Here is the results page (in Russian) (it doesn’t say much about the project itself though).
The hackaton took place last weekend. There were 10 teams, and two won the prize money: $700 for the 1st place (that’s us!) and $500 for the 2nd.
I’ve already posted about the project on this forum.
Here is the project’s main repo:
We set up a new Snowflake network, and demonstrated the possibility of using Snowflake without Tor by setting up Snowflake as a TCP tunnel (much like in this issue), and setting up a SOCKS server on the remote end (server-side) of the tunnel. We then used a browser to connect to the SOCKS proxy through the tunnel (by specifying localhost:1080 as the SOCKS server address), and successfully browsed the internet with it.
The internet speed showed to be several times greater compared to the speed through Orbot.
The initial idea was to set up WireGuard instead of SOCKS, but we haven’t managed to resolve the routing issues associated with it in time.
Major thanks to the organizers, the jury, the mentors, and the participants!
It is an honor!
Since the result page is in Russian, I presume it took place there… and there are no people there who are upset that some folks work on censorship circumvention trying to make it better.
My Snowflake stats show that most of my clients are from there and by a big margin.
Thumbs up on the prize.
Hmmm, a thought just occurred. Maybe it will help them figure out how to block it
Thank you for your work! I had a question: this will require snowflake operators to have some sort of access to a vps/conventional vpn service, right? Also, how hopeful are you that this will take off?
By “snowflake operators” you mean Snowflake proxy operators? No, all they need is to be able to connect to the server where the VPN is running.
I was surprised that the idea was received so well, by several people. I’ve already been contacted by a small VPN provider, they’re considering integrating this.
And the Lantern’s “Browsers Unbounded” project is pretty similar, so I am confident that there is at least demand for such technology.
Even if it’s not gonna be a one big network of proxies, I think smaller-scale deployments also have potential (e.g. you telling your friend to install a proxy so that you can access a VPN through their machine, without any risks for them).
So, I’m pretty sure that something production-ready like this will exist, even if I’m not gonna be the pioneer in this area.