IBM Turns 100, and Creates a Stir

The IBM Personal Computer

Journalists across the Web (mostly) celebrated the 100th birthday of IBM last week, on June 16th. See for instance, coverage at The New York Times, Wired, and Forbes. My history of computing colleagues at the IT History blog also covered the story, with a business history perspective from Joel West. As a former IBMer, I can't help but feel a small twinge of pride at this milestone. IBM's own PR, however, has caused a bit of a stir among the technorati, due to a claim by IBM VP Bernie Meyerson that IBM invented the personal computer. The real inventor, of course, was Ed Roberts, creator of the Altair 8800, as Robert Cringley indignantly pointed out, following up with a zinger:

Among his milestones IBM’s VP of Innovation completely forgets to mention the company having helped automate the Third Reich.

Or was he? Of course, the story is much more complex than this, as Matthew Lasar described in a story at Ars Technica, citing Paul Ceruzzi. There were Intel development systems and hobby machines that pre-dated the Altair (though none achieved the same level of commercial success), not to mention the whole legacy of time-sharing. Interestingly, however, many commenters on the Ars story ignore Lasar's complex analysis to provide their own straightforward answer. Some defend the IBM line, stating, in effect, that the IBM PC was the first personal computer that mattered. Steve Jobs and the Apple II is another popular answer, as is, of course, Roberts. One commenter wrote,

Whenever I think of a PC, it's got to have the GUI. I can't think of anything that defines a personal computer more.

That might make the Xerox Alto, the Xerox Star, the Apple Lisa, or the Apple Macintosh the first PC, depending on one's other criteria. This takes me back to my very first blog post, about the question of who invented the computer. The more I think about these kind of priority disputes, the more I think they could make a very interesting subject of a cultural history--aside from the question of patents and pecuniary interests, how do people get invested in particular outcomes of these disputes? It obviously matters to Cringely that Roberts invented the PC on a level beyond setting the facts straight--why?


Dag Spicer at CHM did a good job of tracking down the earliest machines to be sold for personal use, which I believe used to be available from a column he wrote for Dr. Dobbs some years ago. IIRC he moved back past various obscure machines before settled on the celebrated Honeywell Kitchen Computer, but I may be conflating several of his posts. So that's one way to tackle it: the first personal computer is the first machine marketed for personal use.

I'd answer it another way: the first personal computer is the first machine used predominantly in an interactive way by individual users for extended periods without clear organizational goals. So personal computing is a mode of use, not a kind of machine. The first candidate would seem to the the TX0 given to MIT and used by the famous Hackers from about 1958 onward. See Steven Levy's wonderful book Hackers for more details. In this context personal computing is pioneered with timesharing and minicomputer installations. Which, Chris, would fit nicely with your own research.

That's an interesting point, Tom. The TX-0 also "links" directly to the Laboratory Instrument Computer (LINC), as Joe November showed in his dissertation. The LINC was certainly in some sense a personal computer, intended for use by an individual researcher.

My tendency is to focus on the genealogical question, which leads me to basically agree with Cringely that the Altair was the original personal computer, in the sense that it is the earliest common ancestor of all modern personal computers. The Sol, Apple, Commodore, TRS-80, etc., all were inspired by the excitement around the Altair, and an unbroken line leads from them through the IBM PC, and so on.

Other early commercial machines (from the Kitchen Computer to the Scelbi-8H) didn't achieve the critical mass to trigger a wider hobbyist market. Nor, as far as I can tell, is there a "genetic" connection from the TX-0 and other personal machines developed in academic and research labs to the microcomputer world, at least until the Alto-inspired Lisa and Macintosh.

We have been accustomed to reviving the discussion on "who invented the PC" every time an anniversary of the launch of some celebrated hardware, such as the PDP-8, Altair 8800, Apple II, or IBM PC is up. Some computer historians who like to live dangerously on the fringes of the modern history of computing would even offer a round of beer in a local pub to commemorate little known events such as the birth of Edmund Berkeley's Simon or the announcement of the Kenback 1. On such "who invented the PC" occasions, the definitions of the PC would be bent and twisted to propel a celebrated hardware to the top of the PC ladder.
Typically and unfortunately, such events do not add anything of value to the PC debate. They are backed by the same old texts (the classics!) and the same old selection and interpretation of events, firmly reinforcing the American-centric view of the birth and the sandbox days of the PC industry. Yes, there was the MITS Altair 8800 and yes, its announcement and manufacturing generated quite a forceful waive of microcomputing activities. But while enthusiastically quoting a few thousand orders received by MITS in 1975 for its Altair 8800 (which, in its 1975 configuration, could do absolutely nothing), what's holding these "who invented the PC" debaters from being cheerful about a different few thousand orders received a year earlier for the French Micrals and the Canadian MCM/70s? And what about many Micrals and MCM/70s already in use by individuals in business, research, and educational institutions while MITS was still collecting the Altair 8800 orders, uncertain about its future? These microcomputers sported useful software, powerful programming languages (such as APL), and could be interfaced with a range of peripherals. What are these "Micrals" and MCM/70s" you ask? -- well that's the problem.
The dawn of the personal computer is one of the most intriguing issues in the modern history of computing. Yet, in spite of the large volume of literature on the subject, there is still no universally accepted definition of the PC and, hence, the list of "fathers" of the PC will continue to grow. In 1986, the Altair 8800's creator Ed Roberts told The Computer Museum Report (vol. 17, 1986) that PCs should be distinguished from "hobby machines, demonstration machines, industrial machines and development systems." Furthermore, it was reported that in Roberts' opinion "The PC had to be used for applications typically run on a minicomputer or a larger computer. The PC also had to be affordable, easily interfaced with other devices ... It should have an operating system and ... a reasonably large memory." Clearly, the Altair 8800 in its 1975 configuration would not pass this definition. As far as my take on the PC's definition is concerned, it can be found in the recently published Inventing the PC: the MCM/70 story.