Intellectual property is an issue of hot debate. In the last several years, ideologies have clashed as voices rise up from Silicon Valley, Shenzhen, IP-dependent developers, and the open source community to weigh in on the value of innovation in a copycat world.
For perspective on the impact that IP can have in the tech industry, we turn to a tale from the 1980s during the PC boom when the relationship between hardware and software was markedly different from today. The legal battles between IBM and Compaq over "clone" computers were formative for attitudes about IP today.
The Compaq Story
During the 70s and 80s, there was a massive boom in the computing market. The leader of the proverbial pack was IBM, otherwise known as big blue. At the time, computer companies couldn't agree on a common architecture, which meant that software that worked on one computer model would not work on another. A classic example is software written for the Amstrad 9512m, which would not boot on the Amstrad 8512.
But one company changed all of that by producing a machine that was compatible with IBM computers. This meant that there was suddenly a choice: you could purchase an IBM PC or you could opt for a cheaper alternative that could run the same software. The company that made this breakthrough in the compatible market was Compaq and it became one of the fastest-growing companies of all time, making it into the Fortune 500 by 1986—in a record-breaking four years.
Compaq was a giant in the IBM clone market until it was acquired by HP.
Compaq was created by three employees of Texas Instruments: Rod Canion, Jim Harris, and Bill Murto. The three were generally unhappy with their employment and so decided to make the jump into running their own business. Interestingly, the goal of their company was not to become a giant in the computer industry but instead to have a company that was fun to work at with a relaxed environment.
When the three founders of Compaq were trying to figure out what their business would involve, they noticed that IBM computers were becoming popular. They also noticed a portable computer, the Osborne 1, which had potential—but was not IBM compatible.
The revolutionary idea for their product was to create the first portable IBM-compatible computer. It was not long before they had planned out the machine and obtained the necessary funds from investors.
The Osborne 1 portable computer was the inspiration for the Compaq Portable. Image courtesy of Bilby By Bilby (Own work) [CC BY 3.0]
But one problem remained with their IBM-compatible machine. How would they recreate the IBM BIOS without breaching the copyright? Why did they need a compatible BIOS (basic input output system)?
Software written for IBM computers is dependent on the BIOS. This software determines how the computer interfaces with hardware and handles the necessary calls that the software may make. You can see a list of BIOS interrupt routines found on IBM computers here.
It does not matter how the BIOS completes the instructions; it only matters that the outcome of the request is what is expected by the IBM PC standard. So, for example, reading a PS/2 keyboard could be done using a bit banged IO line or handled by a dedicated IO chip and then sent to the BIOS—the end result is the same (returning the user with the last key press).
One of these ROM chips holds the IBM BIOS, the key to the entire machine. Image courtesy of Germán Martínez By German [CC BY-SA 3.0]
Some companies, such as Eagle, copied the BIOS either by reading the IBM manual to understand which interrupt codes did what or by simply copying the hard-coded ROM chip that holds the BIOS. This breached the copyright that IBM had over the BIOS and so did not take IBM long to prevent such companies from selling more of those machines.
However, there is a way to bypass these types of copyrights with the use of a “clean room”.
The BIOS Clean Room
If you look at the IBM interrupt codes or the BIOS code, itself, then you can no longer make the BIOS as you will use knowledge gained from IBM content, which is protected by copyright law. If, however, you build a computer, examine software for IBM machines closely to see what they are trying to call, and create your own BIOS around that then you have not breached IBM's BIOS copyright. This is how companies were able to produce IBM-compatible clones such as Compaq and Eagle (who had to stop initial production because of the copyright breach).
Compaq sent Garry Stimac to Dallas to obtain some IBM manuals—but when he got his copy and began to read the book, he was hit with a big problem. IBM had put all the information needed to write an IBM-compatible BIOS in the manual which meant that Garry could no longer write the BIOS code without serious implications and potential legal action. Therefore, a clean room was set up, meaning that the BIOS code was created by others who had not seen the IBM manuals.
Inside this computer lies a small ROM chip that defines the IBM machine. Image courtesy of Wikipedia
After running many different pieces of IBM software and finding out why certain programs crashed, they eventually got a working IBM BIOS clone that was 95% compatible with IBM software (no other company could get close without breaching copyright laws until Phoenix Technologies began selling BIOS chips).
Eventually, they had a working machine and released the Compaq Portable in November of 1982. Compaq anticipated sales of up to $30 million. To their astonishment, they ended up making a total sale of $111 million in the first year. Even though IBM had sold 750,000 machines in the same year, Compaq still sold over 53,000 units.
Compaq's Portable. Image courtesy of OldComputers.net
The Rise of Compaq and IBM's MCA
As the years passed, Compaq computers shot up like a firework and had sales of $503 million by 1985. IBM began to realize that they were losing their market share in the personal computer sector and began to devise plans to eliminate competition. IBM saw major issues when software was running better on Compaq IBM clones than the IBM machines, themselves.
Their first attack came in the form of lawyers using 9,000 IBM copyrights in the hope that Compaq had broken just one. Luckily for IBM, Compaq did break one copyright, which forced them into a corner. Compaq negotiated with IBM and ended up purchasing the copyright for $130 million. This, however, did not stop Compaq overall.
In a desperate attempt, IBM made one of the worst business decisions in computing history—it created a new standard that was not compatible with its own IBM software.
During 1987, there was $12 billion of IBM PC software on the market which was in use by companies and individuals alike. These customers relied on the fact that their software and files could be transferred from one PC to another with no issues. IBM wanted to make all clone computers redundant by developing a new computer architecture called Micro Channel Architecture. This would mean that software written for MCA-based machines would not work on IBM PC machines and IBM PC software would not work on IBM MCA-based machines.
But this also meant punishing the IBM PC users who had purchased software from IBM and other developers, essentially making $12 billion worth of software redundant.
IBM PS2 Model 25. Talk about shooting yourself in the foot.
IBM released the Personal System II which had large sales due to IBM's share of the computer market. This did not go down well with clone manufacturers (as they feared that the AT bus would become redundant) and so nine companies banded together (Compaq, Zenith Data Systems, Tandy, HP, Olivetti, NEC, AST Research, Epson, and WYSE) to form the “Gang of Nine” and create EISA, Extended Industry Standard Architecture.
This was a revolution for the clone market because not only did the extension of the current architecture improve machine capabilities but is also retained backward compatibility with software on the market for the older IBM PC-AT based computers. IBM lost the compatibility battle and eventually used PCI, as well as all other computer producers.
Compaq grew into a massive computer supplier and by 1994 surpassed IBM sales in the personal computer market. By 2002 Compaq was acquired by HP and IBM stopped making new PC machines.
The story of Compaq and the creation of EISA is relevant today because intellectual property has become, if anything, even more contentious over the last decades. In the 70s and 80s, clone machines essentially stole profits from the original designers. But if companies such as Compaq had not produced clone machines, IBM would have eventually had a complete control over the computer market, reducing competition. This could very well have also reduced the likelihood of collectively agreed-upon architecture.
The compatibility war was also crucial for hardware companies because IBM expected licensing fees from producers of AT and MCA hardware, whereas producing hardware for ISA (industry standard architecture) requires no license, a standard followed by many companies. This increased the competition between hardware companies, necessitating the variety of hardware available to those machines.
Hopefully, in the next decade, we will see a new architecture for PCs that could replace the archaic PCI, PCIe, and SATA channels and replace with more modern designs for the ever growing computer market.
You can learn more about Compaq's rise in Silicon Cowboys, a documentary on the subject directed by Jason Cohen. I highly recommend giving it a watch!