Cloud computing as new IT paradigm: examining Nicholas Carr’s ideas

The idea behind cloud computing is the sharing of computing resources across many activities. For example, an operator may build a massive data center with tens of thousands of servers and then sell computing power to interested customers. Customers send their tasks through the internet and retrieve in the same way the results of calculations. At the end of the month, the operator may charge the clients to pay only the computing power they have used. The same thing may exist for data storage by giving access to companies that need the capacities of several thousand terabytes.

The great public became aware of the new technology through the bestselling book of Nicholas Carr, “The Big Switch: Rewiring the World from Edison to Google.” The switch here stays for a button linked to the coming out of electric power. Equally, the switch means a change in a computing paradigm that emulates for knowledge economy the ancient shift to electricity in the industrial world. Hence, the author's main argument is grounded on an analogy between early developments of the electrical grid system and today's application rooted in Internet and cloud computing.

The first part of the book sets the basis of the whole argumentation: an analysis of electric power developments some 100 years ago and its tremendous social outcome.

In the first chapter, “Burden's Wheel lays out,” Carr examines the past position of water power, the precursor to electricity, and then explains what these technologies economically means. Carr underlines the exceptional economic impact of general-purpose technologies – those who are the basis for many other commercial activities.

In the second chapter, “The Inventor and His Clear,” Carr reconstitutes the emergence, development, and electric power adoption. He emphasizes that electric power had a false start with Edison's request for continuous electrical current generators. Edison saw himself as a provider for local (every building) electrical generators and associated services ( maintenance, installation, etc.).

The next chapter, “Digital Millwork,” examines the modern history of computing. Here we see the analogy between old client service computing and old continuous electric power generators of Edison.  However, Tesla's generators of alternative electrical current replaced those of Edison's. The following electrical network (allowed by the alternative current) finally displaced the local continuous electric generators. Carr considers that the Internet will do to computing what the dynamo of Tesla (integrated into a grid) did by transforming local electric plants into huge electric power companies.

In this chapter, Carr complains about actual IT costs as being too big for what it delivers. The answer is the cloud computation that will bring computing power on demand exactly as the electric grid give electric power when and where it was needed.

In chapter 4, “Goodbye, Mr. Gates,” Carr paints the future world - a future of virtual computing where physical location and devices based on software licensing no longer exist (hence undermining Microsoft's core business). The comments here are almost in favor of Google and his business model of services via the Internet.

In chapter 5: “The White City,” the author moves back to a historical discussion of how electricity changed people's lives and societies.

The second part of the book underlines the hopes and the dangers of such a paradigm shift.

Electricity drove many revolutions in the years that followed the end of the XIX century, like transportation, middle-class development, and the new mass culture. Carr considers that the switch from local to cloud computing will also bring seismic shifts in business, technology, and learning and that the change is on the run.

The chapter “World Wide Computer” underlines the rampant possibilities of a programmable internet. This chapter focus on how astonishing it will be a world for individuals with infinite information and computing power available to them. 
Carr discusses equally the future of corporate computing. The basic idea is that today's IT will disappear. Until now, if someone needed computing power, he needed to buy servers and install software on them. On the contrary, cloud computing makes the hardware and software transparent. Computing power becomes a commodity that is used as the electric current that we pay to use. The equipment that generates it becomes invisible to the end-user.

The equipment will become abstract when the infrastructure consists of thousands of servers. More than that,  the failure of one server does not cause any break in service. 
That explains, according to experts, the real reason for Google's superiority over competitors. Google does not necessarily have the best search algorithms but has separate intelligent infrastructure software that gives him a tremendous competitive advantage. Many smaller companies will have access to the tools once reserved for enormous undertakings. Productivity and creativity will, therefore, increase.

Cloud computing is a step closer to post-industrial society because a client will not buy servers anymore from a manufacturer but will buy a service. 
Manufacturers like Dell, HP, or IBM will, therefore, be faced with clients that will buy servers by thousands and have enormous bargaining power. A price war will then engage, and computing power will become incredibly cheap and accessible.

Tens of thousands of IT jobs on maintenance will be destroyed. But the software development market will expand. Equally, the application software that can use this unexpected power will explode. The software will move to the SaaS model (Software as a Service), where the software will be installed on cloud computing and used through the Internet. This transition is already underway (via software such as SalesForce.com and many other software packages) but will accelerate.

The chapter “From Many to the Few” discusses the social impacts of a programmable Internet. Carr systematically unveils here the negative consequences. Fewer and fewer people will need to work in a global world of the programmable Internet, but the utopia of equality and the web's local industries will never arrive. The author also underlines the erosion of the middle class.

The chapter “The Great Unbundling” describes the move from mass markets to markets of niches. The section also debates the social implications of the web to create a tribal and increasingly sliced world rather than a unified one. The Internet is strengthening the user's ideas because it allows him to find only others with the same goals. The World Wide Web is increasingly become a zone of niches and not a universal space assumed to arise with the Internet.

In the chapter “Fighting the Net,” Carr discusses the weaknesses and vulnerabilities of the free-flowing of information and the net's structural integrity.

In “A Spider's Web,” he addresses the personal privacy issues associated with the web and the realization that "we live in a world without secrets." This chapter is equally a warning about what it means to do business where everything is recorded and tracked: "Computer systems are not, at their core, technologies of emancipation. They are technologies of control." He points out that even a decentralized cloud network can be programmed to monitor and control.

The last chapter, “iGod,” discusses a futuristic vision of fusion between men and machines.  What would be possible when the human brain can immediately access infinite information, and the tools gain Artificial Intelligence (apparently this is Google’s program)?
These are the questions raised but unanswered in the chapter. As a matter of fact, the book ends by saying that we will not know where IT is going until our children, the first generation to be “wired” from the beginning, become adults.

General evaluation of the book

The main argument is that desktop software will be displaced by Web 2.0 (peer to per and participative internet) and cloud computing. When electricity began its development, many businesses used local sources for power, including local waterwheel or windmill on their own property. As the electrical grid developed, companies were able to get fuel delivered from somewhere else. They didn't know or really care, as long as it came in. 
In the computer industry, the same transition is going on. Instead of using programs on your PC, more and more businesses are using Web 2.0 technology to host their critical mission software somewhere else, and they don't really know where or even care. The advantages of such a transition would be at least economical, and the author cites one source as saying that Google’s computation (probably the largest user of cloud computing through its distributed data centers) can do a task at one-tenth of the cost.

If some aspects are never mentioned (for example, the copyright topics), the author did not fall either in a triumphalist or utopian vision. Some of his conclusions underline the shortcoming of the new paradigm: a further decline in print publishing, a reduction in the middle class, and the continuing erosion of privacy.

But can his core analogy with the electric grid be maintained? The electrical power system is a highly regulated market system (from a strategic perspective, this is quite normal). Can the knowledge economy and the new paradigm of cloud computing be envisioned in the same way? This is a fascinating point which the author never addresses. 
How to understand the actual trend of decentralized electric power generators (for example, by solar panels alimenting electrically autonomous houses)? These new tendencies may affect the pertinence of a similar paradigm shift linked to cloud computing.

The issue of privacy can also undermine all the intellectual build up. Inevitably the high computation power plants of the future will control the data of their clients. And these privacy intrusions might concern not just lay individuals (that may eventually accommodate with it) but the economic data, business plans, commercial secrets, secret inventions, or know-how of different undertakings. 
Is it possible to give all those highly sensitive information to a giant and an almost unique player like Google? Can we trade efficiency with the lack of independence? The answer seems to be negative. But these issues are likely to incite a debate of challenges and possible technical or juridical remedies. The future will decide on the balance between efficiency and privacy, independence, and monopolistic control.

However, one can reasonably agree with the author that ‘cloud computing’ and the pervasive network will be essential in shaping the future landscape of our knowledge economy and society.

Tweet