A friend asked me yesterday if we should be more worried than we are now about the prudence of entrusting so much of our personal data, our business assets, and to some degree our livelihoods to "the cloud." My immediate answer (I always keep immediate answers on hand) is that cloud architecture is nothing to be feared. In fact, the more we understand cloud architecture, the more we may be able to apply it to solving some serious, fundamental architectural issues that face us this moment -- as demonstrated by this week's colossal failure of non-cloud architecture, the BlackBerry e-mail outage.
It occurred to me only later, as we were driving through an intersection that is my hometown's version of Russian Roulette, that if we had as much concern about "the Web" 22 years ago as we do about "the cloud" today, "the Web" wouldn't be the overblown excuse for an electronic flea circus that it has become.
If we peel off the marketing moniker, "the cloud" is a set of emerging, surprisingly sensible, architectural advancements to computing. Up until now, or thereabouts, resources in computing were under the central control of something, which we typically call the operating system. When a resource failed, as eternally spinning hard drives still tend to do, it was the system's job to prevent the system from failing. And its batting average was something less than 1000.
In a large array that only scales up, those failure rates multiply. Big computing systems thus became prone to failure. Keep in mind that the Internet's principal purpose was as a failure management system, a way to route connections around points that could no longer receive them. In logistics planning, the greatest ideas are those which accept failure as a fact of life, and steer around it to avoid derailment and catastrophe.
The decentralization principle in an array of nutshells
Cloud architecture, despite its poor choice of metaphor, is a failure management system. It decentralizes control of resources and de-emphasizes their integrity as singular processors and volumes, in favor of a pooling system that distributes workloads and ensures the existence of data in case of (inevitable) failure. It inverts the formula. In a large array that only scales up, failure rates divide. The numerator becomes the denominator.
If only Congress worked this way. Think about it: What better example is there in the entire world today of a system whereby a single point of failure derails the entire production process and jeopardizes the integrity of the system at large, than the United States Congress?
As a society, we worry about the wrong things. I can't recall the last time a friend, colleague, or Twitter follower (human or otherwise) asked me, "Scott, how worried should we be about entrusting the safety of this nation's system of electricity distribution to a centralized construct that can theoretically be destabilized by a single point of failure exploitable by a laid-off bank teller with a grudge?" Or, "...about tens of millions of people and non-people willingly divulging personal information (some of it theirs, some yours) to a centralized source whose intention to mine that data is a matter of public knowledge?"
Or, "...about tens of millions of motorists traveling every day over bridges whose foundations should have rotted away three decades ago?" Or, "...about a financial system where value accrues through the impending failure of participants to meet their commitments rather than their success?"
We couldn't even speak the word "infrastructure" in the context of the last jobs bill that failed in the Senate this week. The word was stripped from its language, probably for fear that it would point to the real problems that Americans haven't found the stomach to face. There are problems in front of us that we have trained our eyes not to see. And there are solutions in front of us to which we have thus become blind.
Cloud architecture plans for failures and works around them. Its success is predicated on failure as something that can be compensated for. This is 21st century architecture.
This is 18th century architecture. This diagram of the American budgetary process was compiled by the American Geological Institute, based on information supplied by the House Budget Committee. It's a workflow diagram, if you will, but it's predicated on the presumption that the current step will clear in order for the next to proceed. The single point of failure in this workflow isn't actually shown: It's the political process that attributes value to a problem only when it has been elevated to a crisis.
It's the process that prevents us from recognizing any problems as problems, no matter how great or small, until we can raise them to a level that resembles a catastrophe.
1, 2, 3, 4, we know what we're fighting for...
Amid the recent, growing series of "Occupy" protests in many of this nation's cities, a few journalists have asked folks just what it is they are protesting. PBS and NPR have some longer-form examples. Some who are without jobs, who struggle to pay their monthly (and now weekly) bills, who have lost their homes, who have left their families in hopes that they would be better off without them, have immediate answers on hand.
But that's not everyone. For quite a few of these folks, the journalists were the first to broach the question: What is it you want to change? When their responses not immediate, their intermediate responses seemed fairly fuzzy. There's the bailouts, and those are bad. There's CEOs making too much money, and when asked to name examples, you don't get many. Warren Buffett's name came up, which was not as much a surprise as Bill Gates, who's no longer a CEO.
The Huffington Post has compiled a "word cloud" of the terms that come up with respect to stories and textual posts surrounding the protests, in an earnest effort to determine just what they're about. You can't even automate the process of extracting the message from the noise; the word "rights" takes up less space than the word "corporations."
One participant compiled a common thread of texts sent by participants lucky enough to own smartphones, listing the various grievances that come to mind. It would be unfair to presume that any one answer speaks for the entire group, but unfortunately, that could be said about the entire democratic process of late. In any event, here it is in black-and-white, the underlying cause of all our ills from one fellow's perspective. Look out for the evil cloud:
Groups don't know when, but at the right time, North Korea and Syria will add their teams to the internet convergence, strengthening the effect of the DOS that wipes out our silly banker owned coin of the realm, the Chase-Banko'merica electron. No backbone, no bank, no bank, no commerce, no currency (greenbacks) in circulation, no accounting systems in the cloud, for however many days they can keep the run going. Bankers are insufficient to maintain an electronic currency security, and guaranteeing the success of the nations security that the currency represents. The government is required to be the backer of the currency. Its time they moved into the 21th century and secured our nations new electronic system of everyday trade, by nationalizing it and putting it under the Dept. of the Treasury. If anyone is going to make a percentage tax off of every dollar spent in common trade, it should be the US Government,(that is the ONLY agency that is allowed to lay a general fee of such magnitude) and be used to run the other stuff.. The bankers have appropriated a governments function, and it was not recognised at the time. By the right people... ((It makes sense if you don't think about it...)) weschrist
The reset button
There is a growing body of potentially influential people who have come to believe, and to an increasing degree preach, the notion that the Internet is society. Not a medium of communication, but a mode of existence overseen by ICANN, whose constitution was written by W3C and whose local territories are policed by GoDaddy. The inalienable rights of citizens to freely copy software becomes the cause of political parties and campaign funds; and the cloud - the structure that grows in unseen increments behind the cloaked firewalls of corporations - becomes the focus of evil in the modern world.
The winner of the last presidential election in this country spoke of the coming of change. We expect change to emerge from the clouds (the real ones, not the metaphorical one) like Underdog to the rescue, wearing cheerful colors and emanating hopeful rhymes - maybe from America, or maybe in this case from the ever-popular tag-team of North Korea and Syria. We attribute the face of evil to the wrong things.
In the days of the civil rights movement, folks who came together in peaceful protest would find themselves suddenly singing songs, which then became anthems for a cause that cleansed this country. Movements require direction, otherwise they're not movements at all. I sometimes think if the current clusters of protesters were to start spontaneously singing, what we'd hear would be as unintelligible as the excerpt above.
Every so often, a new technology gives us the opportunity to perceive ourselves in a different perspective. I know you may think I've changed topics to the iPhone, but I haven't. There is a sensibility to "cloud architecture" that can be leveraged for an even greater good, enabling us to reconsider how we organize and why. It is past time for us to consider the question of whether, in the emerging system we like to call the "Digital Economy," we, the people are in danger of becoming the single point of failure.