INTERNET SPECIAL ......You Can’t Imagine Internet at 60

You Can’t  Imagine Internet at 60 

The internet turns 30. But it is in the next 30 years that the changes will be truly profound and radical

    The internet just turned 30. That might come as a surprise, as the global computer network seems both older and younger than that. Older because it is such a part of life now — roughly a third of humanity is now regularly online, and its use is so ubiquitous, even people over 40 find it hard to remember a world without it. And younger because it’s still constantly changing, showing us new games, new programes and new fads. Whether it is relaying the latest gossip or teaching us how to do a South Korean rapper’s pony dance, our electronic pal seldom acts a day over 13.
For better and worse, the internet has changed the world, not least for India. But some analysts say that the changes we have seen in the past 30 years are nothing compared with what we may see in the next 30.

Birth of a Network

Although scientists had networked computers as far back as the 1950s, no one had developed a common language that would allow these networks to communicate easily between each other until the late 1970s.
    At that point, the US military had realised that because of the wide variety of communications systems they used, their communications networking problems were bad and only going to get worse. “They would never be able to get out from under this diversity... and they would constantly have to adapt to the future,” says David P Reed, a Massachusetts Institute of Technology computer scientist who was part of the group the military asked to design “an internetwork” to bridge those gaps.
    Reed and his colleagues found the problem an interesting one. “We weren’t particularly focused on military effectiveness, but saw this an early warning of all kinds of challenges,” he says.
    The outcome of their experiments was the Transmission Control Protocol/ Internet Protocol (TCP/IP), the data protocol adopted on January 1, 1983, which made it possible to create a single network of roughly 500 host computers at universities and various US government installations.
    Reed says he and his colleagues had a sense that they were doing something important, but thought TCP/IP would be the first of many iterations. “Most of us never thought that this particular internet, which would be a very experimental thing, would last very long,” says Reed, now an adjunct professor at MIT.
    TCP/IP solved several problems that had vexed US military communication specialists: like, how do you maintain control of a computer network even as you add more computers to it without having it collapse?
    The TCP/IP designers’ answer was: you don’t. Instead, by keeping the structure as simple as possible, not requiring 100% delivery of data, and making security the concern of each end of the system rather than the entire network, they created a structure that had no central control and could easily reroute packets of data if one path were congested or closed.

A Kludge?
However, for at least the first seven years after the launch, it was far from certain that their internet would grow up into the internet. “We expected that a company like IBM would see what we were doing and say, well, I could do that really well, and I could do a much better version,” says Reed.
It was clear a big data network was coming, but TCP/IP wasn’t the inevitable solution, says Andrew Odlyzko, a professor of mathematics at the University of Minnesota. “It’s a kludge, as a technology,” says Odlyzko, a former Bell Labs researcher. “I don’t think it’s optimal in any way, but it works. And it happened to satisfy the needs of the moment.”
:: Bennett Voyles
    As late as the early 1990s, businesses continued to propose alternatives to the TCP/IP-based network. “At the time in the marketplace, there was actually a lot of scepticism about this approach,” Reed recalls. For one thing, engineers didn’t like the idea that no one supported it.
    However, the alternatives proposed by telephone companies and the International Telecommunications Union never caught on. Two factors held them back, in Reed’s view. First, most of the major telecom research labs had “a very planningoriented style” for doing research, he says. Second, they were “hobbled by a desire to go slowly”.

The Growth
But they didn’t have the time they thought they did, particularly as personal computer manufacturers were creating a vast new base of computers already pre-wired for TCP/IP connections.
    By December 1995, 16 million people were online worldwide — and the numbers kept growing, leading to a cultural and investment mania, the dotcom stockmarket bubble that began to pop in 1999.
    However, 12 years later, the reality today seems, if anything, more remarkable than the breathless reports of the Nineties. Today, more than 2.4 billion people use the internet, according to Internet World statistics — and the numbers are still growing.
In all kinds of fields, the internet has had a profound impact. Some industries, such as music and newspapers, have been all but destroyed, even as it has created whole new lines of business, such as search engines, e-tailers and social media. Boston Consulting Group estimated recently that if the internet were a country, it would rank as the world’s fifth-largest economy. BCG analysts say internet growth will continue to outpace the BRIC’s in the short run. They expect it will climb least 10% a year through 2016.
    For India, the growth of the internet has been particularly important. “I think no other technology has transformed the nation more than the internet on three dimensions,” says Jagdish Sheth, a professor of marketing at Emory University’s Goizueta Business School in Atlanta. “First, it has made India globally integrated despite poor physical infrastructure. Second, it has reduced the rural-urban divide especially with the mobile internet. Finally, it has enabled India to have a large IT and ITeS services including [business process outsourcing] and now more highly valued professional services,” he says, adding $100 billion to the economy every year.

What Next?
Nor is the internet close to mature. Besides two-thirds of humanity that have yet to log on, new uses are being found for computer networking.
    Some analysts say that Big Data or what GE is calling the Industrial Internet — the extension of the internet to things — may have an even bigger economic impact than the internet has had in its first 30 years. New sensors on machinery and data over the internet will enable companies to see patterns in a variety of things, from public health to supply chains, and create new products as a result.
    Some of those offerings are easy to imagine, such as car insurance that is priced more competitively because premiums are based on a little black box that monitors the driver’s actual performance.
    Others might seem like the stuff of science fiction but are happening surprisingly quickly. Audi and Toyota just announced that they are demonstrating driverless car prototypes at an electronics show in Las Vegas next week, and in September, the California state legislature passed a law heavily lobbied by Google, a driverless car pioneer, that will make driverless cars street-legal.

Big Data, Big Problems?
However, Odlyzko believes Big Data may also have some risks.
    He worries the loss of privacy, which ubiquitous data collection may entail, could make it possible for companies to model prices on the basis of perfect price discrimination, meaning that you would always be forced to pay exactly what you were willing to pay — a good deal for companies, perhaps, but not necessarily for the consumer.
    “Marx says that capitalism carries the seeds of its own destruction… what I think is more likely to happen is that capitalism through the destruction of privacy and through Big Data is destroying the markets, the foundations of capitalism, and throwing us back into feudalism or, even before that, to small tribes,” says Odlyzko.
    The temptation could also be high to put thumbs on these digital scales. “You’re dealing with people who often have incentives to distort the images that are distributed through the press or social networks, or other things, and who will be influencing this process,” he adds.
Like the Libor scandal? Yes, he says, but the example he has in mind, instead, is the mortgage securities scandal that led to the 2008 crash, which was caused in part by relying on inaccurate data.
Even without bad intentions, trusting too much in models may lead to trouble, says Odlyzko, quoting British statistician George Box’s line that “all models are wrong, but some models
    are useful”.
    “There’s a beautiful area of mathematics that shows certain patterns are unavoidable as soon as you have a large enough system. You find patterns that you’re looking for, even if they’re not necessarily significant.” Mishandle data in this way and, he warns, “you can delude yourself and you can delude other people”.
    Reed says it will be important to think critically about the goals of a project, such as the use of data to pursue terrorists. Don’t let people “use the shiny technology to give credibility to something that would normally be treated with great scepticism”, he says.
    For India Inc, Big Data may present more of a challenge than an opportunity in the short run. Some pundits have warned that data-handling expertise will push the advantage back to the West and away from the developing world. However, Sheth is not so sure. “All major Indian IT services are gearing up for big data analytics,” he says. “While they may lag in the short run, they are likely to catch up.”
    The biggest bottleneck, ironically for a country rich in analytics expertise, is likely to be talent in handling data, according to Sheth. “There will be short-time talent shortage which will be a significant disadvantage. It will also require Indian IT companies to invest in the US and try to gain access to the talent. In other words, neither work can be shifted to India nor Indians can be sent to the US: they need to recruit in the US and compete for the talent,” he says.

A New Metaphor
In the end, however, the economic opportunity may not be the most important aspect of the internet. What may matter more is its usefulness as a new metaphor for human organisation.
    “The fundamental lesson of TCP/IP is that control doesn’t scale. Centralised control doesn’t scale,” says David Weinberger, a senior researcher at the Berkman Center for Internet and Society at Harvard Law School.
    Unlike a centralised system, says Reed, which may keep running well right up until the moment it doesn’t, a decentralised system like the internet will just keep running.
    “One of the things about the internet is that it’s always failing everywhere, and that’s one of the reasons it is strong: because if it is always failing everywhere in little ways and ways that can be quickly resolved, it’s never going to completely go down,” says Reed.
    At the same time, the internet’s lack of central governance, and subsequent internet-built projects such as open source software programmes such as Linux and content programmes such as Wikipedia, have demonstrated the potential value of new kinds of loose collaborations.
    “We really have demonstrated with the internet… that a looser form of collaboration where there are joint interests advanced in working together can work and scale and survive,” says Reed.
    However, the internet’s value as a democratising tool may also not just be in its structure but in its use, and not just as virtual bulletin board, as seen in the Arab Spring rebellions or the Occupy Wall Street movement.
    In his latest book, Too Big to Know: Rethinking Knowledge Now that the Facts aren’t the Facts, Experts are Everywhere, and the Smartest Person in the Room is the Room(Basic Books, 2012), Weinberger argues that rather than see knowledge as something finite that comes out of a book, the internet encourages people to look at it as something that emerges only after extended debate.
    Whether you’re arguing about a new chemical formula or trying to decide which camera to buy, Weinberger says the internet makes it clearer that there are few unarguable facts — and not having those facts makes it more difficult to maintain all kinds of elites.
    “I think we’re in a transition period here,” says Weinberger. The generation that has grown up with the Internet is looking at how decisions are made in organisations and he suspects it won’t like what it sees.
    “The notion that there is a person at the top of the pyramid who knows more about everything… that they are the best person to make the decision, looks less and less tenable,” says Weinberger.
    Governments may resist this shift, he says, but they are going to have to adjust, he predicts. If they don’t, “the disconnect between self-governance on the internet and money and power-based governance is going to become unendurable”, he warns.

BENNET VOYLES a Paris-based business writer SET1301006