The Death of the Internet
The Internet is arguably one of, if not *the*, most significant technological advancements of the 20th century. Through its far-reaching and versatile capabilities, it has spurred an informational paradigm shift on par with that of the printing press in 1440. Since its public release in 1991, it has gone from being a simple medium for the transferal of information to an all-penetrating part of our society. Being used for everything from banking to entertainment, to academia, to even household appliances. Yet for all its importance in the lives of many, it is a common mindset for people to simply accept it as a natural feature of the world, especially given that for an increasing number of people, they either can’t remember or were not alive prior to the existence of the internet. As a result, certain questions about its nature go unasked, such as the issue of who owns the most important portions of its infrastructure, or how its infrastructure actually works for that matter, or even more simply: will it last forever?
Historically, the internet has shown strong resilience in the face of a number of natural disasters and terrorist attacks, which threaten the physical infrastructure of the internet in the case of the former, and the freedoms in that of the latter. It has also stood well against deliberate sabotage, such as the Dyn Cyberattack which, for a time took out a significant portion of the American internet. In the face of all these onslaughts, the network has been able to recover with relative ease. But the question still remains: is the internet as impregnable as it appears, or have we simply been lucky thus far? Several Cambridge computer scientists seem to suggest the latter.
They state that should the greater internet infrastructure fail, it would either be through a failure of the physical infrastructure based upon, such as the bulk power grid; a loss of the people necessary to maintain the internet (such as through say, something as simple as a flu epidemic causing people to isolate themselves in their homes); or through a massive coordinated Distributed Denial of Service (DDoS) attack. They argue, first of all, that there is evidence that some implementations of the Border Gateway Protocol (BGP) the internet’s main routing protocol (which dictates which path signals should take if the default pathway has been compromised) are fragile. The implications of this are that, should someone with sufficient motive and knowledge attempt to, they’d likely be able to take advantage of these vulnerabilities and disrupt the pathways for a significant portion of the internet. Furthermore that there is evidence that should certain local regions be taken down, that the health of the entire system would be jeopardized.
Then we come down to the fundamental nature of the issue. Given that the vast majority of the internet’s infrastructure is privately owned, and therefore developed with the mindset of generating the best possible ratio of stability to cost, one finds the issue that attempting to improve the security of the system as a whole is an extremely costly endeavor that offers little individual incentive, and thus is prioritized by few if any networks.
Though a massive outage is quite unlikely, and potentially not even a cause for immediate concern, a far more disconcerting issue is that of the lack of understanding regarding the mechanisms of the internet, or even more so its potential to perform in the wake of a massive outage. This is not simply due to a lack of public interest in these things, but rather a simple dearth of information. To date, there is no map of the physical connections, traffic or traffic volume, due largely to the simple complexity of the system — while it may be an achievable undertaking to make a map of the physical fiber globally, the sheer number of logical connections that run through these lines would require meticulous bookkeeping to maintain accurate numbers. This is scarcely achieved on individual networks, to even consider such an undertaking on the scale of the entire system is nigh-unthinkable. Furthermore, the few small maps that actually do exist, are seldom shared, for fear that showing a clear picture of the physical layer may pose security threats as the most critical targets to the system are far more easily found this way. But above all else, you come to the most limiting concern. The Internet is massive. It is functionally impossible to give an exact figure on the number of pathways in the internet, or even more simply the amount of data on the network. It’s essentially immeasurable. We lack even a clear picture of the nature of the system we are trying to improve, and therefore have greater difficulty doing so.
But it all ultimately comes down to the tragedy of the commons. As mentioned before, the individual effects of making changes that may improve the health of the overall system are rarely individually beneficial, and in fact tend to be the opposite. So, logically, people tend to not implement these changes, which makes sense from a financial perspective. But it leaves the internet vulnerable in a number of ways.
So here we stand, a relatively unknown instability spreading throughout the most important system in the entire world, and little being done to resolve it. All hope is not lost, as there are investigations being done into what may be done to improve the stability of the internet, but at the moment, the future is quite uncertain. Given how entangled the internet has become in our society, the collapse of the internet could have far reaching and unforeseeable side effects.
For further reading
