As kids, Bill Gates and Paul Allen dreamed of having their own Fortune 500 company.[1] The two became friends (and sometimes adversaries) when both attended the prestigious Lakeside School in Seattle in the early 1970s. But their friendship was mediated by a third entity, a Teletype ASR-33 connected remotely to a computer. As video terminals were rare and expensive, this popular teleprinter was often recruited to provide an interface to computers. Lakeside used the ASR-33 to connect to a timeshare service offered by General Electric (GE) in the Seattle area.

However, based on sharing the resources of a PDP-10, the service soon proved to be expensive. Despite an infusion of $3,000 into the Lakeside computer account by the Lakeside Mothers Club, the boys (Lakeside was an all-male school) soon ran out of computer time. But luckily, one of the mothers of a Lakeside student was a co-founder of a new company called the Computer Center Corporation that offered students computer time on their PDP-10 in exchange for helping them debug their software.

Allen and Gates became quite proficient with the machine, and even after the project ended, they continued to “hack” into the machine. Although they were eventually caught, they nevertheless gained a notorious but beneficial reputation for their hacking. But rather than continuing with hacking, they instead went into business, forming the Lakeside Programmer’s Group with a few other students.[2]

The Lakeside Programmers Group primarily provided computer services in exchange for coding time but provided the foundation for their next enterprise, Traf-O-Data. This new company was formed in 1972 to sell computer traffic-analysis systems to municipalities. They planned to string rubber cables across roads and highways and use a microprocessor to develop statistics on traffic flow.

Their technology was based on an Intel 8008 chip that had enamored Allen and who subsequently seduced Gates into helping him develop a BASIC interpreter for it. By this time, Allen was enrolled at Washington State University, and his dorm became their headquarters. The 8008 was the first 8-bit microprocessor but working with it was a bit awkward. They used an IBM System 360 on campus to simulate the 8008.

At the same time both were hired by defense contractor TRW to develop a simulator for the 8008 chip.[3] The two labored and eventually came up with a workable system for the car counter but it had difficulties with its paper-based printer. Traf-O-Data was not very profitable as it soon found itself competing against free services from the State of Washington. Their final Traf-O-Data product was not really a computer but it provided valuable experience for the two young programmers.[4]

After Gates’ graduation, the two were in Boston, where they took a bold move towards their vision of creating a major computer company. In the winter of 1975, Paul Allen, who was now working for Honeywell picked up the January issue of Popular Electronics at Cambridge’s Harvard Square. Excitedly, he took it to his friend Bill Gates’ dorm room, who was enrolled nearby at Harvard University.

The magazine issue sparked their quest to enter and impact the new computer era. They saw their opportunity to leverage their experience with BASIC and gain a foothold in the emerging microcomputer industry. The magazine showed a low-cost microcomputer built around the Intel 8080 chip and was in desperate need of a programming language. The Altair, as it was called, was actually a kit that had to be assembled by the purchaser. It was marketed by a company called MITS (Micro Instrumentation Telemetry Systems).

Gates and Allen had been following the development of the 8080 chip. Using Traf-O-Data stationary as their institutional identity, they contacted the developer of the MITS machine to offer their services. The two had some experience with Intel chips from their Traf-O-Data days, so they designed a simulation of the Altair’s 8080 chip on a Harvard PDP-10 to build a version of BASIC that would run on the Altair. The key was their questions about how to use a Teletype machine to read and input data. This call convinced the MITS people that they were serious. Just a few months later, Allen flew to New Mexico to present their software. The memory in the Altair was so small that they were unsure their version of BASIC would work. But after a day’s delay and some last-minute software adjustments by Allen, the software worked, and the demonstration was a success. It even allowed them to play a game of Lunar Lander on the Teletype printer.

Allen was offered a job on the spot and became Vice-President of Software at MITS. Gates flew down in the summer and helped out while they worked on their new company at night. On July 23, 1975, the two companies signed a contract giving MITS exclusive rights to their BASIC with royalties going to their new company Micro-Soft. The relationship between MITS and Micro-Soft soon soured. Microsoft was not making much money from the deal. MITS sold BASIC for $75 with the kit while charging $500 for it separately. It became extremely attractive for Altair users to trade bootleg copies of BASIC rather than buying it from MITS. Although the Altair depended on BASIC to do anything useful, MITS saw its business as selling the hardware, not the software. Consequently, marketing the software was not a priority.

On February 3, 1976, Gates sent his infamous letter accusing most Altair owners of stealing the BASIC software by duplicating the paper tape. He claimed that only 10% of them had bought BASIC. Soon Gates got his father, a successful lawyer in Seattle, involved in a lawsuit to get BASIC back from Pertec, the company that had bought MITS in the meantime.

The year 1977 was a decent year for Micro-Soft, with $381,715 in revenues.[5] They got BASIC back and in August began negotiations with Apple to license its programming language for $21,000. Micro-Soft produced versions of BASIC for new processors as they came out, including the 6502 that Wozniak was using for their Apple IIs and the TRS-80 by Radio Shack. The latter’s marketing capabilities made an extraordinary impact on the popularity of the microcomputer, and Micro-Soft’s BASIC was on most of them. When the Commodore PET was designed, Micro-Soft also got the call to provide their BASIC.

Allen, Gates, and crew worked on the project despite the agreement that royalties would not be forthcoming until it shipped the next year. Ironically it was support from Apple that provided a basic level of financial backing for Microsoft. Although Wozniak had designed a BASIC early on for the Apple II, it was not the “floating point” version that many users were requesting. Micro-Soft soon developed a version of BASIC for the Apple II and received a check for $10,500 as an initial part of its 10-year license fee. It was one of the rare times that Gates allowed software to be licensed on a flat-fee basis, rather than requiring a royalty payment on every copy sold.[6] The new version, called AppleSoft BASIC was released in November 1977 and improved and rereleased the following year.

Micro-Soft was very important to Apple in its early days. The Apple II’s bus architecture made expansion possible, and Micro-Soft came up with Softcard to allow the Apple computer to run CP/M. But it was BASIC that was crucial to the success of the Apple II, and Steve Jobs later encouraged Microsoft to create a version for the Macintosh. Unfortunately, this project was a disaster, so Gates strong-armed Jobs to accept it or lose their license for the Apple II’s BASIC. Since Apple Computers was still highly dependent on the sales of the Apple II and BASIC was a strong complementary component of the microcomputer, Jobs killed their in-house MacBasic project to use the Microsoft version. At least until the Apple II computer was discontinued. [7]

In 1978 Gates agreed to move the company back to their hometown Seattle and change the name to Microsoft. Allen, in particular, had grown tired of the desert and convinced Gates to move the company. The three years they were in Albuquerque years were very challenging as the small company (16 employees in 1978) worked day and night keeping up with the new computers and chips coming on to the market. CP/M was becoming the most popular cross-platform operating system due to Kildall’s BIOS that allowed the OS to be quickly adapted for new machines.

In December of 1979, they moved to Bellevue, Seattle, and started focusing on Intel-based 8086 16-bit machines. By March 1979, the newly named Microsoft had 48 OEM (Original Equipment Manufacturer) customers for its 8080 BASIC programming language, 29 for FORTRAN, and 12 for COBOL. During that summer, they developed PASCAL and APL languages as well. The Intel-based microcomputer industry was starting to take off, and Gates and Allen had positioned themselves in the center of it.

Notes

[1] Fortune 500 dream story told by Paul Allen in an interview on Robert X. Cringely’s Triumph of the Nerds.
[2] This information on Allen and Gates’ early years were from Laura Rich’s The Accidental Zillionaire. John Wiley & Sons, Inc.
[3] Allen and Gates became an employee of TRW during his senior year earning a salary of some $30,000.
[4] Traf-O-Data and TRW employment information from Fire in the Valley, pp. 30-32.
[5] Micro-Soft’s 1977 revenues from Laura Rich’s The Accidental Zillionaire.
[6] Licensing of Microsoft’s BASIC to Apple from apple2history.com. Accessed September 19, 2003. .
[7] Microsoft’s early reliance on Apple from Paul E. Ceruzzi’s A History of Modern Computing. Second Edition. Cambridge, MA: MIT Press. p. 265.

Share

© ALL RIGHTS RESERVED

Anthony J. Pennings, PhD is the Professor of Global Media at Hannam University in South Korea. Previously, he taught at St. Edwards University in Austin, Texas and was on the faculty of New York University from 2002-2012. He also taught at Victoria University in Wellington, New Zealand and was a Fellow at the East-West Center in Hawaii in the 1990s.

Category: how IT came to rule the world, technologies of meaning
Tags: Altair > BASIC > Bill Gates > Intel 8008 > Intel 8080 > Micro-Soft > Paul Allen > PDP-10 > Traf-O-Data

How IT Came to Rule the World, 2.6: The PC and the Floppy Disk

Posted on | September 21, 2010 | No Comments

The development of the floppy disk device was a crucial factor determining the success of the personal computer. Originally developed by David Noble at IBM in 1971 as a backup storage mechanism for their System/360’s magnetic core memories, the floppy disk was soon put to use for other purposes. An employee at IBM, Alan Shugart saw the implications of the new device for smaller computers. He realized that the floppy disk could provide a new storage device that was faster, had random access (meaning that one did not have to rewind an entire tape to find desired information) and could be portable. He started a company called Shugart Associates to build and market them. However, for the microcomputer to be effective it would need to combine the power of the microprocessor with the new storage mechanism. For a microcomputer to use the floppy disk, it required a new software package to run it, what IBM had already called a “Disk Operating System”. [1]

It was Gary Kildall who pioneered the first effective disk operating systems for microcomputers. Kildall earned his doctoral degree through the military’s ROTC program and so had a choice of going to Vietnam or teaching computer science at the Naval Postgraduate School in Monterey California. While teaching, he was also hired as a consultant by Intel to write software for its microprocessors, including a compiler later called PL/M. The compiler would allow programs written in languages like FORTRAN on larger computers to be used with the microprocessor. It was used on Intel’s Intellec-4 and Intellec-8, small computers that could lay some claim to the title of the first microcomputers, but were never marketed as such. Later, Kildall began emulating Intel’s new 8080 microprocessor on an IBM computer (Microsoft founders Gates and Allen would soon do something similar to write software for the Altair computer) and developed a new version of PL/M for it. He also decided at that time to write a program to control the mainframe’s disk drive. Using commands developed by DEC to access data from its DECtape, he began to write the code for the new operating system. DEC’s OS/8 and later its RT-11 had been important developments for PHP minicomputer series and showed that smaller computers could compete with the mainframes.[2] Pulling the pieces together, Kildall created his new operating system called CP/M, short for Control Program/Monitor.

Intel didn’t really want Kildall’s OS, but the software soon became the standard for a number of new microcomputers. CP/M was announced as a commercial product in April 1976. Kildall soon quit teaching to form a new company with his wife called Intergalactic Digital Research (later just called Digital Research) to market the operating system. CP/M was soon used by a number of small computers including the Osbourne, the first portable microcomputer, and the Kaypro which is shown below.

As C/PM became the standard for microcomputer operating systems, it inspired imitation. It was the foundation for another important operating system called Q-DOS (Quick and Dirty Operating System) that was bought by a small company called Microsoft as the basis for its own microcomputer operating system. MS-DOS became the software foundation of Microsoft’s empire and the successful early run of the IBM Personal Computer.[3]

Notes

[1] The beginnings of the floppy disk from Paul E. Ceruzzi’s A History of Modern Computing. Second Edition. Cambridge, MA: MIT Press. p. 236-237.
[2] Paul E. Ceruzzi’s A History of Modern Computing. Second Edition. Cambridge, MA: MIT Press was the source for Kildall’s development of CP/M from DEC sources. p. 238.
[3] Gary Kildall’s early story also from Robert X. Cringely’s Accidental Empires. New York: HarperBusiness, A division of HarperCollins Publishers. pp. 55-59.

Share

© ALL RIGHTS RESERVED

Anthony J. Pennings, PhD is the Professor of Global Media at Hannam University in South Korea. Previously, he taught at St. Edwards University in Austin, Texas and was on the faculty of New York University from 2002-2012. He also taught at Victoria University in Wellington, New Zealand and was a Fellow at the East-West Center in Hawaii in the 1990s.

Category: how IT came to rule the world, technologies of meaning
Tags: Alan Shugart > CP/M > DOS > Floppy Disk > Gary Kildall > IBM PC > MS-DOS

How IT Came to Rule the World, 2.5: Intel and the PC

Posted on | September 14, 2010 | No Comments

The computers used for the Moon landing were already out of date when Neil Armstrong walked on the lunar surface in 1969, but the decision to use integrated circuits or “chips” by the Apollo project paved the way for the microprocessor revolution and one of its main offspring, the personal computer.

NASA decided early on to standardize its Apollo flight technology with the integrated circuits (ICs) and nurtured them into reliable, relatively high performance digital processors. Reliability, cost, and the ease of manufacturing these “chips” had been sufficiently subsidized by the space program (and the MAD “Mutually Assured Destruction” missile defense strategy) to the point where integrated circuits and their next stage, the microprocessor, could be used in business related computers. In what would become a common geek term, the PC was the “killer app” for the microchip.

While Jack Kilby at Texas Instruments is generally credited to be the first to construct an integrated circuit, his contemporaries at Fairchild conceived of a production process that could mass-produce the small chips.

While Kilby’s ICs required its combination of transistors, resistors, and capacitors to be connected with gold wires and assembled manually, Robert Noyce and others at Fairchild were developing a literal printing process to construct the ICs. His “planar process” printed thin metal lines on top of an insulating silicon oxide layer that could connect the integral components of the chip. At first they could only connect a few components, but as they refined their “photolithography” method, hundreds, then thousands of connections could be made. By the time the Internet became a household word in the 1990s, millions of transistors were placed on a single chip.

In 1968, with the Apollo project well-established, integrated chip co-inventor Robert Noyce and fellow Fairchild “traitor” Gordon Moore left the company to form a new semiconductor company. What emerged was Silicon Valley stalwart Intel, the future producer of some of the industry’s most sophisticated microprocessors and the latter half of the infamous “Wintel” combination (Windows OS/Intel microprocessor) that would dominate PC sales throughout the rest of the century.

After twenty years of government backing, the microprocessing industry was about to crawl out on its own. And it was the microcomputer that would give the semiconductor industry the legs to become viable in the commercial arena.

Share

© ALL RIGHTS RESERVED

---

Anthony J. Pennings, PhD is Professor and Associate Chair of the Department of Technology and Society, State University of New York, Korea. Before joining SUNY, he taught at Hannam University in South Korea and from 2002-2012 was on the faculty of New York University. Previously, he taught at St. Edwards University in Austin, Texas, Marist College in New York, and Victoria University in New Zealand. He has also spent time as a Fellow at the East-West Center in Honolulu, Hawaii.

Category: how IT came to rule the world, multimedia, the cold war
Tags: Fairchild > Gordon Moore > integrated circuits > microprocessor > Moon > Mutually Assured Destruction > NASA > semiconductor > transistors > Wintel

« go back — keep looking »