Welcome to Boring Topics. This episode is the first in a series on the
decline and/or fall of the majority of the home computer companies of the 70s and 80s.That
was an era of major rivalries amongst a wide array of personal computer companies, fighting
for control of the market.There are few of those companies left standing today, and only
Apple still plays a significant role in personal computing. Even the leader that initially emerged from
the fray as the undisputed winner was only able to enjoy its victory for a handful of
years before going into irreversible decline and today, although still in existence, is
no longer in the business of making personal computers. That company was IBM and its rise and fall
in the PC market is the subject of today’s video. Few companies have entered a market with such
expectations, and even fewer initially met or exceeded them so much. But IBM’s triumph was short lived. In 1980, IBM bestrode the computing world
like a Colossus. They had developed the floppy disk and hard
drive, and not only controlled the mainframe market but also held a massive portion of
the production and distribution of the ubiquitous punch cards that formed the primary input
method for millions of records into mainframes around the world. IBM’s dominance in mainframe computing had
been well established since the 1960’s, when it bet the company on the development
of the 360 series of mainframes, spending the equivalent of 42 billion 2018 dollars
on their development and securing a stranglehold on the mainframe market. History was littered with the carcasses of
companies who had attempted to take on IBM in the mainframe business. By 1980, competitors like General Electric,
Burroughs, Control Data, Digital Equipment Corporation, Cray and Honeywell were bankrupt,
out of mainframes entirely or were primarily competing for whatever business IBM was uninterested
in pursuing. Even those that did find a profitable niche,
like DEC with its line of PDP minicomputers or Cray with its supercomputers, still remained
a fraction of the size of IBM. IBM’s domination of the mainframe computer
market eventually was so overwhelming and came so close to a monopoly that the US Government
took it to court. The resulting antitrust lawsuit lasted from
1969 to 1982 and that could have potentially resulted in IBM going the way of AT&T and
being broken up into multiple “Baby IBM’s”, although the company eventually prevailed
in court. In 1980, IBM’s operations were massive. 340,000 employes, 27 billion dollars of assets,
26 billion dollars of sales, and profits of 3.6 billion dollars. The saying went, nobody ever got fired for
buying an IBM. The IBM sale reps were close friends with
their customers and many heads and upper management of businesses were close friends with IBM
upper management, to the point where they would never even consider looking elsewhere
for their companies’ computing needs, even if IBM’s mainframe offerings were inferior
to what the competition was offering. But by 1980, IBM was also a stultifying bureaucracy,
with layers upon layers of middle management and processes and procedures. Everything was studied and examined from multiple
angels, and any objection, or “non concurrence” in IBM parlance could derail a project. If a project somehow emerged from the byzantine
approval process, it tended to quickly become bloated with additional features, and massive
personnel expansion, as well as become the object of turf battles as various IBM departments
fought for control of promising, career making projects. With all of its competitors seemingly vanquished,
careers were made and broken at IBM over internal turf battles, as managers fought to vanquish
internal rivals in the only form of contest that seemed left. IBM became driven by seemingly endless stacks
of “foils” or overhead transparencies, with an individual’s credibility determined
in a large part by how well they could foresee questions and prepare foils to answer them. Some IBM managers even went so far as to have
overhead projectors built directly into their rosewood desks, so they could be immediately
ready to use their foils or have others show them their own. IBM was not entirely blind to the emerging
PC market of the 1970s, as the first wave of viable personal computers began to make
an impact on an entirely new market. The trinity of the TRS-80, the Apple II, and
the VIC-20 had shown that demand existed for a small computer that could be used by an
individual, not a massive computer that ran an institution and filled up most of a room. Although IBM chairman Frank Cary wanted a
simple machine developed for sale to home users, by 1980 all that had resulted were
two failed attempts, the Datamaster and the 5100 series, and a proposal was even being
floated to just purchase computers from video game giant Atari, and simply sell them with
IBM’s label attached. At a seminal meeting just before the 4th of
July, 1980, this last idea was shot down by an angered Frank Cary, who then listened intently
to a alternative proposal by Bill Lowe, who presented a plan to take a different approach
to developing a personal computer. However his boss Jack Rogers, head of the
enormous General Products Division, said that he had too many projects going on to fund
another one. This angered Cary, who told Lowe that he would
fund the product himself, and told Lowe to take forty people down to a small facility
in Boca Raton, Florida and keep them sequestered away from the massive IBM bureaucracy. Their task was to develop a personal computer
that would catapult IBM into the home computer market, with Cary saying “This is the kind
of machine that will really capture people’s hearts and minds. We have to build this machine.” Although he now had approval to proceed and
with Cary’s backing, a shield to protect himself from interference by other IBM managers
or layers of bureaucracy, Lowe knew that in order to bring the project to fruition within
the year time frame that he had been given, he would need to get his operating system
from someone else. Having it developed from the ground up by
the usual massive team of IBM programmers, who were paid by the “KLOC” or thousand
lines of code, would only result in a massively delayed OS. And IBM didn’t have the time. The goal was to bring the project to fruition
in the space of one year. Following that weekend, Bill Lowe had one
of his direct reports, a mid level executive by the name of Jack Sams, make a call to a
young college dropout by the name of Bill Gates. Head of a small software development firm
of some 31 people, Lowe was hoping that he could buy or license an operating system from
Gates. Through a complicated process that is worth
a separate video of its own, Gates eventually provided IBM with an operating system, one
that would eventually be called MS-DOS. Gates also provided it to IBM with a contract that
gave him the right to license its use to any other company he chose to. With a hubris born of their firm belief that
no other company would ever be in a position to use a license to MS-DOS, IBM had no issue
agreeing to that stipulation, a decision that would come to haunt them down the road. Bill Lowe set up operations in Boca Raton,
but by fall of 1980 he was gone to run another section of IBM’s sparawling operations. His replacement was a charismatic man named
Don Estridge, a 43 year old, 6 foot four inch manager who had been the project manager for
a poorly received previous IBM project, the Series/1 minicomputer. In disgrace and looking for a way to redeem
himself, Estridge threw himself into the project. A man who owned, used, and loved his Apple
II, Estridge was an excellent choice to lead the effort to create a personal computer that
would not only meet what the market wanted, but would also hopefully create a new standard
for home computers in a market currently rife with competing and incompatible standards. As 1980 gave way to 1981, the new computer
took shape. Forced to move quickly by the tight time constraints
they were working under, it was built from chips and parts developed outside of IBM. The computer was sturdy and business like,
with a rugged construction that promised to hold up well under heavy use. When it was released in late 1981 as the prosaically
named Model 5150, it created a sensation. IBM, the most powerful name in computing,
had just put their stamp of approval on the emerging personal computer market. As far as the market was concerned, the personal
computer had moved out of the hobbyist camp and was now a serious machine. The 5150 was a roaring success from day one,
with machines flying off the shelves faster than they could be built. A more powerful successor, the XT, was released
by Estridge and his team in early 1983, with a third even more powerful one, the AT released
in 1984. This propelled IBM to its highest ever share
of the home computer market, 75 percent. But problems were already beginning to appear,
as the weight of the IBM bureaucracy began to exert itself. On August 1, 1983, Estridge was forced to
expand his small independent group into a full business division, named the Entry Systems
Division. Overnight the new division went from four
thousand people to ten thousand. Layers of middle management began to appear
and multiply and the Estridge was no longer free to respond swiftly and effectively to
market pressures. Estridge also no longer had direct access
to IBM’s chairman, instead having to go through a minimum of three layers of management
before reaching Cary. Estridge’s own developers could also no
longer reach him directly, even by phone. Still, Estridge pushed forward, and started
his team working on another new product, called the PCjr. He had noticed that the original PC models
had wound up as more of a business computer than a home computer, so he felt that the
market would respond favorably to a somewhat stripped down PC, one that could be sold through
normal retail channels such as Sears and Kmart. He originally wanted the product to be very
expandable, allowing its owners to add various peripherals and plugins that could eventually
transform it into a full PC. But then IBMs various levels of bureaucracy,
and most importantly, the powerful Management Committee began to stir. The Management Committee sat at the apex of
the IBM managerial pyramid and was the ultimate arbiter of internal disputes between IBM’s
many divisions. It wouldn’t do to release a product that
could threaten an existing product line. IBM was used to dealing with mainframes that
could be sold for many years without too many changes, they were not used to the frenetic
evolution that characterized the PC market. New product lines needed to be brought out
in 18 months or less, but IBM was planning to and had committed to, selling and supporting
the original PC for ten years. So the promising PCjr was scaled way back. Its full keyboard was replaced by a “chiclet”
keyboard that found nothing but derision from its intended users. Most of the peripherals that would have allowed
it to be expanded and upgraded were canceled, prices were raised, and the plan to sell through
retailers scrapped. The PCjr limped onto the market and became
the first major IBM PC flop, with almost no takers to be found. In 1985 the product was quietly discontinued. Although IBM still represented the standard
in PC computing in 1985, and was still selling PCs as fast as they could be made, institutional
malaise was beginning to make its force felt in increasingly unfortunate ways. Development of further PCs was slow, and the
PCjr failure, as well as problems with the hard drives on many of the ATs had damaged
Estridge badly in the eyes of the IBM bureaucracy, where you were only as good as your last project. As IBM vacillated, committees were formed
to explore what the next step should be, with IBM confident that the market would wait as
long as needed for it to come out with the next product. But the market wasn’t standing still. Since the IBM PC used standard off the shelf
components and due to Bill Gates’ willingness to license MS-DOS to anybody that wanted it,
other companies began to clone the IBM PC. Companies like Compaq began to iterate faster
and faster, racing out with new and faster products far ahead of IBM’s slow and delay
ridden processes. IBM also was crippling its software development,
both internal and external. A small company called Software Publishing
was once ordered by IBM to remove a feature from its word processor that it was planning
to release for MS-DOS, that would have competed with another word processor. Software Publishing was also forced to spend
a year rewriting one of its other programs that it felt was ready for release, but IBM
demanded further changes before it would approve it as meeting IBM standards. This delay caused Software Publishing, a small
company that couldn’t afford lengthy development delays, to go ahead and release its product
on competitor’s hardware, which of course ran the same MS-DOS that Microsoft was happily
licensing to all comers. Internal applications that IBM developed were
slow, and tended to have features such as email removed to force them to not compete
with existing products for mainframes. IBM’s inability to grasp that the PC market
was different from the relatively glacial pace of the mainframe market led to them to
yet another of many fateful decisions. Towards the end of 1984, Estridge’s manager,
Mike Armstrong, had decided that a complete rewrite of DOS was needed, one that would
make up for all of DOS’s legion of inadequacies, such as being unable to run more than one
program at a time and having some very unintuitive and arcane commands, even by the relatively
easy standards of early 1980s computing. He also decided that the shiny new AT with
its spiffy new hard drive was powerful enough to last for many years as the main strategic
product of the PC division, a decision that he backed up by personally making hundreds
of calls to customers, with other sales reps making many thousands of calls themselves,
selling customers on the promise that IBM would support the AT for many years, just
as it did its mainframes. The shiny new rewrite of DOS would eventually
be known as OS/2, which is also a story for another video. Suffice it to say that IBM’s new plan was
to work with Microsoft to develop the new DOS as an eventual replacement for MS-DOS. Problems quickly began to occur as the OS/2
project eventually ballooned to more than 1700 programmers (plus management) that had
two entirely separate management teams, and were spread across four sites on two different
continents. Back at the Entry Systems Division, Estridge’s
days were numbered. A man who never liked playing in the politics
and backstabbing that characterized IBM’s hierarchy, he was removed from the PC division
in 1985 and demoted to a show position over IBM’s worldwide manufacturing, with little
actual power. That position didn’t last long though, as
on August 2, 1985 he and his wife were aboard an airliner that crashed while trying to land
in Dallas and they were among the 137 people that lost their lives. The new head of the PC division in 1985 was
the man who had originally proposed it in the first place, Bill Lowe. Although he had championed the PC idea in
the first place, Lowe still did not personally use a PC other than for email. He also began to implement even more layers
of paperwork. While Estridge had shunned memos, Lowe mandated
them. Everything had to have a complete paper trail. Progress on further evolutions of the PC slowed
even further. In June of 1985, Lowe signed a Joint Development
Agreement with Microsoft that not only formalized the codevelopment of what would be called
OS/2, but also replaced the existing DOS contract. Gates got everything he wanted from this. IBM was mostly concerned about keeping its
price for MS-DOS as low as possible, and thus didn’t care if Gates retained the rights
to continue to sell DOS and collect royalties on it from the cloners, who still only accounted
for a small percentage of the total IBM PC market. Gates was also completely free to continue
development on a little side project at Microsoft, a graphical user interface called Windows. IBM and Microsoft had an uneasy partnership
from the beginning of OS/2 development. Their processes were entirely different, with
one of the major examples being that IBM tracked its programmers progress by thousand lines
of code or “klocs”. This encouraged software bloat, whereas Microsoft
programmers were more interested in making shorter and more compact code. There was even a major issue around a Microsoft
programmer taking an IBM piece of code that took up 33 thousand characters, and rewriting
it to a mere 200 characters or 1/160th of the space. OS/2 limped along in development for years
and when it did eventually come to market in 1987, it was so bloated as to require thousands
of dollars of RAM to operate. But by that point other events had pushed
IBM’s market share even further down. By the end of 1985, there were a wide variety
of clones of the IBM PCs from both domestic and foreign companies, which were 100 percent
compatible, meaning that any software written for an IBM PC would run without issue on them. And Bill Gates was happily selling MS-DOS
to all of them, and reaping a fortune in royalties, something he was free to do thanks
to his contract. Bill Gates was also continuing development
of Windows on the side, and the first version was released in 1985, to...lukewarm reviews
and reception. But he continued working on it and thanks
to more of IBM’s mistakes, Microsoft began to wax ever stronger as IBM waned with its
one commanding position continuing to crumble as the cloners kept up a relentless assault
on both price and performance. IBM wasn’t done yet though. With its new OS continuing in development,
new plans for both its mainframes and its microcomputers, and a growing realization
of the danger Microsoft and its operating system posed, Big Blue prepared a massive
counterattack that it was confident would not only reassert its dominance in the PC
arena, but would also bring the cloners to their knees. The new plan was built around a new line of
computers and an architecture called...PS/2. Looking to 1987, IBM was confident that its
new line of PS/2 computers, together with its new OS/2 operating system, would wrest
control of both software and hardware away from Microsoft and the cloners, bringing both
firmly back into Big Blue’s hands. CREDITS "Amazing Plan"
Kevin MacLeod (incompetech.com) Licensed under Creative Commons: By Attribution
3.0 http://creativecommons.org/licenses/by/3.0/ "The Builder"
Kevin MacLeod (incompetech.com) Licensed under Creative Commons: By Attribution
3.0 http://creativecommons.org/licenses/by/3.0/ "March of the Spoons"
Kevin MacLeod (incompetech.com) Licensed under Creative Commons: By Attribution
3.0 http://creativecommons.org/licenses/by/3.0/
