Gray: Brilliant Researcher With An Indelible Personal Touch
In an industry that has consistently exploited its research for rapid financial gains, Jim Gray stood out as caring more about the research than the gains. "Jim used to say, 'I love astronomy data 'cause it's worthless. It's got no commercial value'," recalls Alex Szalay, professor of astronomy at Johns Hopkins.
In an industry that has consistently exploited its research for rapid financial gains, Jim Gray stood out as caring more about the research than the gains. "Jim used to say, 'I love astronomy data 'cause it's worthless. It's got no commercial value'," recalls Alex Szalay, professor of astronomy at Johns Hopkins.That meant the two never had to worry about inadvertently acquiring somebody's else's data and calling in the lawyers as they built up a multiterabyte database like SkyServer.
In a pell-mell, exploitative age, he seemed to have time to execute brilliant work and still care about his fellow human beings. Eric Allman, author of Sendmail, worked with Gray on the Association of Computing Machinery journal's editorial review board.
"A couple of years ago, he gave us a 'brief intro' on new database directions that was incredible. We could have almost published the lightly edited transcript as-is as the entire issue," Allman says.
The January board meeting fell on the birthday of another member, Kirk McKusick, one of the original BSD developers. McKusick writes, "Jim got wind of this and arranged a special dessert for me and lead a melodious refrain of 'Happy Birthday' as it was brought in. Typical Jim."
Gray was thoroughly modern in his knowledge but a throwback to early university intellectuals in the way he shared insights on a personal basis. "He knew an enormous amount, and he spent a lot of time figuring out what others were doing and cross fertilizing what he knew. He was the ideal researcher," says Michael Stonebraker, a former Berkeley professor and now CEO of Vertica Systems.
He would have thought it fitting that an impossibly large collection of satellite data, gathered once he turned up missing, was reviewed by thousands of volunteers through the Mechanical Turk at Amazon, a feat never accomplished before in the search for a single, missing civilian.
"He would be thrilled. Jim was a big proponent of the parallelization of tasks," says Werner Vogel, CTO of Amazon.com. Gray sat on the committee that reviewed Vogel's doctoral dissertation in Amsterdam. "There was a human side to him, the way he'd mentor up-and-coming students. When I was offered the job at Amazon, Jim was the first person I called for advice," Vogel adds.
Gray earned the first Ph.D. in computer science at Berkeley; he worked on System R at IBM, he worked on nonstop computing at Tandem; he was always there at the beginning. Still, he would protest that he was one of many who had contributed to the defining of transactions, the codification of a two-phase commit process that had previously bedeviled programmers each time they sat down to write an application.
"When he received the Turing award for transactions, everyone thought it was right and obvious," says John White, CEO of the ACM. That unanimity doesn't attend every award.
From computing's equivalent of the Nobel prize, the 1998 Turing Award citation:
"Gray's work led to the definition of the desired key transaction properties: atomicity, consistency, isolation, and durability; and his locking and recovery work demonstrated how to build database systems that exhibit these properties."
There have been greater exploiters of breakthroughs, such as Robert Noyce and Gordon Moore; harder-driving entrepreneurs, such as Larry Ellison; and better technology packagers, such as Steve Jobs. But there's been only one researcher like Jim Gray. He won a unique place in modern computing. When he disappeared at sea, there were so many who wanted him back that they joined hands to mount an unprecedented search to find him.
That effort failed, but Allman says, "I can't help but think a lot of good will come out of this. Besides the creative use of technology, the cooperation between industry and government has been one of the best I've ever heard of."
His absence produced a degree of cooperation that might heretofore have been induced only by his presence.
Says Allman, "How I'm going to miss that laugh."
Building A Mobile Business MindsetAmong 688 respondents, 46% have deployed mobile apps, with an additional 24% planning to in the next year. Soon all apps will look like mobile apps – and it's past time for those with no plans to get cracking.
InformationWeek Must Reads Oct. 21, 2014InformationWeek's new Must Reads is a compendium of our best recent coverage of digital strategy. Learn why you should learn to embrace DevOps, how to avoid roadblocks for digital projects, what the five steps to API management are, and more.