chips. It quickly developed into a major storage medium for computers. Magnetic drums, using the same principle but on a hard surface, were in service even earlier. Magnetic disks were invented and utilized by IBM in the mid-1950s. Mmagnetic core memories were another 1950s development.

Magnetic wire was utilized for crude voice recording prior to the development of tape.

Bell also used strips of sound movie film to play back pre-recorded messages, such as passing a phone number to a manual switchboard operator.

Did telephone wires use plastic as an insulator that far back? I thought that came in the 1960s, before that paper, textile, and rubber were used. While some plastics were available in the 1940s, I thought their big growth was after WW II.

Anyway, I don't see in the above where magnetic recording was _perfected_ as a _usable_ technology, or an explanation why Hickman's invention wasn't utilized. Note that many years often go between the time something is invented and the by the time it can be inexpensively manufactured and meet industrial standards. It took ten years for the transistor to be developed into something that would be cheaper and more reliable than the vacuum tube, and several decades more before it finally replaced all applications of the vacuum tube.

The Bell Labs history specifically states that AMA used paper tapes when developed in the 1950s because magnetic recording technology wasn't ready.

Reply to
Lisa or Jeff

A point of disagreement verified by tens of millions of music lovers.

I distinctly remember the group Les Paul and Mary Ford who pioneered overdubbing, tape delays/echos, and multitrack recording in the mid-1950s.

Bing Crosby also invested in the founding of Ampex which was using magnetic tapes for sound recording, and the first 8-track audio tape player/recorder was developed by Ampex in 1954.

More info here:

and

and

Germany had the Magnetophon in the 1930s.

More info here:

and

also claims audio magnetic tape recorders beginning in the 1930s.

Reply to
Thad Floryan

I'm not sure on telephone use, but plastic insulated wire pretty much became S.O.P. in consumer-grade radios in the years immediately after WWII, although cloth insulated wire was also used into the early 1950s. Cloth insulation was the normally used material thoughout the 1920s and 1930s until the start of the War and rubber was used in a lot of sets from the very late 1930s until the War interrupted production. Rubber had a nasty habit of drying out and rotting, causing electrical hazards.

Reply to
Michael G. Koerner

Speaking of insulation lifetime....

What is the lifetime of telephone wire located within a home? I suppose the plastic gray "D" station wire of the 1970s is pretty durable. But what about textile insulation used for the mass of houses built immediately after WW II? What about telephone house wiring of pre-war houses?

When modular phones came out people would convert the old junction box or wall mount to modular, but I doubt they rewired the lines. The outside plant was taken care of by the phone company and presumably old drops to a house would get renewed automatically or if a subscriber complained of trouble.

What happens when those old internal wires are used to carry things like DSL? Is there problems with leakage or capicitance?

Reply to
Lisa or Jeff

I believe that statement is now being taken out of context.

For the purposes of DATA use, I would readily believe that in the 1950's that "magnetic recording technology wasn't ready".

I clearly recall the limitations of open-reel, cassette and other Audio magnetic media even in the 1970's and even the most expensive equipment was a constant battle with precision alignments and cleaning to get the best out of them.

-- Regards, David.

David Clayton Melbourne, Victoria, Australia. Knowledge is a measure of how many answers you have, intelligence is a measure of how many questions you have.

Reply to
David Clayton

The other issue is cost. It's hard for us to believe today, but back in the 1950s electro-mechanical equipment was cheaper than electronics. It wasn't until 1962 that IBM sold more computers than relay tabulator machines, and tab machines remained the most cost- effective choice for some applications for at least another decade.

Even when a customer acquired an IBM electronic computer, they usually still got tabulator gear to handle front-end and rear-end processing, leaving only certain parts of the work to be done by the computer. That is to say, there was considerable work for which it was cheaper to run on a tab machine than an electronic computer, and this continued well into the 1970s. (Most of us have seen computer rooms that had an old sorter or tab printer for auxillary work.)

The Bell Labs history describes the AMA paper tape punch as a [relatively] inexpensive mechanism due to using countersunk holes instead of expensive dies to make the punch (there's a lot more on this subject in the book). I've seen them in operation and they looked very complex. But I suspect in terms of maintainability and first cost they may have been cheaper than an electronic tape drive and reader and associated support equipment, especially in the 1950s.

(Air conditioning may have been required as well for electronics, which most central offices did not have back then.)

The unit I saw in regular service was in the mid 1970s. Why it hadn't been replaced by magnetic tape I can't say, but as mentioned, at that time electro-mechanical gear was still widely used in both the Bell System and in information processing in general. Also, the Bell System tended to build its equipment to very durable standards and then use it for many years.

Remember too that the _peak_ of step by step deployment wasn't until

1974, despite No. 5 crossbar and ESS out by then.

I believe it was cost reductions in electronics into the early 1980s finally made it worthwhile to dump old electro-mechanical gear en masse. This applied to both telephone switching and information technology. (For example, in IT, converting data entry from keypunching cards to key data to tape/disk or on-line systems. Early on-line systems were not cheap--they needed extensive CPU and disk space, plus terminals and line controllers, all of which were expensive.)

Wasn't a widespread ESS in that era the No 5? When did that come out?

Reply to
Lisa or Jeff

I wouldn't. In 1951 the UNIVAC I used UNISERVO magtapes as its primary I/O device. If you wanted to use cards, they later provided offline card to tape and tape to card units.

I hope this excerpt isn't typical of Tim Wu's new book, because it manages to pack a phenomenal amount of wrongness into every paragraph.

R's, John

Reply to
John Levine

IBM would disagree. In 1949 they sought a successor to punch cards and in 1951 that effort came to fruition as the Tape Processing Machine:

I still have one reel-reel stereo audio tape machine from ages ago and I never had to "play" with its heads and cleaning took just seconds since all the parts that could accumulate tape oxide were open and readily available. With the "good" tapes I'd use it was never a problem.

I also (still) have two Teac "Esoteric Series" model 860, arguably the finest cassette deck ever made (picture taken just a few minutes ago), and they never needed any head alignment. The only adjustments required for normal use is bias and EQ for a specific tape. I do have the service manual and it, too, states no adjustments required unless the head assembly is replaced, after which the adjustments are locked-down "with a drop of locking paint."

FWIW, I also have an R. B. Annis Model 25 magnetometer used for checking residual magnetism in tape heads and guides, and demagnetizers to "fix" any problems found. Detail here:

And to be really perverse, I still have a functional paper-tape reader:

that still reads my paper tapes from the 1960s.

:-)

Reply to
Thad Floryan

Cloth was the outer covering of wires around a rubber compound. Older plastics weren't very good either after 10 years or so. Today, we use thermoplastics.

There is a scary amount of wiring out there decades passed its usable life.

Reply to
Adam H. Kerman

They 'usually' work OK. wire lengths are short enough that the "deficiencies" do not rise to the level to bar functionality.

Not those specific deficiencies, as much as inductive pick-up due to the lack of 'twist'.

Some years back I lived in a 1964 construction 32-unit (8 units/floor, 4 floors) apt building, about 6,000 wire-feet from the C.O.. One 50-pair drop to the building, terminated on screw terminals. The 'house' wiring was "quad" (J-K), _in_conduit_, with the 4 vertically aligned units sharing a common conduit.

The Installer's test gear gave a 'stress' reading of "34" at the wall-jack in my unit. It read _"6"_ at the 50-pair termination. I don't know just _what_ 'deficiencies' it was measuring, or whether the scale was log or linear, but it was clear that _most_ of the issues were in the circa 80' of quad wiring from the building DMARC to my unit, and not in the 'F1' run from the C.O. to the building.

Reply to
Robert Bonomi

I'm surprised that nobody has mentioned gutta percha. Gutta percha was amazing stuff: biologically inert, tough and hard at room temperature, but easily formed or molded when heated. During late 1800s and early

1900s, it was used in variety applications: telegraph and telephone wire insulation, golfballs, fake-wood furniture, pistol grips. Dentists used it to fill teeth after root canal surgery. Gutta percha is obtained from the sap of the gutta-percha tree, indigenous to southeast Asia and northern Australasia. [1-3]

The Copper Development Association's website reports that:

| Gutta percha and various rubber compounds were used for insulating and | water proofing the telegraph and early telephone cables. Telephone | cables were employed for aerial, underwater and underground use | around 1879. Early cables were single grounded wires followed by | metallic circuits lines after their development. By 1887 all of the | newly manufactured cables were metallic circuit cables. Some of the | early telephone cable applications were across the East River Bridge, | under the North River between New York and New Jersey and across the | Delaware River from Philadelphia to Camden. There were numerous cable | manufacturers. The cables made by the different manufacturers were | very similar but not identical. The cables contained up to 100 copper | wires. They were insulated with cotton, cotton impregnated with | paraffin, gutta percha or a rubber compound, and then in wrapped | in lead. [4]

References:

[1] "Gutta-percha." Wikipedia.
formatting link
[2] "Submarine communications cable." Wikipedia.
formatting link
[2] "Telephone Cables" from The history of electric wires and cables, Chapter 14. Robert Monro Black, Science Museum (Great Britain)
formatting link
[4] "The Evolution of Telephone Cable." Copper Development Association Inc.
formatting link
Neal McLain
Reply to
Neal McLain

Good point. Gutta percha is what made the transatlantic telegraph cable possible in the 1850s

The company that manufactured it was creatively called The Gutta Percha Company. After a few name changes, it's still around as Cable and Wireless, which runs the telcos in a lot of British colonies and ex-colonies.

R's, John

Reply to
John Levine

I ran the AT&T Silicon Valley UNIX Users' Group for the entirety of its existence. Our meetings were held in the huge AT&T (then) building on Duane Avenue off Lawrence Expressway in Sunnyvale CA [coordinates for Google Earth = 37° 23' 11.38"N, 121° 59' 57.77"W].

Circa 1985 the fine folks of AT&T gave us a tour of the (test) 5ESS located in that building. Amazing. Though I didn't know when 5ESS was "deployed in the wild", it clearly existed then.

This page claims the 5ESS first appeared in Seneca, Illinois (815 Area Code) in 1982.

Reply to
Thad Floryan

I was trained in the 1970's to service magnetic tape equipment using things like tension gauges, alignment tapes etc to get the absolute best out of these things.

I worked on some reasonably high-end equipment at that time that still needed a lot of care and attention to meet the requirements for audio, so that's why I believe that for data use the medium just was too finicky.

I still have a Teac open reel machine holding up some other junk somewhere, as well as some 8" floppy disks which may well shed their oxide now if I still had a drive to put them in!

Reply to
David Clayton

One could trace the beginnings of tape recording back even further to Bing Crosby's postwar schedule. Due to his popularity as an actor and stage performer, he insisted that he be able to record his NBC radio program. This was back in 1945. He'd heard about (and possibly saw a demo of) magnetic recorders the Germans had developed during WWII.

NBC's refusal to allow recordings caused Crosby to walk off his show for half a year, upsetting his sponsor (shows had permanent sponsors in those days) until NBC capitulated.

But it goes back even *further* than that! Both the ABC (a poor spinoff of NBC) and Mutual (an always poor network) had used recordings in one form or another since at least the late 30s, though these were mostly on disc until mag came into its own.

Reply to
David Kaye

It certainly wasn't ready at Safeway's data center on East 14th Street in Oakland in the early 1970s. I had a temp job there pushing a shopping basket filled with punch cards to pick up and deliver to programmers. My first tech job!

Reply to
David Kaye

And... it wasn't really ready. Herman Lukoff's book _From Dits to Bits_ describes a lot of the headaches they had to go through in order to make it work.

"Perhaps the most radical idea which business is being asked to accept is that idea that a reel of tape can safely be used to carry information now being entrusted to visual card files... the adequacy of tape for this purpose has not been sufficiently demonstrated... we are not quite sure that [tapes] are sufficiently safe from accidental erasure, loss of information through breakage, kinks, dimensional instability, flaking, and other such occurrences. Nor have we been satisfied that the devices currently being employed to read and write on magnetic tape can be relied to do so with accuracy." --- M.E. Davis of Metropolitan Life, at the 1953 Joint Computer Conference

It's got better since then.

--scott

Reply to
Scott Dorsey

In the days when I was more ignorant, I bought a spool of this kind of wire. It was twisted. I have no idea if the twisting was tested and avoided any kind of antennuation. Was it?

I kicked myself as Cat5 was available at the time, although more expensive, and I would have avoided some misery trying to keep a modem connection up with a nearby radio station and other sources of interference.

Is there an actual problem with screw terminals, other than the amount of space the consume versus a punch-down block? Is oxidation a concern?

A friend says he used to have a DSL line at his house to help him monitor computer networks at his business on this kind of ancient inside wiring.

Reply to
Adam H. Kerman

'quad' wire has the four conductors stacked in a square arrangement, with a gradual (guessing at circa 1 'turn' per 8",-- don't have any handy to check) twist of all 4 wires _as_a_unit_ -- a 4-strand helix, as it were. Thus, with respect to the other wires in the cable, its just as if it was absolutely straight conductors. Even if there is 'nothing' on the other wires in the cable, you've got inductive coupling from the linear 'single- turn transformer".

'twisted pair' cable (i.e. 'category 'anything') has twist rates for each =pair= in the cable, providing far superior 'cancellation' of inter-pair pick-up. higher 'cat' numbers have higher twist rates, and require subtly different materials to minimize the capacitance between the conductors in a 'pair'. 'Cat 1" (voice grade) and "Cat 2" (up to 4-mbit/sec) don't "formally" exist as standards, although the original classification (by Anixter, a major wire distributor) did include them. ("quad" wiring is one type of "cat 1" -- a catch-all for stuff which doesn't meet any higher specification.) "Cat 2" is adequate for 'ARCnet' and similar (up to 4mbit/sec) over twisted-pair. Cat 3 for 10-mbit Ethernet. "Cat 4" is needed for 16-20 mbit/second (high-end Token ring, the early "100TX-4" fast Ethernet, etc. "Cat 5" for modern 100-mbit fast Ethernet, or limited-distance Gig-E. "Cat 6" for Gig-E or limited-distance 10gig Ethernet. "Cat 6a" lets one run 10gig E over _unshielded_ pairs, for the full distance of the standard. "Cat 7" was the original design for 10gig Ethernet, using individually _shielded_ pairs within the cable. 'Cat 7a' is "theoretically" good for 40gig Ethernet at up 50m, and 100gig Ethernet at up to 15M.

It is, as they say "performance limiting". when the "terminals' for a 'pair' are more than an inch apart, it is _very_ difficult to maintain, say, the Cat 5 required 'twist rate' to within 1/2" of the connection point (which _has_ to be an 8P8C, per the standard).

Also, "as installed", the cross-connects between the termination of the

50-pair drop to the building and the termination block for the 'house' wiring was all _single-strand_ hook-up cable. easily 6'-plus lengths.

Oxidation, as such, was not an issue. While the feed cable was underground all the way from the C.O., each pair went through a 'protector' before reaching the cross-connect terminals.

It's -not- all that 'ancient'. I don't have a hard date for when the industry shifted to using 'twisted pair' for most/all _residential_ wiring, but 'quad' was still being commonly installed in the mid- to late- 1980s. _Commercial_ installations had -- mostly, if not entirely -- cut over to 'twisted pair' in the prior decade, for wiring installs, driven by digital phone systems, the explosion of computer terminals that could use a 'common' structured wiring system -- e.g. IBM 3270 over twisted-pair, vs. dedicated coax. The cost advantage of 'structured' wiring in a commercial environment was a *very* "compelling" argument.

Reply to
Robert Bonomi

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.