Platform Wars and the First Personal Computer

by Thomas More (

Copyright 1996 Thomas More. All rights reserved


You know, I was thinking about Mr. Kraig's letter to the editor last issue, objecting to my poking fun at the Apple Corporation. I responded that I try to find the funny side of things (even Mr. Kraig), instead of taking them too seriously, but I think there is a serious side to this discussion. Can you stand it if I get all serious on you for once?

Well, I've noticed before that Macintosh Users... (:^) seem to be kinda, well, thin-skinned. I can even understand why. They know they've had a superior PC for the past ten years, and that it took that long for Microsoft to apply the PARC GUI to a real multi-tasking operating system for the Intel microprocessor that was even close to matching the Mac in terms of being "user friendly," which means, in computer-speak, that it's easy to use, or at least easy to learn. During those same ten years, the installed DOS and Windows base increased more than ten times the rate of the Macintosh.

Don't you love it when I talk technical? So what's this got to do with surfing the World Wide Web, you ask? Everything. Soon you'll be getting the Web on the same cable on which you get your phone, TV, newspaper, junk mail (yeah, it'll be there), etc. But you knew that, didn't ya? Getting back to the Mac, though: the problem is, the corporate world went for IBM in the 1980s (they thought it was a new typewriter, so they got one for every secretary), and the Macintosh, the superior PC, was relegated to the art department of large corporations, at best.

And now the last of the aerospace engineers who had held on to their Macs for the past ten years are being persuaded to give them up, in order to be on the office intranet with all the neat Microsoft Office stuff. Some Apple defenders claim that Microsoft placed a "mole" in the Pentagon to bring this about. Whether that's true or not, it's obvious that there is no justice, even in PCs.

But don't despair, Mac lovers. Did you know that Apple Corporation, the mother of the Mac, last year was bigger (in sales) than Microsoft? Maybe the pendulum swings both ways. Maybe Apple will step up with the next generation of PCs, perhaps one you can carry around in your pocket so that, with a wireless modem and a display that you wear like a pair of glasses, you can sit under a tree in the woods and access your office or the World Wide Web.

Personally, I'd love a new Macintosh. Do you think they'd give me one if I wrote a nice poem about it, or something? I have a Pentium 120 running Win95 (which is finally on a par with the Mac), but I put it together out of spare parts, for less than $1,000. I don't think I can buy an equivalent Macintosh for that kind of money, can I? What is the reason for this disparity? It's something called open architecture, which will be explained below.

Okay, if you're read this far, you must be really interested in the future of PCs, so here's an explanation of the terms above. A PARC GUI is a Palo Alto Research Center (PARC) Graphical User Interface (GUI), that was developed by Xerox Corporation (owners of PARC). GUI means the icon-based "point-and-click" concept, to be used instead of typing cryptic gobbledygook to have the computer open a file or execute a program. The Xerox marketing people didn't know what to do with this technology (strange as that sounds now), but Steven Jobs did. That, ladies and gentlemen, is the story of the Macintosh, in a few words. It took the typewriter people quite a while to catch up, by adding a PARC-type GUI to the Intel-based PCs. What now, Apple Corps?

If you are interested in where PCs are going in the beginning of the next century (coming right up, folks), you really need to know where they've been. Like some people, they have a rather shady past; not in the "criminal" sense, but as in "origin unknown." For example, no one knows for sure who first used the term "personal computer." There are pundits who would tell you that they invented the term, but we're going to look way back, with the help of a friend who has been messing with PCs longer than most of us. His name is Dale Thorn, and he deserves a great deal of credit for doing the research for this article, and for being involved in the marketing of scientific and personal computers during their early development, way before the term "personal computer" was used in the popular press. Like Xerox and IBM, Hewlett-Packard never envisioned a consumer market for PCs, although they used the term "personal computer" in their product literature before it ever appeared in the popular press. Their 1960s desktop computers were classified as "scientific," or technical workstations. The shady past of PCs goes back a couple of generations, in terms of the application of chip technology. The ancestors of PCs also include calculators and game machines, as well as the more sophisticated "scientific" computers.

So what happened during the gestation of the PC as a result of the "calculator wars" of the early 1970s? Well, Texas Instruments dropped the price of hand-held calculators so much that it nearly bankrupted a smaller competitor, Commodore, and seriously affected the direction of the largest player in that market, Hewlett-Packard. TI cast another die later in the 1970s, with its TI-99 "home computer," which was originally designed as a game machine, the BASIC programming capability being added as an afterthought, perhaps for appeal to a few computer-hobbyists.

Apparently the TI-99 was directly responsible for the development of the Commodore 64, since Jack Tramiel vowed to drive TI out of the PC market the same way that TI drove him out of the calculator market, a few years earlier. And he did, forcing the price of the TI-99 from over $1,000 to $150, and finally, when production ceased, to a sell off at $50 each. Many of us who had one of those cute little machines bought two or three at that price, just for spare parts!

About the same time, Steve Wozniac quit his job at Hewlett-Packard, to join Steven Jobs and form Apple Computer Corporation. By 1980, IBM decided to catch up and produced its first PC, using an off-the-shelf Intel microchip, industry-standard architecture (the layout or wiring of the "bus," which routes signals in and out of the microprocessor, and to "ports," which are connectors to devices outside the computer, such as a printer, or a cathode-ray tube (CRT), similar to a television. In fact, the first IBM PCs, like the TI-99 and Commodore 64, used a standard television set for the computer display. By the way, IBM anticipated a big market for their first PC. They planned to sell 250,000 of these machines over two to five years. They had no idea they would sell 250 million of them, so many that it would greatly affect their core business of large mainframe computers. Never underestimate the power of personal computing. It's changing the world, you know.

That, in a nutshell, is the story of events leading up to the "platform wars" between users of Apple and IBM PCs. However, there are many more details to this story, starting with a definition of exactly what a PC is. There is quite a lot of history of the development of PCs before Apple or IBM started making them. You will see that both the major combatants in today's platform wars were really latecomers in the birth of PCs. Because there is a lot of information in Mr. Thorn's book, "An Alternative History of Personal Computing," each of the following condensed subjects can be examined separately, at your pleasure. If you are as fascinated with this subject as I was, you might want to visit each of these subjects, maybe more than once. Step into Mr. Thorn's time machine, and let's see where PCs really came from!

Part I: The First PC

Definition of the Term "Personal Computer"

The definition of a PC is somewhat controversial from a historical perspective, although this term seems clear today, given millions of IBM clones running Microsoft Windows in one form or another. I've decided that there are essentially three factors determining whether a computer can be seen as a personal computer:

Candidates for "The First PC"

Following are two candidates for the world's first-ever personal desktop computer.

The first was the Hewlett-Packard Model 9116 circa 1966, which an individual could buy and take home, plug it in and turn it on, and then write or run software based on an enhanced version of FORTRAN (a high-level programming language suitable for general-purpose computer applications).

The second was the Altair circa 1974, which an individual could buy and take home (I don't have a clear fix on the comparative differences in setting these two computers up), plug in and turn on, and then...I don't know what you could actually do with the Altair.

Note: The two candidates above ignore several HP personal desktop computers made between 1966 and 1974. I selected these two, not as the first PCs in the absolute sense, but as the first in each of two categories, which are explained below.

Pros and Cons of the First PCs

Based on my experience with various electronic calculators and computers from the early 1970s to today, there are some arguments both pro and con in designating either of these two machines as being the first- ever personal desktop computer. However, this easily disputes the oft-heard claim that the Altair was the very first PC.


Pro: Not a kit; ready to go after modest set-up.
Con: Expensive. $20,000 or higher. In 1966 dollars this meant you had to be a doctor or have equivalent means to buy this computer. Skipping a few meals would not be enough to make the difference if your income was average.

Pro: Productivity tools included with the computer enable short-term software generation plus a viable upgrade path.


Con: Required considerable time and electronics hobbyist-type knowledge for a usable installation.
Pro: Not cheap in 1974 dollars, but still a fraction of the cost of an HP-9116.
Con: Lack of productivity tools implies a hobbyist item, but a new Basic interpreter gets the ball rolling.

Other Candidates for "The First PC"

Given the above rather serious constraints (for personal computing per se), we might be inclined to pick other time periods and computers as representing the "first real breakthrough" for truly personal computing, for most average users. If you subscribe to mass-marketing logic, as it pertains to an actual consumer product, you might point to the Macintosh computer which, from the perspective of its user interface, is certainly a breakthrough; however, one could just as easily point back to the HP 80 series computers (easiest to program and learn), or the HP Touchscreen DOS computers of late 1983.

First Use of the Term "Personal Computer"

"Personal Computer" as an industry buzzword has been credited (in popular journals) to an article in Rolling Stone in 1974, but the term was used before that in HP product literature, and in an HP Journal article in early 1974 to describe a hand-held computer without I/O ports, but with a set of applications packs, and a magnetic card reader with which to install them.

Hand-held Computers

Hand-held (or palmtop) computers have had an interesting history outside the realm of desktop machines, and have served as viable substitutes for larger computers (considering limitations of screen and keyboard size, clock speed, disk storage, and so on) for only a short time now, albeit at a prohibitive price for the average person. For example, you can obtain a pocket-size PC with MS-DOS version 5.00 in ROM for about $400, but adding a "flash"-memory card with 165 Mb of storage space will raise the price tag by another $2300.

HP Makes First Actual Desktop Stand-alone "Personal Computer"

Uses That Term in Product Literature

My inclination is to credit the origin of personal computing to HP products, primarily because they actually did come about first, but also because they were ready-to-operate, with applications software included, as well as with special tools to enhance the existing software or create new software. The following paragraph applies mostly to the so-called "technical" workstation computers, but since they necessarily preceded the non- "technical" machines, their relevance to what came after them cannot be overlooked.

HP Operating System and Programming Language

HP Workstation Basic languages were not only full-featured and powerful, with instruction sets not matched by the most sophisticated "C" compilers of today, they were also the operating system of the computer, so that the "DOS command line" was not just DOS, but a very complete language as well. One reason why the Basic languages offered such good performance was the fact that they were not interpreters in the traditional sense; rather, they employed threaded and tokenized coding much like the Microsoft QuickBasic 4 and 7 editors, and they utilized string allocation techniques which provided the best features of the interpreters (ad hoc string allocation) along with those used in "C"-language compilers (preset length where current length is determined in an ad hoc way).

Two Classes of PCs

At this point, you might suspect that the issue of what's a PC is either not at all clear-cut (at least from a historical point of view), or that this article is prompting an "apples and oranges" discussion, when there are really two (or more) historical classes of PC: the machine for casual uses, (word processing, spreadsheets, and recreational purposes), and the machine for scientific, engineering, and other technical applications.

Overlap Between Classes

This overlapping of computers and applications did not mean you could stroll into a retail store, buy a $1000 system, then obtain some "shareware" software for scientific computing, and thereby get a viable workstation for technical applications. By the same token, you could not expect to find the best word processors or database programs running on the so-called technical computers back then, although today you can get nearly anything you want for the mass-market machines, whether Macintosh, DOS/Windows, or UNIX workstations.

Pulling the Plug on Retail Stores

This two-classes viewpoint was actually a hallmark of PC sales in the early 1980s, and the issue was resolved to some extent when the major manufacturers (HP and others) pulled their "scientific, engineering, and technical workstation" computers out of the retail channel permanently. Needless to say, there was some overlap between the two classes even back then, so if you used a "technical" computer, you could find a spreadsheet or word processor for it, and if you had a non-technical computer, you could get technical software for that computer, if not from a retail store, then probably through a user group for a user- maintained shareware library.

Part II: The Apple

Large Base of Software

The Apple II had a profound significance in the field of PCs, in spite of the fact that other brands and models having similar capability were available at that time (circa 1980) through major retailers, including Radio Shack and their own extensive product line. The deciding factor that made the Apple II the biggest thing going was undoubtedly the huge software base that built up around the machine. It was not uncommon to see a person walk out of a computer store with a cart full of diskettes for the Apple II, including games, business software, math/CAD applications, and just about anything else you can imagine.

Selling "An Apple a Day"

A typical retail store chain selling the Apple II from early 1981 to late 1982 moved (as we liked to say then) "an Apple a day" per store, usually with diskette drives, monitors, and other items that would be standard on today's PCs, but were extra back then. Selling an Apple per day per store brought in enough revenue to pay all the bills, and everything else that went out the door was gravy.

End of the Apple II

The downside of the Apple II was ultimately Apple itself, due no doubt to their commitment to the mouse- and-icon GUI (graphic user interface) in its first mass-market success on the Macintosh. Other factors contributing to the Apple II's demise included a lack of industrial-grade floppy diskette drives combined with the lack of a practical hard disk option (at that time), not to mention a Basic language which would allow insertion of data into existing strings, to preclude "garbage collection," a serious problem on the II.

Macintosh vs. Microsoft

With the introduction of the Macintosh, Apple stood more or less alone for several years in GUI-based PCs, not seriously challenged until the market was flooded with PC clones which contained pre-installed Microsoft Windows. The fact that the IBM PC and its clones continued to outsell the Macintosh from 1984 through today (by nearly ten to one) demonstrated to the PC world that, contrary to conventional wisdom, the user interface was not the key to widespread acceptance of PCs in places other than corporate offices. In fact, the momentum to place PCs in people's homes and other casual-use environments was well under way even before the aforementioned flooding of the market with Windows-based clones took place.

New Applications Drive PC Market Expansion

In my view, each successive level of PC market expansion is fomented by the introduction of a significant new application, the latest examples of which are communication programs for accessing the Internet through the World Wide Web, as well as satellite-based wireless paging, fax, and email services. It's important to note here that it might not necessarily be the availability of new hardware and/or software per se that provides the impetus for expansion of the PC market; rather, it's the consumer awareness of these new services in concert with the hardware and software which are evolving simultaneously to facilitate the use of those services.

Part III: Big Blue Catchup (IBM PCs)

Entrepreneurs Build Huge New Software Base

The last great (and relatively sudden) transformation of the PC market was the advent of the IBM PC in late 1981. Tens of thousands of entrepreneurs got word of the new computer-to-be, and they quickly began developing what would eventually become the largest collection of interchangeable software in the world, dwarfing the already-large quantity of Apple II applications, in number as well as in capability and performance.

Importance of Open Architecture

Whatever IBM originally intended, the openness of the PC architecture made it possible to write software in 1981 that (with some exceptions) operates satisfactorily on the most up-to-date Pentium-based PCs of today. Knowing how to get around incompatibilities has become something of an art form in the PC world, especially for Windows 95 systems, but encountering glitches here and there is a far more agreeable situation than having to convert to IBM from the Apple II, a frequent problem for computer users in the 1980s.

Part IV: Portable PCs of the Future

Desktop vs. Portable Computing

Arguments concerning the merits of desktop computing vs. portable computing have been propounded at length for many years, nearly as many years as the smaller machines have been around. What's becoming more obvious now is that people are seeing a need to communicate from wherever they happen to be, and that communication is quickly evolving from voice-only telephone connections, or faxing from local businesses which provide these services, to individuals' being able to manage their business communications from wherever they happen to be when they feel like doing so.

Business Communications and/or Information Access

It's easy to visualize portable computing and communications as primarily for business or technical work, but the main factors inhibiting persons who don't carry on such business from enjoying the convenience of on-the-spot access to information that would be useful to them are the lack of seamless integration of hardware with software, the costs associated with such equipment, and most importantly, the absence of cooperation between vendors whose benefit derives mainly from having proprietary products.

Personal Productivity With a Tiny PC

My first PC was a programmable pocket-sized machine, with 2 Kb of RAM memory, which I purchased in 1975. Using that computer, I was able to reduce the time spent on my normal job functions by half, and use my newly acquired free time to analyze our inventory and shipping methods. I reduced our on-hand inventory over a few months to two thirds of its normal size, then, while holding inventory to a minimum, I decreased back orders from an average of 100 items or so to ten or less at any given time. One could suggest that this is not personal computing per se; however, when I left that position, I took my computer with me, and it continued to enhance my productivity.

The Radio Shack Model 100 Portable

Late in the 1970s, Radio Shack introduced their legendary model 100 with its 40-column-width screen. A very small and eminently portable computer, it remained popular for a number of years; nonetheless, I wondered at one point why so many people would use a 40-column computer in lieu of an 80-column-wide laptop computer, once those became widely available. Finally, it occurred to me that people who prepare text for newspapers and various other periodicals almost never output their text in formats wider than 40 columns, hence, a perfect fit! Eventually, this and other CMOS computers in the laptop-size class lost out to laptop computers having conventional memory and a built-in hard disk drive. CMOS computers do have advantages over conventional laptop machines, however, operating as long as 25 hours from a single battery charge, if there are no moving parts and the screen is not back-lit. Laptops usually have maximum operating times of less than five hours, and in the case of the later Pentium portables with their new Lithium- Ion rechargeable batteries, on-time may be as little as two hours.

HP Makes the First Plug-and-play PC

When Apple introduced the Macintosh computer in January of 1984, Hewlett-Packard brought out a new machine as well: a pocket-size model with their own proprietary operating system, with up to 400 Kb of free RAM memory at a ridiculous price ($4,800 per Mb), and with a single I/O interface which could support over 900 peripherals simultaneously, including the industry standard RS-232, IEEE-488, and so on. This was the first computer of any kind that could actually serve in the ideal (well, almost) role as both a desktop machine, with large keyboard, monitor, printer, etc., plugged into its multi-peripheral I/O port, and as a pocket portable by simply pulling the plug and popping it into your jacket pocket. There are valid reasons why this particular device faded away, but I've always felt instinctively that this kind of convenience, in a more modern configuration (MS-DOS and a 386 processor, for example), would become the most sought-after product of all time, at least as far as personal electronic items are concerned.

The Ubiquitous PC of the Future

Fast-forward to 1996, and you see a few more people using laptop computers in planes and airports than what you would have seen a dozen years earlier, but you'll more likely see people using palmtops, particularly those which are termed "organizers," having Windows-like capability built in, and with linking software included for direct connection to desktop computers. The organizer-class palmtop computers are frequently sold with internal modems, and if they have cellular capability included, then the sky (pun intended) is truly the limit. As time goes on, these miniature computers will enjoy significant improvements in screen legibility, keyboard and screen-pointer operations, memory and disk space, processor speed, and compatibility with desktop software. As these things take place (and when prices become more in line with other portable consumer products such as cameras, stereo tape and CD players with headphones, cellular phones and so on), we will witness a time when more data is processed daily on personal pocket-size computers than all other types of computers, simply because of the greater number of people using them. That is the beauty of personal computing.

During the week, Thomas More writes rather technical stuff about business software, but on weekends he likes to loosen up a bit and write about WWWiz stuff.