It is easy to survey the realm of computing and see that Windows is dominant (at least on desktop computers, and on lots of laptops in offices). But Windows did not always have dominance; it had to fight its way to the top. Windows had to replace PC-DOS/MS-DOS, it had to fight off OS/2, and it had to beat a number of other (smaller) contenders.
Much has been written about the transition from PC-DOS to Windows and the competition between Windows and OS/2. There is one factor, I think, that has received little attention. This one factor, by itself, may not have made the decision, but it was a factor that favored Windows.
That factor was the laser printer. (Specifically the Hewlett-Packard LaserJet printer.)
Laser printers were desired. They were expensive, which dampened their acceptance, but people wanted them. They were quieter, they were faster, and they produced better quality output. They could provide different typefaces and they could print graphics. And people wanted quiet, fast, high-quality output, especially with graphics.
One could use a laser printer with programs in PC-DOS. It was not always easy, and it was not always possible. PC-DOS provided few services for devices; just enough to send data to a parallel port or a serial port. Applications that wanted to use sophisticated devices (such as laser printers) had to build their own drivers. (The same issue was present for video cards, too.) Thus, when purchasing software, the first question was "Will it support a laser printer?". Some software did, some did not, and some supported laser printers poorly.
Windows supported graphics, video cards, and laser printers from the start. Windows was built around graphics; the first release of Windows was a graphics program. Windows also handled device drivers, allowing a device to have a single driver for all applications. If a program ran in Windows, it could print on all of the printers supported by Windows. Windows was graphics.
In contrast, the first version of OS/2 worked only in text mode. OS/2 users had to wait for version 2.1 to have graphics. Microsoft and IBM (developing OS/2 jointly) focused on multitasking and memory and security.
The difference between Windows and OS/2 was that orientation. Windows was an operating system for a PC; that is, an operating system for a video display board that had a processor and memory and storage attached. OS/2 was an operating system for a minicomputer; very good at multitasking for a use that communicated through a character interface. Even though PCs at the time had video boards, OS/2 pretended that the user was sitting at a terminal.
But people wanted graphics. They wanted graphics because they could see the print-outs from laser printers. They were willing to pay lots of money for laser printers, to impress their co-workers and their bosses.
Windows had graphics. OS/2 did not.
I cannot help but think that laser printers helped Windows win over OS/2.
(I do recognize that other factors contributed to the success of Windows. Those factors include licensing arrangements, marketing, and compatibility with PC-DOS applications. I think laser printers are another -- unrecognized -- factor.)
Today, we casually accept that just about every device works with Windows, and that we can print from any application to any device (laser printer, ink-jet printer, and even PDF file), and that it all works. The computing world of 2020 is very different from the world of 1985.
But maybe we should be looking forward instead of backward. Windows won over OS/2 because it met the demand of the market. It provided graphics on screen and on printouts. It gave people what they wanted.
Today, in 2020, what do people want? And which companies are providing it?
Showing posts with label market pressure. Show all posts
Showing posts with label market pressure. Show all posts
Wednesday, February 5, 2020
Tuesday, February 23, 2016
Why the IBM PC became a standard
The history of the PC is simple: After several companies, including Apple, Radio Shack, and Commodore, built and successfully sold microcomputers, IBM entered the market and its products were adopted as the standard. Yet I think there is more to it than that.
The IBM PC was a popular computer. Some historians claim that its open architecture and its superior design lead to its success. The architecture was open, more open than some but not all competing products. The design was good, but not superior -- and we are still paying for some of the compromises IBM made in those early days.
It was more than openness and design that made the IBM PC the standard. There were three other factors.
The first was a set of compelling applications. The "killer application" was the spreadsheet. Considered mundane today, the spreadsheet was a large step up from the manual methods of calculation that preceded the PC. It was far better than doing math by hand or even with the assistance of a mechanical or electronic calculator. Spreadsheets provided more automation, allowing for sophisticated formulas.
The other compelling (although perhaps not a "killer application") was word processing. Personal computers could be purchased for slightly more that electric typewriters, and word processors gave us the ability to compose, store, retrieve, and revise documents.
The second factor was a sense of urgency. Managers may not have understood personal computers or foreseen how they would change businesses, but they knew (or believed) that computers were the way of the future and they did not want to be left behind. Perhaps the fear was a remnant of the 1960s space race (with its jet motors, robots, and computers) or perhaps it was caused by the hype of IBM's advertising. Whatever the reason, there was a fear of becoming obsolete, of missing the computing boat. That fear pushed companies to adopt computers.
These two factors were enough to drive the PC revolution, but they were not enough to force the IBM PC as a standard. There were many competing systems, several which ran MS-DOS and Lotus 1-2-3, just like the IBM PC. There was a third factor, one that moved businesses into the IBM fold. That factor was the cost of PCs (IBM or otherwise).
Personal computers in the early 1980s were expensive devices. The price for a typical business system was in the $3000 to $5000 range. (And that was in 1980 dollars!)
The expense of a microcomputer meant that business had to think carefully about their investment. A wrong choice would mean a significant loss of capital. Businesses wanted to avoid that loss, so they went with a safe choice: IBM.
Those three factors (compelling applications, a sense of urgency, and risk avoidance) led to the IBM PC as a standard.
* * * * *
Looking at today's market, there is no analogous situation. Personal computers are inexpensive, standardized boxes. But what about new technologies?
Phones and tablets are inexpensive, with no compelling application for businesses. There may be a sense of urgency to build mobile apps for customers, but note that customer apps are outward-facing, not inward-facing like the spreadsheets and word processors of the PC revolution. Equipment for inward-facing applications can be standardized, as the equipment is under your control. Outward-facing apps cannot be easily standardized, as customers choose their hardware. That's a big difference between the PC revolution and the mobile revolution. There is not sufficient force to select one design as the standard -- and we have competing designs, Apple, Android, and Microsoft.
Cloud systems inward-facing, but not compelling. There is no sense of urgency to move one's systems to the cloud. Cloud systems are not expensive either -- at least not in the way that PCs were expensive. If anything, cloud providers focus on the ability to "pay for what you use". It's no surprise that there is no standard cloud system.
The same holds true for Big Data: inward-facing, but there is no compelling application, no sense of urgency. Sure enough, there are multiple "standards" for Big Data.
* * * * *
We don't have a standard for a smart phone, or tablet, or cloud system, or big data. Without the three factors of compelling applications, urgency, and risk avoidance, I think we will see multiple solutions for some time.
The IBM PC was a popular computer. Some historians claim that its open architecture and its superior design lead to its success. The architecture was open, more open than some but not all competing products. The design was good, but not superior -- and we are still paying for some of the compromises IBM made in those early days.
It was more than openness and design that made the IBM PC the standard. There were three other factors.
The first was a set of compelling applications. The "killer application" was the spreadsheet. Considered mundane today, the spreadsheet was a large step up from the manual methods of calculation that preceded the PC. It was far better than doing math by hand or even with the assistance of a mechanical or electronic calculator. Spreadsheets provided more automation, allowing for sophisticated formulas.
The other compelling (although perhaps not a "killer application") was word processing. Personal computers could be purchased for slightly more that electric typewriters, and word processors gave us the ability to compose, store, retrieve, and revise documents.
The second factor was a sense of urgency. Managers may not have understood personal computers or foreseen how they would change businesses, but they knew (or believed) that computers were the way of the future and they did not want to be left behind. Perhaps the fear was a remnant of the 1960s space race (with its jet motors, robots, and computers) or perhaps it was caused by the hype of IBM's advertising. Whatever the reason, there was a fear of becoming obsolete, of missing the computing boat. That fear pushed companies to adopt computers.
These two factors were enough to drive the PC revolution, but they were not enough to force the IBM PC as a standard. There were many competing systems, several which ran MS-DOS and Lotus 1-2-3, just like the IBM PC. There was a third factor, one that moved businesses into the IBM fold. That factor was the cost of PCs (IBM or otherwise).
Personal computers in the early 1980s were expensive devices. The price for a typical business system was in the $3000 to $5000 range. (And that was in 1980 dollars!)
The expense of a microcomputer meant that business had to think carefully about their investment. A wrong choice would mean a significant loss of capital. Businesses wanted to avoid that loss, so they went with a safe choice: IBM.
Those three factors (compelling applications, a sense of urgency, and risk avoidance) led to the IBM PC as a standard.
* * * * *
Looking at today's market, there is no analogous situation. Personal computers are inexpensive, standardized boxes. But what about new technologies?
Phones and tablets are inexpensive, with no compelling application for businesses. There may be a sense of urgency to build mobile apps for customers, but note that customer apps are outward-facing, not inward-facing like the spreadsheets and word processors of the PC revolution. Equipment for inward-facing applications can be standardized, as the equipment is under your control. Outward-facing apps cannot be easily standardized, as customers choose their hardware. That's a big difference between the PC revolution and the mobile revolution. There is not sufficient force to select one design as the standard -- and we have competing designs, Apple, Android, and Microsoft.
Cloud systems inward-facing, but not compelling. There is no sense of urgency to move one's systems to the cloud. Cloud systems are not expensive either -- at least not in the way that PCs were expensive. If anything, cloud providers focus on the ability to "pay for what you use". It's no surprise that there is no standard cloud system.
The same holds true for Big Data: inward-facing, but there is no compelling application, no sense of urgency. Sure enough, there are multiple "standards" for Big Data.
* * * * *
We don't have a standard for a smart phone, or tablet, or cloud system, or big data. Without the three factors of compelling applications, urgency, and risk avoidance, I think we will see multiple solutions for some time.
Subscribe to:
Posts (Atom)