Man of his time

“The illusion of order is that we can learn to predict everything.”

That was the point driven home by Jaron Lanier, “the father of virtual reality” and a Time Magazine 2010 Person of the Year, when he spoke on technology and society at the Pennsylvania College of Technology’s Centennial Colloquia series Monday night.

Lanier, who worked for Atari before founding the first company to sell virtual reality equipment, was part of some of the biggest advancements in computer science in the last three decades – such as the technology that made a website like Wikipedia possible – but his view of the current “information-sharing age” was not favorable.

“We think that just by observing and sharing all of this data we will figure out how everything works, what the meaning of life is,” he said, “but it doesn’t work that way.”

Lanier explained that in the late 90s and early 2000s, nearly every sector of the economy and society was looking at computers and what they could do with data, and businesses especially were interested in how computers could be used to their greatest advantage.

He said when he was working as a consultant for what was, at the time, the nation’s largest health insurance company, an executive had the idea “to do the exact opposite of what health insurance companies usually do.”

“Normally, to increase your profit, you’d try to insure as many people as possible,” Lanier said. “His idea was to use data and information to predict who would be the least likely to get sick and to go after insuring them.”

The problem? There is no such thing as a completely accurate prediction.

“Statistics are not always what happens,” he said. “Statistics might say you’re going to do one thing, but for no predictable reason, you change your mind, or something else random would happen.”

It’s an unstable business model, he said, and one that could come crashing down on a single deviation.

However, businesses were – and still are – eager to use data to plot behavior and spending patterns in search of the ultimate “wealth machine,” a program or algorithm that never will stop producing a profit because it can accurately predict human behavior, he said.

“Some (people) have an advantage when it comes to dealing with this information because they have better resources – think Google,

your bank, Verizon, etc.,” he said. “They pull it all in and look at it to try and determine what people want to buy or if they should get a mortgage, but it’s not always accurate.”

Those inaccuracies, he said, can have drastic consequences, such as the auto industry bailout a few years ago.

It’s why he believes in a different type of information sharing, one that is monetized but is “fluid and voluntary:” users pay individuals or groups that have done the research and compiled the data, so that the most accurate information is shared and the right people are paid for it.

Lanier believes that middle-class jobs already are being threatened by the idea of “free” information.

He gave the example of language translators who have seen a huge drop in work because now everyone uses the Internet for translation.

“Information is not free,” he said. “People still do the research and write the papers and input data, but the problem is that users are not paying for it.”

Translation programs, he pointed out, do not undertake the action of actually translating. Instead, they create a “mashup” of existing translations, ones that were done by actual people and run an algorithm to determine how closely it matches what has been entered.

This will happen in more and more sectors of society, Lanier suggested, and since only a small group would have the enormous wealth that comes from using this “free” information, our economy could become extremely unstable.

The next installment in the speaker series is April 22, when Penn College professors Rob Cooley and Mark Noe present “Google Meets Aldo Leopold: Information, Technology and the 21st Century” at 7 p.m. at the Klump Academic Center Auditorium. Admission is free and open to the public.