We are now at a point where the internet has become so centralized that a single company can bring it to its knees. What started as a military communications system 53 years ago and has become a basic human right is now so intertwined that any fallibility could wreak havoc and affect millions of people.
The now ubiquitous Internet began life as a truly decentralized communication system. Beginning with ARPANET and progressing to the launch of the World Wide Web, the decentralized nature of the Internet remained such until the Web became increasingly important and vital. Subsequently, centralized services and providers emerged, and while they improved the user experience at the time, they also undermined it.
No more downloading software to install on a computer or server for yourself or your employees. Now we can open a browser and do almost anything without installing an app.
Everything is based on the cloud.
As with all markets, only a small number of organizations have become the world’s leading cloud service providers. While simplifying the big picture, given the importance of cloud-based solutions, you could argue that the modern internet is run by Facebook, Google, Apple, Tencent, Alibaba, and Amazon.
Even if you look beyond cloud providers, the story of the internet’s continued centralization is similar.
Through the acquisition of its own Messenger app, Meta owns almost all instant messaging communications in non-Chinese regions. By owning Messenger, WhatsApp and Instagram, Meta dominates the flow of our conversations between us. In China, the two most popular messaging apps, WeChat and QQ, are owned by Tencent and are responsible for almost all communications and transactions in the country.
We could go on. Almost all data carried around the world passes through the top 25 telecommunications companies, and almost all smartphones run on a choice of two operating systems. On virtually every metric, the intended decentralized design of the Internet is under threat, and with centralization comes greater risk.
On June 21, 2022, Cloudflare – a content delivery network and Distributed Denial of Service (DDoS) attack mitigation company – suffered an outage that affected traffic in 19 data centers. These 19 locations handle a significant portion of global traffic. An error in their messaging to the internet routing protocol led to an outage that broke some of the most well-known and essential websites and services, including Discord, Shopify and Peloton.
The issues were problematic for users of Cloudflare’s DNS lookup service, but were far-reaching for everyone on the internet. Some estimates suggest this outage affected 90% of the internet, and it’s not a “Black Swan” moment – there have been hundreds of outages since the start of 2022.
Blockchain and Web3 solutions are also not immune to the shortcomings of current internet technologies. Since most Web3 projects are hosted on cloud services such as AWS, this outage put an end to many supposedly decentralized Web3 applications and services.
The question we need to ask ourselves is how did we let this initially decentralized Internet become so centralized that a single vendor failure can destroy it.
Centralization is the product of those who seized the opportunity to evolve before others. There’s nothing bad about Cloudflare – it provides a great service that millions of people need. By providing the best perceived service and value, customers have turned to her. But as a result, everything from your blog to giant corporations depends on its service.
So while the protocols that allow the Internet to exist and function are decentralized, some services have near-monopoly control because they were the best options available in the beginning to help scale and protect it.
And now the infrastructure is monopolized; when it fails, the entire internet fails.
The magnitude of this problem is almost too great to imagine, and it will get worse. The Internet is going to grow 30 times over the next eight years.
Before looking at what can be done, we need to understand why today’s Internet is not fit for the future.
The Internet runs on a routing protocol called BGP (The Border Gateway Protocol), sketched out in 1989 by engineers on the back of “three ketchup-stained towels.”
It served us well for a while and we built many upgrades to alleviate its issues, but in the end these are band-aids and not the long-term surgery needed.
The problem with the modern Internet is that it is essentially a collection of private networks run by individual Internet Service Providers. Everyone has a network, and most connections happen between those networks. Even if you’re chatting with your co-worker on Zoom and you live in a city (or are in the same apartment), that signal will likely travel much farther.
Your physical atoms may be five meters apart, but the pixels that display your two-dimensional form relative to each other may have taken a transatlantic journey.
Networks are managed only locally. Routing decisions are made locally by providers via the BGP protocol. There is no shared knowledge and no one controls the entire connection route.
Using these public ISPs is like using public transport. You have no control over where it goes. The vendors own the cables and everything. In this system, ISPs have no incentive to provide good service.
And we’re not talking about the speed of your internet at home. We are talking about the ISP connection to services like Netflix.
This was the origin of the net neutrality battle in the United States. Netflix was sending too much data (around 1GB of data per hour when streaming a standard definition TV show or movie and up to 3GB of data per hour when streaming HD video), and ISPs in the US threatened to cap it unless Netflix paid. .
And that’s how it goes. The only way to get more reliable service is to pay ISPs dearly for high-speed private connections. It’s the only way for big tech companies like Amazon to manage their data centers. But the biggest irony is that there is enough infrastructure to handle much more growth.
70% of the Internet infrastructure is not used because nobody knows these routes and ISPs do not have a great solution to monetize them on demand. They prefer to work on the basis of fixed and predetermined contracts, which take a long time to negotiate and sign.
There are other issues. For example, the Cloudflare outage happened because the internet is built on trust. The network expects everyone to be a good actor, and if someone, maliciously or otherwise, sends false information, they can easily snap.
We trust these actors. We trust these centralized providers. We even trust organizations to keep our sensitive information and the Internet, as a whole, private.
The HTTPS protocol ensures that the data we send, whether it’s a personal photo or a credit card number, is encrypted. But this encryption is based on a certificate provided by a specific organization, such as LetsEncrypt. If someone tampers with it or if a government orders everyone to install their certificate, all your data will be at their disposal.
That is why without trust systems are essential. It’s not just a marketing term created by Web3 preachers. The system is not susceptible to multiple attack vectors and can exist even in a hostile environment.
The blockchain provides a natural solution to this problem. At a fundamental level, this technology allows us to create decentralized systems that are secure by design and governable without a single central entity.
It can also be private by design due to public key encryption. Public key cryptography is a cryptographic system that uses pairs of private and public keys. Anyone can encrypt a message in the system using the intended recipient’s public key, but that encrypted message can only be decrypted with the recipient’s private key.
Imagine a system where participants such as content companies like Netflix and ISPs can dynamically buy and sell bandwidth, move their data through the best possible route, and avoid being locked behind a centralized provider, formerly the only way to reach the ladder. Users, in turn, will get a more reliable connection with lower latency, which is not so important for web browsing, but very critical for online gaming and future Metarverse applications.
Bloomberg estimates the metaverse industry will be worth $800 billion by 2024, based on its analysis and data from Newzoo, IDC, PWC, Statista and Two Circles. This will require significant additional investment in Internet infrastructure. The software we use to run it must be able to support the growth we envision and conforms to the decentralized principles that are one of the cornerstones of the Metaverse and Web3.
The demands of future media channels and technologies, such as AR glasses or any version of the metaverse, will exceed anything the current version of the internet can provide us with right now. And the greater the role it plays in our lives, the greater the consequences of any failure will be. Now is the time to address this issue before we lose access to this fundamental human right for so long that it results in significant economic, infrastructural and personal losses that affect our well-being.
Domas Povilauskas, co-founder and CEO of Syntropy.