For quite some time China has been blocking access to Facebook from Chinese networks, but recently there has been rumors that the ban was lifted in some areas, and that Facebook was renting office space in Beijing. Some people say that this is because China wants to promote its own social network sites, or that the sites they block are harmful for the people and country. It is very difficult to find argument which would be satisfactory, and all of them seem to be partial and some even surreal. Maybe the only thing is that politics of power in the cyber is a rapidly evolving subject and contrast to that most governments are not used to it. And then the main question is that will they ever be?
When Bloomberg writes in the headlines that “consumers” want to have Facebook, but that Chinese regulators refuses, its like two completely different viewpoints put against each other not discussing on the same terms. Will this argue ever find a conclusion?
Dailymail Blocked in Thai Network
Similar to this event is the recent Twitter ban in Turkey, or the ban of Dailymail UK newspaper in Thailand. The list could be longer and certainly will be in the future. The main problem is that the banning parties are bound to their historical concept of territoriality and means to secure it by deploying the monopoly of physical power. However the cyberspace is by nature borderless, even when you can imagine some artificial borders such as country level domain name systems, which are dissolving now, or some national routers, which are only partial by their scope and violate the concept of freedom of packets to travel around or peoples freedom to choose their location or communication devices. Hence in essence the blocking mechanisms are trying to re-establish territorial control and identity structures for the times in which people are more and more distracting themselves from the old concepts of territorial stability. This all has happened party because of the freedom of movement, easy of travel and of course networking, but mainly it is because the times are changing and no governance system is static, fixed and eternal. So, its tried to be argued here, that the issue to consider at the social media bans, as any other web bans, is more of the act of trying to re-vitalize the nationalism in the way of enforcing the binding of the individual to the cyber-state, than just about the ban itself or its legality or practical arrangements.
The world of social media applications is wide and deep, and as so many other phenomena of modern performances they are also considered from somewhat limited angle. In order to understand and approach the question on what kind of spheres of influences the social media applications build and carry with, one would need to understand the history of the web development and something about the underlying technical innovations which enabled the massive expansion of the number of users for any particular web application.
For the massive scale of users, such as Facebook or Twitter might have, not any single web server could do the job. So you would need many of them, and some way to keep them in synchronized. But that is not enough, the underlying database should also be distributed across the continents, as there are so many users posting their updates all the time.
This article tries to bring light to the question of the technical background of the social media applications, and at the same time evaluates and considers the penetration level and mechanics of them. Why certain social media apps announce as they are global, but then there reality is not so much so? And from the other angle, why some of them try to restrict access on using them to only a certain group of people or society?
Fear of Invasion?
In the old nation-based world, there was always fear between them that if the neighbor is going to launch an offensive invasion to their territory. Unfortunately this still happens, but more often in the cyberspace rather than because territories per se. Therefore when the social medias are becoming more and more core elements in the modern performances of power, so does the fear re-appear for the authorities of those services that if the neighbor social media system will invade, if not completely occupy “ours”.
So, one could hear being asked and spied upon the question: Will the Facebook invade China? Or vice versa, will the Chinese Weibo replace Twitter in U.S.? While these questions are interesting, one hardly can provide an answer here, but only try to get the reader to realize the underlying tendencies.
Social media applications and tools are very important part of the modern governance performance, and the prime model to build the attaching identity to the sovereign. Hence its understandable that various countries aim to secure their social media applications while at the same time try to reach them as far out as they can.
However at the same time, the world is rapidly moving towards the post-industrial times when people are moving more and more across the continents, and not any particular government seem to be liable to force people in the traditional way to stay inside of this territory and remain working. Instead many governments and modern khalifa try to restrict the freedom of movement by declaring worldwide surveillance systems, global social media applications and global outreach of communications and even location. “There is no need for borders, as we know all the time where people go and are, we just need to ensure that there is no place for them to hide in the cyberspace,” might some people conclude.
All this here should not be taken so that it was arguing that the competition between social media was just about the struggle to gain social control. While it can be seen as being that, I would argue that its more about identity politics rather than control. Since the effective control mechanisms were merely financial or labour movement related, and there is little in the blockage of popular public websites to do with control, it could be labeled mere as a public relations campaign.
At the first sight, it might appear as irrelevant task to invite even one-fifth of the world population to our website. But don’t start sending your invitations yet! Your servers might not be able to carry the load. How then the worldwide, well almost, social media web application does the job? There are two issues here, one of them is the more important one: distributed nosql databases. The other thing is the web load balancers, but they are not so interesting here, although they enable the web edge delivery to scale up, and together with intelligent DNS, distribute users across the practically unlimited amount of servers hosting the social media application.
However the first innovation, which originates to Google, is the NoSQL database which is distributed across the again unlimited amount of nodes, where each can perform as master and a slave. Well you ask, we have had the old Oracle databases distributed and running for quite many years before Facebook, so where is the innovation here? Yes, that is true, you need to hear, but the old Oracle databases were based on the highly and strictly structured databases, where the data integrity, strong relations and distributed transactions were important. And it did all this on the cost of the performance and data structures. And it incurred high costs. For the social media companies, similarly as to Google in the early ages, both were unbearable; they could not compromise with the performance, scalability nor with the costs. Hence what became to be was anti/SQL, noSQL, with the aim to provide lightning fast distributed database where there is no sql any more. The intent to give up SQL was because it is the language which enforces the strong typing of the data structure, strong relations between tables and does not do well with distributed transactions.
Have you ever wondered why there were in the first place #links in the Twitter? Or why Facebook introduced the newsfeed? The core reason is that they are views to the distributed farm of NoSQL databases. They provide you the fast, immediate, result set from the worldwide database farm. And as such, they are only partial from the databases point of view. But they work. They give you the results you need, the results you are happy with, not the results which were perfect from the data structures point of view. In the world of tweets, it simply does not matter if one user across the globe gets your latests tweets a bit later than another. He just needs the tweets, most of them, and fast. This is what distributed NoSQL database guarantees to you, it will give you whatever it has, quickly, and leaves out the boring stuff, like hard typing and relations. The application layer, the Facebook, does it for you then later if needed. The same applies when you consider creating a tweet; in the world of SQL, a global write transaction might need to lock down the whole world in order to serialize the two conflicting data updates or inserts, and still one of them could fail, but in the world of distributed NoSQL, there is no locking. World must continue going around, there is no reason to try to stop it while someone is posting a tweet. In the world of NoSQL since it wont need to guarantee that your tweet is seen everywhere at the same time, nor that it has valid relations, you can post your tweet, and boom!, you are done. The database will then distribute your tweet across the globe while others are also tweeting and posting. No need for them to stop and wait until you are done.
So, in this light, the form of invasion in the world of social media, is not only question of the application itself, but also the implementation and competence acquired to run the technology needed. None of this is proprietary today, so you can start building your own distributed database today!
Censorship, the Ultimate Safeguard?
In most of the countries there are corps or robots of various size to browse the relevant websites and applications which are considered as important for the people and also considered mostly because of their volume of users or some other interest to be to raised to attention. However here one should remember that even when one regime operates hard post-publication deletion processes as part of their routine daily operations, some regimes, while not aggressively and openly deleting some classified content, nevertheless have similar perverse interest on what “their” people are up to and discuss about. This could be compared to the analogue of the publicly announced and celebrated myth of disappearance of border controls when crossing the old national lines in EU. However there is no reason to assume that the interest of the national authorities on where people move, if they leave and when might they return had somehow vanished, quite opposite. This way discussing about the benefits of the missing border controls, the physical queues at the crossing points, misses the essence that the interest and means to gain the knowledge in the borderless system might be even more intrusive and deeper, and just realized via other means.
This all being said, there should be no reason to argue just that ours or their way of the sovereign to exercise his interest to the subjects was somehow morally or technically better to others. However the paradox is that the technical infrastructure of distributed non-relational databases, and heavy use of treelike structured datasets is practically immune to the deletion of the content, and assumes that everything that happens in database, leaves there, forever. For example, have you ever tried to delete a commit in GIT repository? Therefore its understandable that some companies would not comply with the Chinese policies, but at the same time they could be argued against that they comply to so many other policies, requests and interventions in the name of the cultural compatibility.
While the world progresses towards the post-national times, the more and more rare will the coherent surveillance or censorship regimes be, at least in the terms of the legacy national territories. This all is evident by the fact that the Chinese Facebook ban has been raised in some of the areas while holding it at the others (clearly ignoring the fact that there are no geographical connotation to the ip address space, but an artificial one).
How would the Facebook and Twitter then solve the case of two somewhat opposing interests. in order to enter the Chinese markets? Well there is only one option, to compromise creatively and give up the arrogance to upholding only domestic intrusive mechanisms. One needs to remember and pay attention to the co-operation and friendship these companies have towards some other national intelligence authorities.
Still the whole issue of global equality of censorship or surveillance is somewhat absurd, since during all times in the history, people have conducted evolutional discussions, developed new ideas and collaborated, but what is important the means of doing so are very diverse. Hence it should be seen from the optimistic point of view that the diversity grows and maybe even some fragmentation is seen. Reducing the whole question of the freedom to speech to the censuring of Facebook is somewhat naive, isn't it?
Maybe one example could bring light to the otherwise sometimes one-sided approach. I was some months ago visiting an Internet Governance seminar where had an unique opportunity to discuss with some Russian comrades. They were very worried about the case of Youtube where they had experiences multiple times cases where some material they considered important in the light of the freedom of speech was silently deleted by some other authorities. One can just guess, but there was no reason to assume they were spreading false information. They were very concerned that there are no means for them to raise complaint or anything, as diplomatic cables and processes are developed for the use of all other than web content deletion claims handling process.
Turkish Twitter-Ban as a Performance
Maybe you have read about the recent event where Turkish government implemented a Twitter ban. Most of the name servers of the country were set to serve false DNS data to users looking for twitter.com. This address was not the Twitter server, but one run by the government and which contained just a notice that Twitter is closed here. This was part of the process where european countries and also globally various legacy governments started to acquire capabilities to block some domain names, or even restrict traffic to some ip addresses. How effective is this? In the face of the masses, it might serve the function of fear more than actually limit the communications across the borders. And for some people that is really the actual function of this kind of performance.
Similarly the implementation of the closed or restricted, private, social media appliances, is also part of the modern performative power. And this power includes the narrative of the other, someone who might invade, make an offensive and “take away our facebook”. This is, one needs to remember, no more real than the practical DNS hijacking implemented to divert users from twitter to local banner.
So, based on this, one could assume that yes, most of the social media applications, the companies that run those, and the governments who support them, want and need to expand. They mostly want and need to ensure that their outreach is having a good coverage, but as most of them, even the best ones, are in practice unable to do so, they need to perform as they were. They need to ensure that everyone thinks they have global coverage. Facebook needs to ensure that it has global coverage, as so well the NSA needs to demonstrate via proxy persons that it has the global capability which others just dream about. It cannot say it aloud, but it benefits and might even be happy that someone goes out and tells people what they are able to do.
At last, one needs to ask, in the light of the recent Ukraine crisis, which has demonstrated almost complete silence in the cyber domain, that why in this conflict there has not been seen a need to raise national cohesion in the form of similar website banning. Maybe they are operating on the level which does not any more involve the old concept of nation-building or restoration nor upholding of national identity politics, in the form of trying to perform Twitter bans. However as reported, the Russian invasion of Crimea recently included the takeover of local Internet Exchange points too, but so far this has been understood to be only an event related to the infrastructure and not related to any web application.