Roger Needham: A very forward man

Whatever you use your computer for, the chances are that Roger Needham had a hand in developing it. Wendy Grossman looks back on a remarkable life

Roger Needham, the executive director of Microsoft's Cambridge Research Lab, died at the end of February of an incurable cancer that had been diagnosed only a couple of months earlier. It's commonplace to say when someone dies that he or she was universally loved. In Needham's case, it seems to be true; it would be very difficult to find someone who disliked him.

It's also surprisingly hard to find anyone – almost certainly including anyone reading this – who hasn't used something that Needham innovated. If you've ever typed a password into a computer, it probably used the "one-way" encryption system: your original password is irreversibly encrypted and stored; subsequent login attempts are also encrypted, and compared against what's stored. The original password is never stored in a decrypted form, except in your head. He also demonstrated how to make a computer do time-sharing (serving many users) over networks in 1968, when nobody thought there was much point in it. Needham, of course, begged to differ.

Yet his formidable intellectual talent and knowledge gave him none of the arrogance that characterises some of the computing world. He served as a local councillor, and liked to spend evenings in the pub arguing about politics with his neighbours. When you met him, he was unfailingly kind and patient, explaining things thoughtfully and deliberately if you didn't understand them.

"He was part of my life for nearly 30 years," says Andrew Herbert, his successor as head of the research lab, who began as Needham's PhD student in 1975. "At every stage of my career, Roger was there with advice; it's hard to believe he won't be there now. There are many in the industry who regard themselves as people Roger helped along the way."

Needham was known for his aphorisms, most of which he credited to others. One that is often attributed to him (although he credits it to Sir Hans Kornberg) is: "Serendipity is looking for a needle in a haystack and finding the farmer's daughter." This comment applies as much to the type of basic research he did – which sometimes seemed futile at the time, only to be proved valuable later – as it did to his concerns about the consequences of IT.

It's commonplace to think of the past 50 years as an information revolution. Yet Needham thought the opposite. "I believe that it has not happened yet," he said in one of a series of lectures he gave at Aberystwyth in 2000. "So far, information technology has made incremental changes to our lives rather than revolutionary ones." For example, even though airlines depend on computers, if they did not exist the experience of getting on a plane in London and getting off it in Amsterdam would be pretty much the same.

Needham laid the groundwork for a number of important computing fields. But his speciality was security, and he worked on authentication protocols and access control, creating a whole area of computer science. In more recent years, he spoke often about the issues of privacy being raised by today's information systems, and became a supporter of the Foundation for Information Policy Research.

The hype about "internet time" makes it easy to forget a point that Needham made many times: discoveries in computer science do not in general get into users' hands very quickly. "The pipeline is long," he said at the lab's launch in 1997. "Look at the mouse: that was invented 20 years ago, and it took that long to move from the laboratories to being ubiquitous. An awful lot of research is viewed at the time as being futile. We did a demonstration in 1968 with MIT using a transatlantic telex line, showing how you could use the same system at such a distance. The pundits said, 'It won't work, it won't be economic, and people won't want to work that way anyhow.' " Yet mainframes haven't gone away, and time-sharing is essential for all sorts of businesses; just think of airlines, which rely on vast time-sharing systems stretching across continents.

One of the biggest changes during Needham's life as a researcher was the fundamental way that computer research was carried out. When he began, you researched computers by making them. "One of the things that sets us [the Cambridge computer laboratory] apart from most computer science departments," he said in 1993, "is we still make things." Hands-on construction clearly mattered to Needham: he and his wife, Karen Spärck Jones, built their house with their own hands while completing their PhDs. He was also a keen sailor, and owned a boat built in the 1880s to bring the catch back.

In 1997, at 62, he became the founding director of Microsoft's first research lab outside of the US; he was last at work there only a few days before his death. The company had tried to lure him to Washington some years before – both Rick Rashid, senior vice-president in charge of Microsoft Research, and Nathan Myhrvold, overall head of Microsoft Research, knew his work and wanted him. In the end, when Needham preferred to stay in Cambridge where his wife was (and still is) Professor of Computers and Information, they came to Cambridge to get him.

To a lot of people at the time, the Microsoft move was surprising, but as Needham noted, at his age the concept of "career risk" was non-existent. Myhrvold has said that the best way to recruit for a research division is to look for "talent and discontent". Needham fitted that description because of the increasing financial pressures within academia. "I spent 35 years at Cambridge surrounded by brilliant people," he said, "and I rarely had sufficient money to hire them."

On 17 February, knowing he was ill, computer scientists from across the globe gathered in Cambridge to celebrate his 50 years in computing, each giving a short paper, many of which are available as video clips on the Microsoft Research website. (www.research.microsoft.com/~aherbert/ needham_50_5.aspx).

By then, although mentally as sharp as ever, Needham was physically so frail as to need a wheelchair. On being presented with the bound volume of papers, he commented on the difficulty that computer scientists face in having to invent the theory of what they do as they go along.

Noting that he'd "contributed a few bits" to this effort, he pulled out a hard hat and put it on, saying, "But at the end of the day, I'm an engineer." There was laughter and applause. Eleven days later he died.