The Open Letter to Hobbyists: The Day Software Became Property

Abstract

On February 3, 1976, a 20-year-old Bill Gates published an “Open Letter to Hobbyists” accusing the personal-computer community of theft: “As the majority of hobbyists must be aware, most of you steal your software.” The letter is remembered as the founding document of the argument over whether software is property to be sold or knowledge to be shared, a fault line that still runs through the industry fifty years later. This article examines what Gates actually wrote, why the hobbyists were copying in the first place, the awkward fact that Gates had built the disputed program on a computer he was not supposed to be using for it, and why every attempt since to shame or engineer copying out of existence has failed.

A World Without Software Companies

In 1975 there was, in effect, no software industry for personal computers. Hardware was a product you paid for; software was something that came with hardware, or that you typed in yourself from a magazine listing, or that a friend handed you on paper tape. The idea that a program was a separately owned, separately priced good (that copying it without paying was stealing) was not yet a shared norm. It was a norm that had to be invented, and the invention was contested.

The context was the Altair 8800, the MITS kit computer whose January 1975 Popular Electronics cover launched the personal-computer hobby. The Altair was sold as a bare kit for a few hundred dollars; on its own it did almost nothing. What made it useful was Altair BASIC, the interpreter that Bill Gates and Paul Allen wrote for it, the founding product of the company they had just named Micro-Soft. (For the language itself and its wider reach, see BASIC: The Language That Democratized Programming.)

BASIC was what turned the Altair from a box of blinking lights into a computer you could program in English-like commands. And almost nobody was paying for it.

The Letter

By late 1975 paper tapes of Altair BASIC were circulating freely through the Homebrew Computer Club in Silicon Valley and beyond. The most-repeated origin story: a pre-release tape was copied at an Altair demonstration and reproduced en masse, with club members reportedly making dozens of additional copies that then spawned hundreds more. To the Homebrew culture (steeped in the hacker ethic that “all information should be free”) sharing a program was not theft but the whole point.

Gates disagreed, loudly. On February 3, 1976 he published his “Open Letter to Hobbyists” in the Homebrew Computer Club Newsletter (and, subsequently, in other hobbyist publications). Its central charge was blunt:

“As the majority of hobbyists must be aware, most of you steal your software. Hardware must be paid for, but software is something to share. Who cares if the people who worked on it get paid?”

He backed it with numbers. Gates claimed that less than 10% of all Altair owners had bought BASIC, that the value of computer time spent developing it exceeded $40,000, and that the royalties actually received worked out to a rate that made “the time spent on Altair BASIC worth less than $2 an hour.” His conclusion was a demand: that people who copied software be treated as thieves, and that anyone reselling pirated copies be excluded from the community.

The response was overwhelmingly hostile. Hobbyists argued that software genuinely was different from hardware, that the community which debugged, extended, and evangelized a program deserved a share in it, and that Gates’s commercial framing missed the point of what they were building. But the letter had done its work: it forced the community to have the argument at all. The proprietary-versus-free schism that still divides the software world (Microsoft on one side, the open-source movement on the other) traces its cleanest origin to this single page.

The Man Making the Accusation

There is a large irony buried in the letter, and it is worth stating plainly because Gates did not.

Altair BASIC was not written on an Altair: Micro-Soft did not own one. It was developed on a DEC PDP-10 mainframe running an Intel 8080 simulator, and the PDP-10 Gates used was at Harvard, where he was an undergraduate. That machine (in the Aiken Computation Laboratory) was funded for defense-related research, and using it to build a commercial product for a private company was not what it was for. Harvard opened an inquiry into Gates’s heavy, unauthorized use of the machine.

The inquiry ended without formal discipline: Harvard had no written policy governing personal use of the machine (the professor who controlled it had tolerated personal projects) so there was no rule to convict Gates of breaking. The university’s response was practical rather than punitive: it restricted access, and Gates and Allen finished the work on commercial time-sharing systems they paid for themselves. Still, the man who told an entire hobbyist community that “most of you steal your software” had built the very software in question on computing resources he was not entitled to use for a private company, and escaped consequences on the absence of paperwork.

Misconception: “Altair BASIC was written on an Altair”

It was not. The MITS demonstration in early 1975 was the first time the code ran on real Altair hardware; the interpreter had been developed entirely on a PDP-10 mainframe using a software simulation of the Intel 8080, because Micro-Soft had no machine to test it on. The story matters because the same borrowed-mainframe episode is what triggered Harvard’s investigation of Gates. See Myths and Misconceptions.

None of this makes the hobbyists’ copying legal or the letter’s economics wrong. Writing software genuinely does cost money, and Gates was correct that a market in which almost no one pays cannot sustain professional development. The irony is not that Gates was a hypocrite about money; it is that the entire early history of personal-computer software was built by people (hobbyists and founders alike) helping themselves to whatever resources and code they could reach, and that the boundary between “sharing” and “stealing” was being drawn in real time by the people who stood to profit from where it landed.

Fifty Years On

The half-century since has been one long test of the letter’s premise. Gates’s own answer (a moral appeal, shaming copiers as thieves) did not stop the copying; it arguably made “software should be free” into an organized movement rather than a vague habit. The industry then tried to solve the problem with technology instead of rhetoric: copy-protected disks, dongles, serial numbers, activation servers, and finally full digital rights management. Each generation of protection was broken, usually quickly, and the cracked releases fed the warez scene that industrialized software piracy far beyond anything a Homebrew tape-copier imagined.

What actually reduced casual piracy was neither shame nor DRM but a change in the shape of the product: software as a subscription and a service, where the valuable part lives on a server you don’t control and updates never stop. You cannot copy a paper tape of something that isn’t a discrete artifact anymore. Gates lost the argument he started (copying was never defeated) but the industry he helped found won anyway, by making software into something there was less and less to copy.

Dead End: Winning the Argument by Moralizing

The Open Letter’s rhetorical strategy (persuade people that copying is theft and they will stop) is a technique that has failed continuously for fifty years. Every anti-piracy campaign built on the same premise (“You wouldn’t steal a car”) has been mocked as often as heeded. The lesson the industry eventually absorbed is that copying is governed by friction and convenience, not by moral conviction: people pay when paying is easy and the alternative is a hassle, and they copy when it is the other way around. Streaming beat piracy where lawsuits and lectures never did. Gates’s letter is remembered less as a winning argument than as the moment the argument began, and as an early, unheeded demonstration that you cannot moralize a copying problem out of existence.