Lotus 1-2-3 turns 30: Mitch Kapor on the Google before Google
Surviving tech's bubble, but not Microsoft
Before Apple and Google turned computing into a webified, personalised and mobile experience, there was Microsoft. It was Microsoft that set the computing paradigm with a layer of software called Windows, which made computing personal, powerful and affordable when married with Intel chips. But before all of them, there was Lotus Development Corp, with its Lotus 1-2-3 software - so-called because it integrated three elements: spreadsheet, database and graphics. In many ways, the 30-year-old software package laid the foundations for the type of productivity app that's so ubiquitous in the modern computing experience and yet so important it has been thrown into the cloud by Google with Docs, Apple with iWorks and Microsoft with Office 365.
Lotus 1-2-3 went on sale 30 years ago on 26 January. It was a simple spreadsheet package that tapped a need among business people to quickly store and manipulate their numbers, to slice, dice and fiddle their data. The spreadsheet conquered business from Wall St to the UK high street as a tool to record results and chart projections.
1-2-3 wasn’t the first spreadsheet for the PC - that honour went to VisiCalc on the Apple II - but it was so successful that it transformed the PC from something for nerd hobbyists such as Bill Gates and Steve Wozniak to a tool for serious business users on a budget.
“It was a product that really legitimised the PC worldwide as a general business tool for non-technical users more than any other product... People were buying PCs to learn 1-2-3,” co-creator Mitch Kapor tells us in an interview on the 30th anniversary.
“It was as big as anything that’s big today,” said Kapor. “It was the Google or Facebook of its time. The market size was orders of magnitude smaller but the magnitude was the same.”
1-2-3 sold so fast it blew apart Kapor’s first-year sales projections by $50m, selling at $495 a copy, and helping persuade Kapor to take his year-old Lotus Development Corporation (LDC) public the same year – LDC was formed in 1982. For comparison purposes, just two years later Bill Gates was selling Windows 1.0 for $1 a copy. Gates IPO'd Microsoft 11 years after it was founded, in 1986.
'It was as big as anything that’s big today... it was the Google or Facebook of its time. The market size was orders of magnitude smaller but the magnitude was the same' - Lotus 1-2-3 designer Mitch Kapor
Yet, it’s Excel from Microsoft – not 1-2-3 – that’s king of the numbers three decades on. What's the story?
Circumstance favored 1-2-3 in the early 1980s. Advances in processors, memory, interface, engineering and materials combined with a coming together of geeks, engineers and entrepreneurs around that time saw the construction of the first PCs. Motivated by idealism or profit, people delivered systems that meant you no longer needed the budget of a big corporation or a small country to buy a computer and run it.
Dan Bricklin and Bob Frankston created Software Arts and built VisiCalc, the first spreadsheet, in 1979 for Wozniak’s Apple II. But VisiCalc faced two problems: firstly, the Apple II was too pricey for business – selling at up to $2,638 for a system with 48K of RAM – and relatively weak on memory. Secondly, a new machine had appeared offering greater memory: IBM’s 5150 PC. The 5150 appeared in 1981 and initially came with up to 64Kb of RAM, but Bricklin and Frankston were too busy fighting Personal Software on VisiCalc to develop the application so that it could take properly advantage of the IBM machine’s power.
Enter Kapor, who’d been a product manager for VisiCalc on the Atari at Personal Software and who became the designer for 1-2-3. He worked with Jonathan Sachs who worked on architecture and coding. They created Lotus Development Corp in 1982.
“They [Bricklin and Frankston] had a very weak entrant for that machine that didn’t take advantage of the IBM PC... people wanted to build large spreadsheets and they couldn’t," Kapor says. "The VisiCalc for PC was hard to use. I said: ‘Let’s build something that really takes advantage of these capabilities’.”
What Kapor and Sachs delivered was a powerhouse that packed elegance with ease-of-use. And they managed that not on Windows but on PC-DOS.
“I felt the key to unlocking the market was making something ordinary people could use. It had to be better than single-letter commands,” Kapor tells The Reg. “We spent lot of time on how to work efficiently, so we had slash commands with the keyboards so people quickly developed muscle memory.”
1-2-3 packed more commands than the competition - it automated tedious charting, detailed spreadsheet manipulation and allowed you to calculate formulae using rows and columns instead of just row and column order.
And, unlike the rivals, 1-2-3 wasn’t hard to understand or difficult to use. There were jargon-free prompts, explanatory messages, and fail-safe mechanisms in case of mistakes. 1-2-3 was the first program to make full use of the IBM keyboard enabling commands via the F and slash keys. 1-2-3 packed a tutorial package and context-sensitive help – all features that had one reviewer enthusing about 1-2-3’s attention on “human engineering.”
It packed in so much that 1-2-3 exceeded the 5150’s memory, soaking up 256Kb and making it possibly the first case of the software pushing the hardware.
Kapor in 1982 had forecast first-year sales for 1-2-3 of $3 to $4m based on the performance of Personal Software, which had made $12m annually and had shipped 300,000 copies of VisiCalc. 1-2-3 made $53m in year one, selling at $495 per a copy on a PC bundle of the day that was priced $5,000 - which included a printer and monitor. “We had no idea - nobody had put out software like that before,” Kapor said of his sales estimates.
At last - a credible IPO!
Kapor decided it was time to go public, so LDC floated at $18-a-share in 1983. The IPO was sooner than planned but Kapor reckoned he’d wanted to take advantage of the first tech bubble around him – a string of now mostly forgotten-about personal computers had been going public: Vector, Victor and Eagle.
It was now or never.
“I was looking at this and I thought the quality of the companies going public is decreasing rapidly: most of those companies had no chance and they were foisting themselves on the public. So I said we can go public now, much earlier than we planned, or we will have to wait a couple of years because the window will close. I was not a student of Wall Street but I was a quick study. I thought getting liquidity and having currency to do acquisitions and cash in the bank would be worthwhile,” Kapor said.
It was a dizzying experience. “I woke up every day and I wouldn’t say I was constantly terrified, but I was worried it would all evaporate, [that] I would make one bad decision and it could all fall apart because it had come so quickly,” Kapor says.
It didn’t evaporate, but this was certainly the high point for 1-2-3.
Thirty years on here’s a test. Ask yourself a question: What’s the word you use when talking about your “spreadsheet”? Answer: “Excel”, in the same way you say “Google”’ when referring internet search or “Tipp-Ex” when talking about correcting a mistake on a sheet of paper using a quick-drying white fluid. 1-2-3 still exists as a spreadsheet owned by IBM, which bought LDC for $3.5bn in 1995, but 1-2-3 and the Lotus brand are barely seen outside of blue-blood enterprises and IBM shops.
What happened? A couple of things happened – and one of them was Microsoft.
LDC stumbled twice. First, 1-2-3 lacked a word processor, something that became a staple of personal productivity and left a hole in the suite for somebody else to fill. A word-processing app had been planned but it was dropped because of difficulty in coding and hitting deadlines. Lotus returned to word processors in 1984 chart, spreadsheet and docs package Lotus Symphony for MS-DOS from a then-unknown developer Ray Ozzie, who went on to develop the hated Lotus Notes groupware. While Symphony was still largely based on the spreadsheet, it was never as successful as 1-2-3. “We had the right idea but the wrong architecture,” Kapor says.
“I woke up every day and I wouldn’t say I was constantly terrified, but I was worried it would all evaporate, [that] I would make one bad decision and it could all fall apart because it had come so quickly” - Kapor
Next problem was that LDC had stumbled on its re-writing of 1-2-3 from macro assembler to C (to make it more easily portable). LDC delayed release of the rewrite until 1989 and even then outcome was not one but two products: the rewrite had produced a code base so big it couldn’t run on low-spec machines, so LDC responded by developing version 2.2 on assembler code that targeted PCs lacking extended memory and version 3.0 for “high-end” PCs.
In the meantime, rivals mushroomed. One of these was Microsoft, whose Multiplan spreadsheet was replaced by Excel for the Macintosh in 1985. Excel landed on Windows 2.x in 1987. As Windows grew, so did Excel, and the spreadsheet software was swept into Office in 1990, along with Word – both built under Charles Simonyi - and PowerPoint, which Microsoft had bought.
“It became clear to me by 1984 that Microsoft was likely going to be the big winner in the PC software apps and operating system category, partly because of the dynamics of owning and controlling the operating system: that gave you enormous power and I came to see Bill Gates was fierce competitor,” Kapor says.
“He was smart and ruthless in pursuit of business objectives and I was not as ruthless: I’m super competitive, but... my goals and values are not about crushing the comp and I don’t have the kind of personality to carry out a strategy like that. It presented a dilemma.”
The C re-write and competition from Microsoft happened in the absence of Kapor - he left LDC in 1986 and was replaced as chief executive by former McKinsey consultant Jim Manzi. Running a large and growing public company wasn’t to Kapor’s tastes – he preferred startup mode and entrepreneurism, he tells us.
Calculating the outcome
Kapor now reckons he’d have done things differently, staying longer and building a more sympathetic board and management and cultivating his base inside LDC.
“Had we done it the right way in 1984, things would have been different,” Kapor says. “We get credit for the right idea about what people wanted but we didn’t do it the way we should. So we lose credit on that,” Kapor says.
“With hindsight it seems so clear there are a fundamental set of productivity tools and if you are a company that makes productivity apps then having a strong offering in each is the right strategy. Microsoft gets credit on that - and for executing.”
Kapor says the lessons he learned from those heady LDC days have been applied to "hundreds of other companies". Kapor has since been a founding investor in Real Networks, Second Life's Linden Research and the first commercial ISP - UUNET. He was also a co-founder of web privacy watchdog Electronic Frontier Foundation and a founding chair of the Mozilla Foundation.
Any regrets that despite 1-2-3's role in shaping computing culture, it's Excel that took the money and the market? “It’s now long enough ago that I don’t have regrets,” Kapor tells us. ®