Reflections on Internet/Information Policy

Global War for Internet Governance is a great primer on the architecture of the internet and a balanced view of the policy challenges for internet governance. I particularly enjoyed the section on net neutrality and the discussion of the governance structures for regulating it. The author of the book is both an engineer and a policy wonk, which allows him to explain the challenges of net neutrality in a way that  is accessible and comprehensive. Before you read any further, check out this Daily Show clip for a refresher on net neutrality, or if you don’t have time, read below.

Developments in processing power now allow for internet providers to check traffic moving through their cables for its contents, size, and origin. This means that internet providers can charge heavy data users more, restrict access to certain types of content, and charge websites that use more data. This seems like an economically sensible idea: if you use a company’s infrastructure more then you should pay more for it. But, there are plenty of reasons to distrust this model and support net neutrality, besides your innate distrust and for your local internet provider.

Because internet providers make an economic argument against net neutrality, it is important to address the economics of these companies. Internet companies require huge upfront costs to lay down fiber optic cable and then recoup the costs with high (often monopoly) profits later on. Some of these companies only spend 12% of their operating costs on maintaining and updating these systems. The high costs of starting an internet provision business make it so that new entrants are rare. Internet providers try to limit their costs by merging with other companies which reduces their costs and increases their profits. This practice of merging with other companies makes anti-trust supporters nervous, and rightly so considering internet service providers history of monopolies. In 2002 the FCC classified internet provision as an information service which deregulated the industry. Regulators thought this would increase competition but instead found that service providers stayed out of each other territories in order to increase their profits. This practice, obviously, had a negative effect on internet service for Americans. 

Although, in general, I’m in favor of net neutrality, and I’ll explain in shortly what net neutrality means to me, I was always bothered by its supporter’s argument that all internet traffic should be considered equal. DeNardis covered these concerns. First, deep within the internet backbone there are legitimate technical reasons for treating different traffic, especially for administrative traffic and to guarantee quality-of-service. Second, the limited and shared bandwidth of cellular-based internet connections give even more reason to differentiate the last-mile traffic to guarantee fair usage of the shared medium. I’ll give a concrete example to explain these points. A live video stream and a cellular phone conversation are not equal: the video stream requires significantly more bandwidth and is thus more “expensive”. On a bandwidth-limited connection, if an internet service provider (ISP) gives equal opportunity for both the video or audio conversations, the video stream would mean sacrificing the audio call quality for others sharing the connection, who are using a fraction of the bandwidth of the video stream. Not only that, the video stream would require more processing and more power consumption of the ISPs routers. So it’s reasonable to require the video stream user to pay more than the audio-only user. This is especially true on a cellular network were call quality, and priority of calls over other traffic is part of the service-level agreement (SLA).

My view is that net neutrality means that an internet backbone or an ISP (and routers in between, if any) should be mostly neutral to the source, destination, and content of internet traffic. Exceptions can be made to neutrality between different sources and different destinations if the bandwidth usage exceeds that of other users. In practice, for back-bone connection’s bandwidth is effectively unlimited (or at least the marginal cost of adding more capacity is trivial compared to fixed costs), and the same is true for newer, especially fiber-optic, last-mile connections. So regulations to limit monopoly profits for ISPs that want to extract a lot more money than the marginal cost for high-bandwidth usage customers do make sense.

The internet can undermine or strengthen the authority of a regime in a multidimensional fashion, however it is often viewed as a subversive force. As The Global War for Internet Governance highlighted, systems of technological control such as the internet are layered, which means there exists many different focal points of power through which market and governmental entities can exert influence. One key component of internet regulation and governance is access. Access can include the ability to possess a laptop with an internet connection, being able to afford the per-minute fee at an internet cafe, or belonging to a municipal broadband network. With increased access to the internet, citizens are endowed with the ability to develop a new realm, a new modality of participation in civil society. Participation in online civil society can take place in a chatroom/web forum, or working as a journalist for a web-based news organization, such as Wikileaks.

Governments, viewing the implementation of new internet communication technology as inevitable, have implemented mechanisms limiting the internet’s capacity to influence the political ideology of its citizens. In a piece from the Carnegie Endowment for International Peace titled “The Internet and State Control in Authoritarian Regimes: China, Cuba, and the Counterrevolution,” authors Shanthi Kalathil and Taylor C. Boas assert states can respond to this potentially harmful problem reactively or proactively, and they often use internet technology to do so.

Reactive measures to control the internet include:

limiting the diffusion of internet technology,

censorship,

mass surveillance of internet activity,

the complete proscription of the internet.

On the other side of control, proactive actions can incorporate:

the dissemination of propaganda globally,

the creation of a national internet that is separate from the worldwide web,

the formation of state-sponsored internet services that positively influence the populace’s opinion of the government,

or even engage with other countries internationally through cyber warfare.

These proactive strategies reveal a pattern: states that have particularly strong involvement in the development of their internet often have the greatest degree of stability. Our book this week was published in 2014, but in 2001 Kalathil understood how technology impacts society, with or without state control. The author’s conclusions regarding her research reveal a shrewd foresight into the potentialities of the internet, saying “As access to the Internet increases or other intervening variables evolve over time, Internet use may indeed play a greater role in challenging authoritarian regimes - perhaps even contributing to their eventual demise.” If only she could have known about the Arab Spring, the political revolution largely known for the apparent use of social media services like facebook and Twitter to organize protests and assemblies, particularly in Egypt and Tunisia.

To hear more about how the diffusion of internet technology influences the arena of international relations, watch this TED Talk by David Rothkopf, CEO and editor of Foreign Policy Magazine and foreignpolicy.com.

https://www.youtube.com/watch?v=NA7krbsdXFA

I’ve summarized his potential causes for concern he details in the above video. I’d love to hear what you guys think about these possibilities.

When 80% of the world’s population has a smartphone, as is predicted to happen by 2020, nearly every human being will be a part of a man-made system for the first time in history. The implications of this has yet to be realized, and governments are largely ignoring what may come.

The low economic cost associated with cyber warfare may lead us into a period Rothkopf calls a "Cool War,” in which no doctrine exists shaping how decisions and responses should be made, and digital conflict wages unceasingly.

Cyber-nationalism vs. cyber-internationalism - evinced in Trans-Pacific Partnership negotiations regarding global internet regulation, states are weary about NSA surveillance programs and want to protect the privacy of their citizens. As a result, internet connectivity across countries can be limited, as nationalized intranets or large firewalls are preferred to global internet access.

The United States no longer appreciates or encourages technological development, especially with the dissolution of the Office of Technological Assessment. The September 11th terrorist attacks engendered a paranoid myopia in the United States government, and the idea of “national security” became a top priority.

I agree that the reactive and proactive measures you describe limit the freedoms of citizens of authoritarian regimes. However, I don’t think there’s much we can directly, and successfully, change about it. We’ve seen it over and over, direct interference with other nations “development” does more harm than good. The best we can do it to be good role models.

There are two ideas that I wanted to take from Bartlett’s The Dark Net. What is the value of the dark net and what is the impact of the dissociative effect of the internet. Bartlett paints a pretty bleak picture of the dark net, where people can find and do anything they want, much of which we’ve banned in society. I think the question we must ask is, what is the value of the dark net? Or, does the dark net need to have value? Some would argue that the dark net is the last place where people can be as free as they want without some sort of authority controlling them. With what we have discussed about surveillance and tracking, it is not hard to understand why individuals would want to go to the deeper parts of the web in search of privacy. And as Barlett argues, it is the last hope of the internet being a libertarian dream world of complete freedom. Sadly, the freedom is often used to do some pretty dark and disturbing things. But there have always been back alley, black markets, shady parts of town, where people can find the same things your find on the dark web. The web, just makes it a whole lot easier.

The second point I’d like to discuss is the idea of the internet creating a dissociative effect on people. Barlett’s explains “the idea that a screen allows you to disassociate your real self from your online behavior, to create fictitious identities and alternative realities, in which social restrictions, responsibilities and norms do not apply: as if the online space is somehow separate and different. (121)”  This dissociative effect appears to bring out the worst in many people. What gives a troll the motivation to doxx someone online (see pg 16 for explanation of doxxing). I can’t imagine they would interact with a real person in the same way in the real world, but when online they can be a different person. The Dark net makes it really easy for people to do really bad things. It is simple to buy drugs online, so simple that it may make you forget that what you are doing is illegal. We’d all like to think that we are moral beings, but as Barlett highlights, many of the people caught doing terrible things on the web, did not see themselves as amoral, but rather they had disconnected their acts on the dark web to the real world. Like I said before, the dark web as place to engage in illegal activities isn’t new to society, but what makes it different is the ability for the internet to disconnect us from the act that we are committing.

This all connects to HBO’s new show, Westworld. The show is about a man-made world that has been created for people’s enjoyment, full of life-like robots and endless storylines to entertain the guests. Visitors are transported to the wild west and can do anything they want. In essence, it is the virtual reality version of the dark net. It gives people the ability to dissociate from their normal selves and do whatever they want in a world that has no consequences. The same way in which people today can go onto the dark web and become a different person and still convince themselves that the person they are outside of the internet is not the same as the one that is online.

I was a big fan of this week’s readings mostly because it relates to a lot of the research that I am doing on my own about anonymity, cybercultures, and trolling. 

As I was reading through this stuff, and as I have been doing other research for other courses, I have been trying to keep a tally, of sorts, that lists everything people say is useful about anonymous networks. So let’s look at some of the benefits that Bartlett brings up (and feel free to respond and add to my list anything that you picked up from Bartlett or others):

innovation

subverting or circumventing powerful (sometimes read authoritarian) regimes

increased privacy

increased freedom

increased liberties

???

PROFIT

(Yes, I have adapted this meme for a different setting. I know this is not proper implementation.) This, obviously not exhaustive, list seems to line up pretty well with the list I already have from other people. The point is always made that individuals repressed by authoritarian regimes can use anonymous networks as a means to protest, circumvent unjust laws, and so on. This is a totally fair argument, but I think the others are a bit spurious. The idea that anonymous networks are inherently offer liberties and foster greater innovation seems to not really be inherent at all. In fact, a lot of the innovation Bartlett points to is only relative to the fact that sites are evading being taken down or getting caught. If the sites (or, more specifically, some of the actions on them) were not being investigated or if they were not working against the threat of being taken down, would we still see them as innovative? If, following Bartlett’s logic, the dark web starts to be accepted and normalized as more people and companies move there, won’t that slow this, supposed, inherent innovation?  Regarding the idea of more liberties, I think this is a weak argument too. Not because I disagree, but because I find it short sighted. Of course these anarchic communities are going to offer more liberties. There is only liberty there. But when humans organize in groups and cooperate, they create certain policies which constrain certain liberties. A reputation economy is all well and good, but we have seen the problems that creates with cases like Uber and passenger discrimination or just general conflicts between that kind of system and access. The liberties that are made most apparent in these anonymous networks are, more often than not, those which irl societies have agreed are, for various reasons, criminal, unnecessary, or unjust - hate speech, human trafficking, buying and selling certain narcotics, etc. 

The old web is often seen as the utopia for cyberlibertarians, and to some extent it was, but it was also only that way for a very specific set of people. As open-sourced as Tor is, the so called free space is, practically speaking, still only available to certain types of people too, and the inequalities of our physical world are only made more apparent in these spaces under the guise of increased freedom and liberty. 

Cyber War: The Next Threat to National Security and What to Do About It, written by Richard A. Clarke and Robert K. Knake, is a fast and enjoyable read and a good, if somewhat outdated, primer on cyber issues. Unfortunately, however, the authors engage in a fair share of fear-mongering and hyperbole about potential cyber attacks. Clarke (who also writes fictional thrillers and perhaps let his vivid imagination run a little too wild) and Knake regurgitate debunked hacker horror stories, giving especially the first part of the book a prophet-of-doom-like character. More disconcerting for the critical educated reader, perhaps, is the book’s complete lack of reference material, footnotes, or even an index. This stands in stark contrast to the 100+ (!) pages of well-sourced endnotes that in last week’s Data and Goliath reading by Bruce Schneier. Leaving these shortcomings aside, I’d like to comment on a few important aspects raised by the authors of Cyber War.

First, the authors define cyber war as “actions by a nation-state to penetrate another nation’s computers or networks for the purpose of causing damage or disruption.” This immediately brings up the question of non-state actors. Can’t a non-state entity engage in warfare? While certain forms of offensive cyber actions, such as large-scale distributed denial-of-service (DDoS) attacks, require substantial computing power, other forms of causing damage or disruption don’t. In the first presidential debate, even Donald Trump admitted that we cannot be sure whether specific cyber operations could be attributed to the Russian Federation or to a 400-pound hacker sitting on his bed. Joking aside, if only actions by a nation-state can constitute warfare, why did the U.S. bomb Afghanistan post-9/11? It seems like a broader definition is warranted.

Second, is ‘warfare’ even an appropriate term in a cyber context? Which actions in cyberspace should qualify as acts of war and who decides? Can activity in cyberspace have kinetic consequences such as property damage and loss of lives? We’ve seen clear cases of the former, most notably  the Stuxnet worm that reportedly sabotaged and physically destroyed centrifuges at Iran’s Natanz uranium nuclear facilities. After Stuxnet became public, former CIA director Michael Hayden famously said “somebody crossed the Rubicon.” There is no doubt that the U.S. and many other countries face cyber espionage (stealing secrets) and cyber sabotage (messing with processes/equipment) against media organizations, government entities and corporations. However, despite the doomsday hype, so far, there is no documented case of a cyber war-related human fatality. Apart from the technical and legal questions on what constitutes an ‘act of cyber war’, there is the political question of whether the U.S. even wants to define it clearly, especially since that it has an interest in using its offensive cyber capabilities against adversaries as well. Long shrouded in secrecy and done in the covert realm, U.S. cyber operations are increasingly being carried out through Cyber Command (Cybercom) and are now becoming more public. The U.S. thus faces a tricky balancing act between using its vast offensive capabilities and inviting (legitimate?) retaliation.

Third, one of most compelling and thought-provoking discussions in Cyber War is the asymmetric nature of the cyber threat. Similar to the fight against terrorism, the defender has to find and eliminate all exploitable vulnerabilities; the attacker only needs to find one! A highly IT-dependent and interconnected country like the U.S. is much more vulnerable to cyber-attacks than a more isolated and closed country such as North Korea. With millions and millions of systems (civilian, military, commercial, etc.) online, the U.S. faces a staggering number of potential vulnerabilities. The fact of the matter is that perfect security is unachievable in any useful system. Interestingly, the inherent vulnerability of what we today call the cyber realm was pointed out already in 1967, before ARPANET, the precursor to the internet, was even created. 

First, I completely agree with you that a broader definition of cyber war is warranted.

Second, I disagree with the implication that because “so far, there is no documented case of a cyber war-related human fatality” the thread of cyber war is diminished. I don’t think the use of atomic bombs and loss of life necessary to reach the current equilibrium (this is separate from whether it was necessary to end WWII). I think the threat is legitimate, but it’s probably kept at bay because the major players understand the threat. However, the complexity of attributing attacks does make the threat different from conventional warfare.

Third, as you mention, the asymmetric nature of the cyber threat is interesting. I agree that “the fact of the matter is that perfect security is unachievable in any useful system,” but we can get near-perfect security for highly-critical systems.

Clark’s Cyberwar and the recent political circus, has made be think a lot about the role of cyber warfare on our international political system. This is the first presidential political contest we have seen where the hacking of private information has been so prevalent. While we can’t say for sure that is Russia or China is responsible for hacking the DNC, it is safe to say there is foreign entity that is intentionally trying to change the outcome of the election. Of course, foreign governments have been trying to impact the elections of one another for centuries, and it is something the US has been a big fan of. Here is a nice rundown by the huffington post - The U.S. Has Been Meddling In Other Countries’ Elections For A Century. It Doesn’t Feel Good.

But the level we see it at today is concerning. And maybe what is more shocking, is that the public does not seem to care very much that it is happening.

Let’s take the leaking of DNC emails as an example. Instead of viewing this as an attack on American sovereignty by a foreign entity, the real conversation is about the information that is released. People within the DNC were fired, it impacted the Democratic National Convention, and maybe affected the polls for that week. But there was pretty limited public outrage towards the perpetrator. The harm of the attack is always felt by the victim, and blame is almost never placed on the attacker. We even had one candidate call for Russia to hack Clinton and her emails. But our focus was on the content of the attack and not the attacker. Maybe it is because we are generally interested in the information that is released, or because it is really difficult to actually put the blame on someone (I think at least, maybe the CIA, NSA, FBI, know who is doing it and is not telling us) This makes cyberwar a uniquely powerful tool. The United States can be attacked by a foreign actor, and there is little conversation from the public. We actual indirectly encourage the attacks by the way we react to the information leaks.

To quote a review from the Wall Street Journal, Data and Goliath is “unable to make intelligent trade-offs among individual privacy, national security and business efficiency. ” If I didn’t have to write this assignment, I might just leave it right there, but I’ll elaborate. The book’s large scope makes it a good overview of the subject but the author doesn’t dive in deep enough into any one topic to give the reader an understanding of the cost-benefit of losing privacy. This lack of focus makes the book sound kind of like a conspiracy theorist who saves all his emails on his own personal server (which he does) ranting about privacy.

He does talk about the benefits some applications give us in return for relinquishing our privacy. He mentions traffic analysis and benefits to urban planning from GPS location services but never tries to reconcile these benefits with some of the risks. Performing a cost-benefit analysis on privacy would be a difficult task, but it seems like the easy way out to not give a fair description and equal time to both sides of the argument. This inherent bias towards privacy is particularly clear in his short discussion on video surveillance.

The author’s first mention of video surveillance comes when he writes about how annoyed he is that there is public video surveillance. Without a hint of self-reflection, he then talks about how video surveillance enabled the capture of the Boston marathon bombers and London subway bombers. He then fails to seriously reconcile the costs and benefits of between his annoyance and the catching of terrorists .  There have been benefits to mass CCTV systems and Britain has seen a drop in property crime possibly due to their installation. Futuristic blimp technology and license plate tracking with video surveillance are also covered in the book but the author doesn’t give either method enough time to understand the risks and benefits of the technologies. Every technology has different implications and it is unfair to label all technologies that violate any amount of privacy as a part of a slippery slope that should never be touched. 

The cutting edge of camera surveillance has moved much further since the release of this book and complete recreations of movements of people and objects in a city can be now be recorded . This technology was developed in Iraq and was tremendously successful in reducing the number of roadside bomb attacks. The footage taken from a plane allows for analysts to “wind back time” to identify people and cars that were in an area where a bomb exploded and then look to see where they came from, who they were with, and where they live. If first heard about this technology in a Radio Lab podcast that is worth a listen. This surveillance technology did violate people’s privacy but it also saved soldiers lives in Iraq and was implemented in Mexico to take down a major cartel. This technology is spooky for sure, but would you give up a little privacy to save a soldier’s life? 

Having said that, I think Schneier’s opinion is very clear that we have too much surveillance. I also agree that the cost of surveillance outweighs the benefits. However, as I discussed in my post, the policy challenge comes from the fact that different people have wildly varying perceived costs and benefits of surveillance. If everyone reads this book people may start to see the additional cost of surveillance, but I doubt we would have a consistent answer as to how much surveillance we want.

The hack of Sony by the Guardians of Peace (operating on behalf of the North Korean government). The breach of the Office of Personnel Management by [agents of] the Chinese government. The release of DNC emails by WikiLeaks. The hack of the DNC and other campaigns, which the Obama Administration has just accused the Russian government of today. 

These are a just a few examples of some of the high-profile attacks that have occurred over the last several years. The recent attack on the DNC is perhaps the most troubling. I don’t think we’ve seen such clear evidence of a foreign government attempting to influence the result of an election in another country.

In one of our previous classes, someone applied the logic of comparing actions before and after the widespread use of technology and the internet. For example, could we do things, like conducting surveillance on individuals, before technology? Has technology completely changed this, or is it simply a new mechanism to do the same thing? 

Admittedly, technology has completely changed how we conduct espionage and surveillance. We can do many things that we could not previously do. But I would argue that this is a somewhat necessary step. The reality is that we are fighting asymmetrical actors – sometimes acting on behalf of governments and sometimes acting on their own accord. State actors are currently not the biggest threat – although it is very possible that tensions with Russia and China could escalate in the near future. 

America and other countries are facing asymmetrical threats from non-state actors, violent extremist organizations, and lone wolves. And citizens have a certain expectation that their governments will keep them safe. We’re generally ok with surveillance of foreign governments, but not of the domestic population – which Schneier draws a clear distinction between. But the reality is that the biggest threats may be from homegrown terrorists in isolated shootings and attacks. How are law enforcement agencies supposed to prevent such attacks without at least some level of surveillance of its own population?   

We’ve talked in the past about cyberspace being the new Wild Wild West, but Schneier’s characterization of the intelligence world certainly makes it seem like the true Wild Wild West. Governments are spying on each other in a multitude of ways, and they are spying on their own populations. They are spying on their adversaries and their allies. The U.S. intelligence community is composed of 17 different intelligence communities. There are various international partnerships, such as the Five Eyes, Nine Eyes, and (not to be redundant) the Fourteen Eyes.

This goes back to the fundamental tension between security and convenience that I’ve talked about before. But in this case, the interplay between these two is very interesting. The mass collection of data and centralization of surveillance are meant to make securing the population easier, and more convenient, for governments. However, the very nature of centralization makes us less secure. Specifically, if all our data is centralized it becomes more vulnerable to attacks by foreign governments and criminals. Further, we’re not just proposing centralizing data, we’re also going to centralized real-time surveillance (e.g. street cameras in London and reconnaissance satellites). As @michael-pearson the hack of DNC is a great example of this. What if a foreign government hacks into the NSA or the DOD’s reconnaissance satellites? What if a criminal organization hacks into the London camera system to aid it in getting rid of its enemies?

Correlation does not equal causation. Big Data: A Revolution…spends a whole chapter showing how correlations within big data are now the hard and fast rule in retail, health, and government. I feel like this chapter was missing the rule of “correlation doesn’t always equal causation”, often pushed in the sciences. Big data is often pushed as the only way to gather information, and the book even has pushed so far as to say the scientific method and hypotheses are no longer strong forms of research. It bothers me that Mayer-Schonberger is making the case that predictability and probability will replace structured methods such as methodical experiments. Not everything needs to be rushed and “fast thinking.” I can understand Walmart’s use of correlations between hurricanes and selling more Pop-tarts, but once my health insurance starts rising because they think I watch Netflix for hours, then I have something to be worried about.

Whereas personalized advertisements may feel creepy, in the end they are meant to benefit us as consumers. I am more concerned about another point that @jmraye​ brought up:

but once my health insurance starts rising because they think I watch Netflix for hours, then I have something to be worried about.

Again, I think competition comes to the rescue. It may very well be that watching Netflix for hours increases my odds of bad health outcomes (but I’ll keep the debate about correlation vs causation for another post), in which case I am fine with a slight increase in my insurance premium. For every dumb insurance company that unreasonably raises my rate because I watch Netflix for hours, there will be a smarter company that analyzes the data better and only slightly raises the rates.

This week we explored the “mystical” world of big data. The first question I asked myself, was what big data really means. While the book gave a relatively complicated definition, I gathered that big data is really just a way to aggregate lots and lots of unstructured data and then make sense of it. Really, the concept is not new – what has changed is the technology that enables us to make sense of huge amounts of data. In Big Data: A Revolution that Will Transform How We Live, Think, and Work, Viktor Mayer-Schönberger and Kenneth Cukier introduce us to a fantastical world, where businesses use data to make better products and services that fulfil needs we didn’t even know we had. Using big data, Walmart will never run out of pop tarts before a hurricane and Netflix will suggest the next movie or tv show we should watch based on our viewing history.  

Mostly, the book deals with how companies can use data garnered from their users. The authors call the process of capturing information about the world “datafication”. This process is opening our eyes to information sources we have not previously thought of but can be of great use. They advance uses such as IBM’s touch-sensitive floor surfaces which are able to identify us by our weight and the distribution of it. By using huge amounts of data, the authors advocate that we are freeing ourselves from small sample data, which can be unrepresentative. However, the authors fail to point out that a digital divide exists in our societies, where not everyone has access to digital services, let alone a computer.

Mayer-Schönerger and Cukier also tell us that we should drop our “obsession” with causation. Instead we should use big data to establish correlations between things we never thought would be correlated, such as the increased sales of pop tarts before hurricanes. They advocate for an approach that looks more at the what than at the why. While the reasons behind phenomena might be obscure to us, the fact that correlations exist in the data should be enough to base decisions on it. That is a dangerous thing to do, especially in public policy.

As President Bartlet taught us in the West Wing episode “Post Hoc, Ergo Propter Hoc” (btw, that’s episode 2 of season 1 if you want to watch it on Netflix), correlation does NOT imply causation. That may not be a great problem for companies, since they are more interested in the fact that pop tarts are more popular before hurricanes rather than why they are more popular. In public policy, however, understanding the why is really important, otherwise we might make wrong decisions with potentially nefarious consequences. This website gives a great overview of weird correlations that exist in data but can hardly be said to be connected. For instance, US spending on science, space and technology is positively correlated with suicides by hanging, strangulation and suffocation. Now, this is an extreme example, but if we took it to mean there is a clear connection, then we might assume that reducing US spending on science would reduce suicides (!).

Approaching problems by first looking for correlations and then causality might also help avoid the common pitfalls of assigning false causality.   First, you have to identify variables that exhibit some interdependent relationship. Only after that step, can you investigate whether the relationship is causal.  

Furthermore, a correlation-first approach may be a good for another reason. It eliminates the biases and pre-judgements we tend to have when trying to analyze and solve problems, especially policy ones.

I’ll also add that that as “Governance in the Information Era” explains, in addition to the study of causation we need to understand the unintended consequences of policy perhaps by using computer models, perhaps driven by big data, to understand the consequences of policy.

Forgive me for the impending geek-out, but did Interop make anyone else think about Battlestar Galactica? Just me? Okay, I’ll explain:

Authors John Palfrey and Urs Gasser explore how technology is becoming increasingly standardized and interconnected in their book Interop: The Promise and Perils of Highly Connected Systems. They argue that a stable theory of interopality is needed, especially as policymakers try to find a balance between its costs and benefits. If you don’t quite get the concept, just think of using your Facebook account as proxy when creating a new account on another site – this is a form of interoping. The benefit is that using an intermediary system (Facebook) improves the convenience of using another system (Pinterest, for example) for the first time. But, that comes with a trade-off: another site now has access to your Facebook information, increasing your security risk. This is just one example of interoping, but instances of interoping are infinite given how the Internet operates. Most forms are fairly innocuous in-and-of themselves. But, on a larger scale, these issues are much more pressing.

Now, back to Battlestar Galactica…

In this science fiction universe, humans had long been in a war against a race of cylons that they created. Bent on human extermination, the cylons launched a cyber-attack that set off humans’ nuclear weapons. This wiped out the Twelve Colonies of Man and their military vessels, as the ships were connected to the military computer network, which was also cyber-attacked. But, the Galactica, an aircraft carrier that fought in previous battles but was being decommissioned, had never been linked into the network. As a result, it was unaffected by the cylon attack. The principle of this extensive plot summary is that if the ship had been interoped with the network, we would have never gotten 14 books, 23 comics, 2 films, and 5 television series about Galactica. And, that would have been a tragedy. But, it also illustrates that interoping comes with inherent risks (like the extinction of mankind, for instance).

The challenge for policymakers, then, is to assess when more interloping is in the public interest and when it is not. This feat is particularly tenuous given that, as Palfrey and Gasser point out: “there is no single form or optimal amount of interoperability that will suit every circumstance.” Therefore, “most of the specifics of how to bring interop about [must] be determined on a case-by-case basis.” To achieve this, they recommend collaboration between the public and private sector. Although many of the concepts that they introduce, such as “optimum interoperability”, are fairly amorphous, they still argue heartily that policy is needed to manage risks related to interoping. Palfrey and Gasser note, “the law can play a central role in ensuring that we get as close as possible to optimal levels of interoperability in complex systems.”

As I’ve argued before, there’s always tension between privacy, security and convenience. If we want to have the convenience of using the internet and the tools and technologies built around it, and especially if these tools and technologies are interoperable, we have to accept that security is not guaranteed. There’s nothing that government, policy or even private companies can do about it. They can try to minimize the potential for attack but they can never prevent it. It’s the same problem as in the physical world: you can probably live your life in a hidden underground bunker for maximum security, but how convenient is that?

The Collins English Dictionary defines private property as “land or belongings owned by a person or group and kept for their exclusive use” and William Blackstone, the influential author of the Commentaries on the Laws of England wrote that ownership is the “[s]ole and despotic dominion which one man claims and exercises over the external things of the world, in total exclusion of the right of any other individual in the universe.” In our digital age, interoperability plays an increasingly important role in safeguarding our private property rights. In Interop: The Promise and Perils of Highly Interconnected Systems, John Palfrey and Urs Gasser write about the dangers of a lack of interoperability. Without interoperability, consumers are at risk from being locked-in into systems because companies actively prevent interoperability. Consider, for example, how difficult it can be to switch from an iPhone to an Android phone. Apple has created a closed ecosystem from which it is very difficult to transition to another. Apps you buy cannot be transferred, iTunes music may be DRM (Digital Rights Management) protected and your settings will not be transferable. As such, our ownership about our personal data and property (e.g. apps) is being restricted by private companies in their own interest.

Palfrey and Gasser find that interoperability leads to more consumer empowerment through increased user choice. Indeed if one is not locked-in, then it is easy to switch between systems. Interoperability also enables more competition between companies. Especially in the internet age, new entrants without any way to access existing user information are at a disadvantage. However, if all companies were forced to create interoperable systems, then new entrants would have better chances of succeeding and competition would foster greater innovation. Yet when interoperability is not given, customers are forced to stick with inferior systems due to the high switching costs. In an internal memo, a senior manager at Microsoft once wrote: “It is this switching cost that has given the customers the patience to stick with Windows through all our mistakes, our buggy drivers, our high TCO (total cost of ownership), our lack of a sexy vision at times, and many other difficulties […] Customers constantly evaluate other desktop platforms, [but] it would be so much work to move over that they hope we just improve Windows rather than force them to move.”

Also, is it really true that we have “despotic” dominion over our property? You can modify your car so that it’s not street-legal anymore. Sure, you can keep it in your garage, but how useful is it to you now as property. Same thing with land. There are restrictions on what you can do with it – you’re restricted by local zoning laws all the way to federal laws that criminalize certain activities. Point is, we don’t have total control over our property anyway. Further, what keeps you from losing your property? It’s not the fact that it’s physically there and you have guns to protect it. It’s that we have public policy, and a government that can enforce the policy, that explicitly gives you certain rights.

If it’s the job of government to protect your rights, wouldn’t it make sense for us to have public policy that requires interoperability of digital content, such that our property rights are preserved. Also, government can protect our rights even when corporations file for bankruptcy (e.g. FDIC-like agency for digital content as an independent agency as I proposed here).