My 25 Years at Digital Equipment Corporation (1966–1991)

I don't have many readers here, but some stories deserve to be written down anyway. I worked at Digital Equipment Corporation from 1966 to 1991 — a quarter century that took the company from a scrappy minicomputer maker in an old wool mill to the second-largest computer company on Earth, and then into the decline that eventually ended it.

When I joined in 1966, DEC was already proving something IBM didn't want to hear: that smaller, cheaper machines you could actually sit down and work with beat room-sized mainframes fed by punch cards. The PDP-8 had just shown the world that a computer could fit on a workbench instead of behind glass walls. That was the spirit I walked into — engineers who cared more about getting the machine right than about the quarterly numbers.

The 1970s were, in hindsight, the best years. The PDP-11 reshaped software (Unix and C were both born on one), and then Gordon Bell's team built the VAX https://www.youtube.com/watch?v=O89XmbCkeQ4&list=LL&index=1— 32-bit machines that could go head-to-head with IBM mainframes at a fraction of the price. Revenue went from a few hundred million to over $14 billion. We were, for a while, unstoppable.

But I also watched the cracks form from the inside. DEC's product groups competed against each other so fiercely that customers sometimes got contradictory advice from two DEC salespeople in the same week. And when Dave Cutler's team proposed Prism — a new RISC architecture and operating system meant to move DEC beyond the VAX — management killed it in 1988 to protect the existing VAX business. Cutler and most of his team left for Microsoft that same year. What they built there became the foundation of Windows NT. I remember the mood inside the company when that happened: frustration, but also a sense that leadership simply couldn't imagine betting against its own cash cow.

By 1991, DEC posted its first full-year loss. Ken Olsen — the man whose name was inseparable from the company — was forced out the following year. I left in 1991, right as that chapter was closing, watching a company that had beaten IBM at its own game struggle to believe the same logic could be turned against itself.

I still think about DEC's real legacy less in terms of stock price and more in terms of philosophy: the idea that the person doing the work should be able to touch the machine directly, not wait behind a priesthood of operators. That idea is now so basic to computing that nobody credits it anymore. But those of us who were there know where it came from.

If you also worked at DEC during those years, I'd love to hear your own memories in the comments.