Wednesday, May 13, 2009

Medieval secrets behind Microsoft research

Thomas Edison is famous for lightbulbs and phonographs, but his major invention was Menlo Park, the world's first industrial research laboratory. Taking up two city blocks and stocked with everything that could conceivably be useful for researching, it provided the model for the great corporate labs of the 20th century IBM, Bell, General Electric. The general principle is easy to grasp: if you give creative, disciplined people the chance to have good ideas and the means to follow them up, good things will follow.

At a Microsoft Research's open day at Cambridge recently, this line was promoted heavily in the keynote speech: fundamental research helps generate brand new technologies that give companies competitive advantages. In some cases, that's unarguable. If Intel shut down its research, it would die overnight. But what would happen if Microsoft stopped doing research? Based on the researchers' demonstrations, the answer is 'not very much'.

The ideas on show in Cambridge, while good and interesting, did not address Microsoft's core problems, nor even any of their minor ones. There was research into ecological systems, into displaying networks of influence, into low-power network hardware, into capturing people's lives as a time line. Try matching those with any known Microsoft strategy or any conceivable one in a way that makes compelling sense.

With Intel, you can see the research up on the screen. I've had a briefing from a solid-state physicist on a new transistor design and seen it emerge as a major strand of processor strategy three years later. With Microsoft, it's hard to trace such developments easier now than it has been, but the link between bright idea and bottom line is very weak.

It's not as if you can't just shut down fundamental R&D. Apple had a classic operation in its Apple Research Labs, which did good work in networking, data recognition, human-computer interaction and so on. It was highly thought of, and some of its legacies live on in QuickTime and many other ideas spread throughout the Web and software development.

It ran for 11 years. Two months after he regained control of the company in 1997, Steve Jobs shut it down.

Apple survived the operation. Microsoft would too, purely in terms of the products it makes. That's because big company research in this instance is not primarily about the science and technology, it's about marketing marketing of a kind Apple decided it just didn't need, but Microsoft needs more than ever. And they're not 21st- or even 20th-century ideas at work: the principles go back much further.

This is hidden by a common misconception about fundamental research. Ask someone to describe a scientist or inventor at work, and you'll get a picture of an eccentric loner crouched over a bench in a deserted lab late at night. What research is, really, is networking. It's about influence and status and competition and being the first to know, all of which only work well in a large group of peers. Becoming part of a big industrial R&D operation while continuing an academic style of working is landing a plum; there's money, and brand recognition among the laity. And so, the entire research community warms to the idea and its sponsor.

This works all the way up the chain. The director of a big R&D organisation will find doors open and ears pricked at the highest level of universities and government; the divisional head will be lionized at conferences. You can see why Jobs found this a useless distraction in building the Apple brand, which stands for direct user experience. Microsoft, on the other hand, relishes every chance to become part of the infrastructure of influence. Individuals don't buy Microsoft products: they get them from organizations or by default. Microsoft's most important marketing is to those in power, and what could be a finer advertisment of your suitability for partnership than a big building in Cambridge University filled with happy academics? It worked for King Henry VIII: it can work for Steve Ballmer.

In its own way, the company is influenced too. Being a high-profile public supporter of a social good like fundamental research means you have to look very much as if you take it seriously. The first open-source MS software came from MS Research's work in IPv6; the academic necessities of collaboration and openness feed both ways and that does the company more good than it'll ever admit on the record.

What research is for, in cases like Microsoft, is status. It's a very tax efficient, with many valuable and wonderful side effects that occasionally benefit the company, but its primary task is marketing. It's proof, even in these most distressingly modern of times, that patronage works.

This article was originally posted on ZDNet UK.

  • Microsoft seeks the right touch for Surface
  • Microsoft steps up online search R&D in Europe
  • Microsoft trumps Apple in battle of the brands


  • Microsoft seeks the right touch for Surface
  • Microsoft steps up online search R&D in Europe
  • Microsoft trumps Apple in battle of the brands
  • No comments: