In the last 10 years the phrase virtual private network, or VPN for short, become mainstream. Today, every teenager knows what it is and why it is necessary to be installed at almost any smartphone, laptop or tablet. Twenty years ago 99% of Internet users could...just use the Internet. Without crutches.
But, the situation had been slowly but surely become worse year by year. Restrictions gradually imposed by governments lead to more than 1.2 billion monthly VPN users, according to Globalwebindex. The Global Mobily VPN Report 2019 claims that VPN apps were installed 480 million times on smartphones all over the world during the last year. It is a 54% year-over-year increase. The market grows by leaps and bounds.
Why? Some people want to protect their privacy and personal data. Others just want to see favorite Netflix series or access to other restricted content. Or to send a message to their family. The majority of people don’t do anything special - they just want the Internet. Its “full” version. They don't need anything more.
Today centralized VPN providers completely own the market. It is not surprising. However, centralized VPNs have some obvious limitations, that every meticulous user will see:
- They could store log files. And no doubts a lot of them do it;
- They can be blocked and it’s why the majority of VPNs doesn’t work in mainland China (due to static network footprint);
- The users of centralized VPNs are competing for the service bandwith;
In opposite, a decentralized VPN approach looks quite attractive:
- If the VPN is decentralized, the attacker needs to own the majority of network nodes to get a picture of participants net activity;
- Blocking dVPN is a more complicated task due to the dynamic network footprint and set of used IPs;
- The bandwith of decentralized VPN network can grow with users demand thanks to economic incentives and ability to permissionless participation;
So, if this idea is so attractive, where is at least one mainstream dVPN with millions of users? Let’s answer this question later and check historical picture first.
Did you know that decentralized VPNs existed before Bitcoin’s appearance?
The first mention in academic sources regarding the p2p VPN network comes back to 2005 and was called ELA. Two years later the first commercial decentralized VPN was launched: hola.io. It is still in operation today, but some researchers claim that it is unsafe and even malicious. Other examples include SocialVPN and VPN Gate. The latest one demonstrated a succesful case of providing access to web sites, restricted in mainland China. They used an approach, which comes down to a set of dynamically changing IP addresses.
Why dVPNs didn’t become a mainstream approach for building VPN-based business? The answer is simple - there was a lack of technology. A distributed VPN system requires a private and secure transaction layer for bandwith exchange and implementation necessary economic incentives to agents. The economic incentives are a core part of every anonymous and permissionless blockchain system: you cannot trust people, you can trust incentives and code. For example, in the TOR network actors don’t have incentives to run an exit node besides collecting users’ data. So the community has reasonable suspicion that special forces run the majority of nodes in the network and therefore track at least part of users.
The state of decentralized VPNs now
Now at the end of 2019, it looks like a decentralized VPN built on top of blockchain technology is a “must-have” for the rapidly growing VPN market. But, the dVPN is still a not-realized dream of the community. There are several well-known projects such as Mysterium, Sentinel, Orchid, Lethean and the VPN0, the Brave browser-based dVPN service. In our opinion, each project is in the early test phase now. It is a bounty phase, during which early adopters test the network by launching nodes and would get rewards. The majority of these projects (first three) were funded at least 1-1.5 years ago and it is great to see first results.
We think, that dVPN is not only about unrestricted Internet access, but also about private and secure p2p data transmission layer. Our team is developing dVPN solution, named Kelvin or KELVPN. It is a completely open-source project and based on its own VPN client, developed from scratch using plain C. Also, we first in VPN space implemented post-quantum cryptography, including ZK-SNARK payments to achieve highest level of security.
All the mentioned projects are obviously competitors from one side and clearly, have different market niches from another. Some of them, like Orchid and Kelvpn, are focused on onboarding VPN providers into the network, while others are focused on ordinary users as hosts first. It will be an interesting race, for both users and teams behind dVPN.
So, what is expected to see in 2020?
In 2020 there will be even more new VPN users all over the world. I assume at least 25% growth of the user base. The blockchain technology becomes more matured and trusted by the community. So, the dVPN startups will have a unique opportunity to onboard millions of new users, which will be driven by lower costs, absence of logs and good performance of the service. If in 2019 Defi was a "killer case" for crypto, in 2020 VPNs will take the lead. It’s simple - VPNs have at least 10 times a bigger user base and aimed at the real everyday issue, important for every person.
Let's see - December 2019 is a test month for all of the mentioned dVPN solutions, so we will track the progress in the next posts.
 S. Aoyagi, M. Takizawa, M. Saito, H. Aida, and H. Tokuda. Ela: a fully distributed vpn system over peer-to-peer network. Symposium on Applications and the Internet, pages 89–92, Feb 2005.
 slipstream/RoL,D.O’Cearbhaill,S.Slootweg,IceMans/RoL,infodox, pathfinder/braenaru, APT1337, spoonzy, LeShadow,. Adios, hola! or: Why you should immediately uninstall hola. http://adios-hola.org/, 2015.
 P. S. Juste, D. Wolinsky, P. O. Boykin, M. J. Covington, and R. J. Figueiredo. Socialvpn: Enabling wide-area collaboration with inte- grated social and overlay networks. Computer Networks, 54(12):1926– 1938, 2010.
 D. Nobori and Y. Shinjo. VPN gate: A volunteer-organized public VPN relay system with blocking resistance for bypassing government censorship firewalls. In Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation (NSDI 14), pages 229–241, Seattle, WA, 2014. USENIX.