Billionaire SAS co-founder keeps on coding

SAS Executive Vice President John Sall, who developed the data visualization tool JMP, remains its chief architect

John Sall doesn't have to work. As one of SAS Institute Inc.'s four co-founders, the 60-year old has a net worth of $3.1 billion, according to Forbes' estimate, ranking him the 196th richest person in the world (just ahead of George Lucas, Steven Spielberg and Ralph Lauren).

So why does he stay at the Cary, N.C.-based business intelligence giant, 33 years after starting the company with his graduate school professor, James Goodnight, who remains SAS' CEO?

"It's always been my job to be a statistical software developer," Sall said in an interview earlier this week in Chicago, where he was in town for a SAS user conference.

"I've invested my life in it. My net value, it's all vapor, tied up in company valuations. It's not real. I don't go out and buy yachts or do other strange stuff. I spend most of my time working," he said.

SAS' John Sall
SAS' John Sall.

Perhaps we can credit Sall's Midwestern roots for his drive. Born and raised in the central Illinois city of Rockford, he earned a history degree at Beloit College in Wisconsin and Northern Illinois University. It was while studying statistics at North Carolina State University that he met Goodnight.

Sall was one of the lead developers for SAS during its first decade. But with the launch of the Macintosh in the mid-1980s, Sall got interested in the computer's graphical visualization capabilities.

He started working on a data visualization tool called JMP (pronounced "Jump"), an acronym for "John's Macintosh Project."

The software took off with the release of a Windows version in 1994. Out of 250,000 users, most run JMP on Windows, followed by the Mac and a small number on Linux.

Its users are primarily research scientists and engineers, as well as manufacturing experts following the Six Sigma methodology. They use JMP to design and record data from experiments, or keep quality control statistics.

For instance, most leading semiconductor vendors have long used JMP to help them improve chip yields, said Sall, though he declined to name any of them apart from National Semiconductor Corp.

Unlike most similar software that are aimed at helping statisticians display and confirm known trends, JMP is what Sall calls a "discovery" tool that helps researchers see and interpret anomalies.

For instance, the Chicago Botanic Garden used JMP to analyze DNA from the tropical breadfruit and determine that the seedless, starchy fruit was created by the deliberate hybridization of two fruits, the breadnut and the dugdug.

Elixir Strings uses JMP to help create its good-sounding, non-breaking guitar strings.

This emphasis on discovery is why the celebrity guest speaker at JMP's user conference earlier this week was the journalist and author Malcolm Gladwell, who wrote the popular books Tipping Point and Outliers.

"Journalists are like detectives. A lot of our customers are like that," Sall said. "They don't just want to prove things you already know. That's lawyer stuff. They want to look at statistical outliers and figure out new things."

Although still a desktop software product, JMP is so robust today that the latest, version 8, can display graphically millions of rows of data.

While JMP's core customer remains those researchers and engineers, the product has in the past several years also been touted for mainstream business users as a front-end for SAS users.

An executive vice-president at SAS, Sall remains JMP's chief architect, running a team of 100, including 20 developers -- a sliver of SAS' 11,000-strong workforce.

That's fine by him. "SAS makes all of the money. JMP is the fun product," Sall said.

FREE Computerworld Insider Guide: IT Certification Study Tips
Editors' Picks
Join the discussion
Be the first to comment on this article. Our Commenting Policies