Imagine you’re Ray Ozzie, born half a decade after Turing outlined his Test to see if a machine is truly intelligent. You’re lucky enough to attend a high school with ARPA funding, ARPA of internet forerunner ARPANET fame, with Teletypes connected to time sharing mainframes, with a teacher looking for volunteers to figure out how to use the thing.

FORTRAN was finalized when you were two. And now, in high school, you’re coding in FORTRAN and Basic in downtime at school, a string of words that would have been nonsensical a few years prior. Logic would have you major in Electrical Engineering; luck would place you at the University of Illinois Urbana–Champaign. And UICI was where everything from the first eBook to the first graphical browser, Mosaic, would be invented.

Coding was manual, tedious, in those late-1970’s days. “In my classes, we were using card decks. We were having to ... write out our machine problems on paper,” Ozzie reminisced with me on a call, then “right into the key punch, submit our decks, and get our lists back.”

“Directly across the street, there was this room with all these orange screens, with people doing things,” said Ozzie. “I had to figure out what was going on, and managed to wrangle an account on the system.”

PLATO, it turned out, was the orange-and-black system, the University of Illinois’ early educational computer. By day, programming was painstakingly hand-coded. By night, “at 10PM ‘till 6AM,” said Ozzie, here was this computer of the future. “People were just using it for all these wacky things.”

“Online community. There was an early form of email called personal notes. Discussions. There was a thing called Talkomatic, which was essentially online real time chat a character at a time—not line at a time. No matter what you were doing, you could press an escape button, it would open up a little box on the bottom of the screen and you could type in who you wanted to talk to. There was a special account called Consult where you could share your screen, and they could see what you were doing and talk to you while you were doing it.”

“It was all of this online interaction,” described Ozzie, at a time when the majority of people had never touched a computer.

So imagine Ozzie and friends’ shock when they graduated and went into industry, only to discover that the future was far from evenly distributed. “We were just shocked at the lack of use of the computer as a communications device,” he recalled. You could have taken that as a given, that industry was doomed to stay behind the state of the art. Or you could take that as a mission, to bring the state of the art to the masses. Slowly, over weekends, a plan came together. You’d seen the future, and knew you had to do your part to make it more equally distributed. And so, Ozzie along Tim Halvorsen and Len Kawell, friends from the University of Illinois, set out to build the future that David Wooley’s Notes had shown on the PLATO computer.

“It wasn’t like I had a choice, you know?” said Ozzie. “I had to do it. Because it was just super compelling, and everybody was ignoring it.”

Inventing while the paint’s still wet

On to the drawing board, in 1984, coding the first bits of what would come to be called Lotus Notes in a callback to the PLATO Notes. There was a whole universe to invent. Collaboration on terminals connected to a mainframe—the model used by PLATO and other early networked computing models—meant everyone was technically working on the same computer.

The PC moved the actual computer onto everyone’s desks, and the center of gravity shifted. No longer was there a shared storage. Every computer running the same operating system wasn’t even a given. The ideas of email and collaboration were in the air but the specifics were mostly left to be invented. The words to describe collaborative software weren’t even finalized yet, much less keyboard shortcuts and user interface conventions.

“We had opinions on computer support and collaborative work, based on observations that we had on PLATO,” said Ozzie. “But we weren’t thinking in the realm of standards.” Bringing the magic of collaborative computing out of the cloistered halls was work enough.

So the team set to work. “We estimated that maybe 50-60% of the development calories were spent on memory management, because the resources of the era,” said Ozzie, at a time when a computer might come with only 256kb of ram. A single platform wasn’t enough, either: “We did a simultaneous ship on OS/2 and Windows in ‘89,” before expanding to two variants of Unix, Mac OS, and NT.

“Resource allocation was based on one very simple thing: We are a collaboration product, so the more people we can get to use it, is more important than how beautiful it is,” said Ozzie. If push came to shove, support for an additional platform was worth its weight in less-polished interfaces.

Because polish is in the eye of the beholder, and if you’re doing something new where the standards aren’t set, you’re just trying to see what sticks. Or as Ozzie stated it, “When you’re early, you’re just trying to get a match between user and technology fit—what they’d call product-market fit today.”

And Lotus was doing everything. Email, with rich text, attachments, and push notifications of a sort years before similar features would trickle down to universal email. But not just email. It could be anything you wanted, an app that you could morph into a CRM or help desk, a tool where you could correlate data or collaborate on documents.

“It was like Microsoft Access, it was born in an era where we idealistically believe that users should build their own apps and that they should be pliable to their own needs,” said Ozzie, designed to work the same on half a dozen platforms yet somewhat uniquely from each platform itself. It was early, and “it was destined to be fragmented.” It hinted at what web apps were to bring in Salesforce or HubSpot-style CRMs and automations, with collaboration and notes that hinted at modern Notion and Confluence-type tools.

“The technologies are inside baseball,” said Ozzie, when asked about how he thought about the standardization process that made email universal and, in the process, left Notes feeling like retro-futuristic software. “They’re very important to the dynamics of how things grow because of standardization and commoditization. I think that’s orthogonal to why [technologies] become popular.” People care about getting things done, about building tools that fit their needs. On a long enough time frame, that includes interoperability with others, one of the many reasons that standard email won.

But when you’re new, when you’re just starting out, when you’re trying to invent the future from scratch, you have to just start. A closed system meant that you could do things your way, as a shortcut to get the world collaborating before the world decided how collaboration should work. “If you’re going to build something, you have to be opinionated about how certain things are architected and built,” as Ozzie put it. It meant for their team building something that shipped on disks in boxes, software that had to get it right the first time and could not be infinitely tweaked with over-the-air updates. It meant knowing that you might get obsoleted, but in the process, you’d get to invent the future.

“Somebody has to pioneer the concepts,” said Ozzie, “and I feel good that we contributed a little bit in that direction.”

The race is not to the swift

Because in the end, Lotus Notes didn’t set the standards for email or database-powered application development. It was loved and hated in equal measure, loved for its flexibility (“A well run Notes system was a real pleasure to deal with,” reads a top Hacker News comment) and hated for its lack of conformance (it showcased the “survival of the unfittest,” as Jeff Atwood put it, offering “death by a thousand tiny annoyances.”).

Ozzie learned to take it all in stride, chatting with “the tongue-in-cheek @IHateLotusNotes ‘global support group,” and “proudly wore the T shirt they sent me,” he says. And to this day, when “Somebody says ‘I used to use Notes back in the day,’ I say ‘My thanks and my condolences.’”

“It’s disappointing to see people criticizing your work. But you know what? You have to develop a kind of thick skin when you do those things.”

Because the goal wasn’t perfection, wasn’t to solve collaboration forever this one time. It was, instead, to turn this glimpse of the future into reality.

“As long as people are using our work, and it’s providing value, how could we not be overjoyed?,” said Ozzie.

People used his creation, cared enough to criticize it, to, years later, fill threads with recollections of the best and worst of times in software he’d coaxed to life. It’d been successful enough to be acquired, then given him a chance to build an adjacent dream at Groove before it, in turn, was acquired.

And here, years later, Ozzie was happy about it. Happy about having been there at the cusp of the personal computing revolution, to be able to look around today and see people using the technologies he knew one day we’d all use, even if not in the exact format he’d imagined.

It’s easy to feel pessimistic, when the ground under your feet is always shifting, when the technology you mastered yesterday is obsolete tomorrow, when the thing you set out to build doesn’t take the world by storm. Tech prompts those feelings like no other industry, with its ever-changing, Moore’s Law-powered restless energy.

Yet most of us would be lying if we didn’t say that was what attracted us to tech in the first place. “The thing I’ve always loved about our industry is that about every five to eight years there is some major platform transformation,” said Ozzie, from mainframes to minicomputers then microcomputers. What’s new when we’re in university and early in our careers is the future; what’s new mid-career threatens our very livelihoods and the experience we’ve painstakingly built up.

There’s been, for years, a feeling that we’re overdue for the next post-iPhone device, something to keep the innovation cycle going. Yet increasingly it looks like LLMs are that next shift—and for once, the doom may be grounded.

“Right now, people are worried about their jobs,” said Ozzie, for good reason. There’s a big transformation that’s going to happen across all industries because of the hollowing out of the middle. And I don’t want to trivialize that in my excitement.”

Yet with that caveat, he couldn’t help but imagine that the world of creating your own apps, Notes and Access style, that “maybe it will emerge again in an LLM world.”

“Ever since the Notes era, products like Notes, Lotus 1-2-3, they enabled the domain practitioner to potentially solve problems themselves. They’re democratized problem solving. A real estate officer or a plumber can solve their needs without having to involve third parties and explain to them the need. It’s going to benefit so many people,” said Ozzie, with palpable excitement. His grandkid had coded a game; a captain on a whale watching excursion he’d been on was scheduling stuff for his business with Claude.

It was always the promise of computers, that they’d take care of all the tedium. “It portends the ultimate realization of removing all that friction between you and... creating,” he said.

“I don’t know what’s going to happen to the software industry. But I do think we’re in for, over the next five to ten years, major transformation that is going to be quite exciting.”

Which is where that optimism comes back in. One could look at this as a glimpse of the future, and think that you have no choice. You have to build what’s next, because you’ve seen the future and how could you not be excited at building while the very rules are being rewritten?

So I asked Ozzie how he stayed positive, optimistic, excited about the new. How he didn’t look at how the world passed Lotus Notes by, and regret it. How he could be directionally right about networked computing while failing to build the actual software most people would use decades later. How he continued to find AI exciting even saying that “you can’t be in denial about how scary all of this stuff is.”

And he relayed an anecdote. “A long time ago, somebody told me—and I couldn’t shake it—that optimism or positivity can be local. It’s enjoyable when it’s local. But negativity is viral.”

“As a leader, you have to come in every day with a smile on your face regardless of the stress level, regardless of the negativity, because if people sense anxiety or insecurity, it spreads virally.

“And so I’ve enjoyed being around optimists,” said Ozzie. And so did I, walking away from Zoom with a bit of the wide-eyed wonder having rubbed off.

We’re standing on the shoulders of giants, with tools at our disposal that the earliest computer scientists knew we’d one day have. Time and chance happen to us all. How could we not make the most of it, and build that thing we’ve always dreamed of?

The world’s not perfect. Never will be. The future will arrive half-baked, the winners far from what we’d have expected at first. But optimism is a choice. A choice that birthed the ideas behind the things that make up a modern life. That’s worth smiling about.