Have people forgotten that crypto is not casino tech but resilience tech for people? No we haven't! Heres a huge upgrade to the HOPR network getting us ready for the launch of Gnosis VPN! Start running a node now to earn $HOPR revenue from Gnosis VPN very soon ⤵️

What does it mean for an Ethereum wallet to be "self-sovereign"? What if a wallet's dev team walked away or turned evil one day? How much control and ownership over an account does the wallet give the user? Here's how Walletbeat evaluates self-sovereignty, and why these attributes matter 🧵

Update Telegram Desktop immediately! A critical vulnerability has been discovered that allows attackers to gain access to your account and secretly steal files from your computer. The attack uses links in the format tg://. If a special command is constructed, clicking the link can cause a malicious script to send a selected file to the attacker’s channel - without requiring additional confirmation.

October 5, 2026 A new letter was filed today by SDNY: The DOJ is still coming after me with everything it has. They really want to see me convicted. It has been 1,139 days since my nightmare began. They arrested me when my daughter was just 3 years old, and escorted her out of our home. She is 6 now, and I don't know how I would ever explain to her that I'm being sent to prison for decades - for writing code. Meanwhile, today the Department of the Treasury has admitted it was wrong about the mixer rules, citing a "chilling effect on legitimate activity." Source: Yet in my case, the DOJ argued at the April 9, 2026 hearing: "...even the legitimate transactions that went through Tornado Cash became illegitimate because they also facilitated the criminal ends of the money transmitting of the money laundering and of the sanctions evasion..." (Docket 290, p.43 The non-criminal branch of the U.S. government says: it's bad policy. The criminal branch says: every transaction is a crime. The Crypto War is still being waged against developers and open source, cementing an extremely dangerous precedent in the United States of America.

Me talking to Claude evolves like human communication Day 1: Good morning! Could you please build a progressive web app showing the mentioned data as a semi-log plot and investigate the trend line? 1 year in: Lol those numbers are weird. need stat son web, make have Dune vibes

We're thinking about this at Gnosis VPN. A mixnet like HOPR can in principle route every packet over a different path to a different exit. The hard part is using that in practice. Gnosis VPN establishes prepaid sessions to an exit. Relays are paid per packet and the exit can only answer through SURBs (single-use reply blocks) the client sends ahead, so it can reply without learning who you are. In theory sessions could be tiny and switch constantly, e.g. keep 100 open and cycle them per target IP. But: 1. Every session has setup overhead through the mixnet 2. Sizing prepayment for tiny sessions is hard: Prepay too little and the session runs dry mid-transfer and stalls while it tops up. Prepay enough to be safe and most of it expires unused. 3. Each session needs its own SURBs sent upfront, so uplink bandwidth scales with the number of sessions rather than with traffic The deeper problem is that the VPN's network interface only sees packets. It can't tell where one request to a server ends and the next begins, and each TCP/QUIC connection is pinned to one exit anyway. HTTP/2 puts many requests on one connection, many sites and apps sit behind the same CDN IPs, and TLS resumption and cookies relink whatever you manage to split. The VPN could try to guess request boundaries from timing, treating an idle gap on a connection as the end of a request. That isn't great either: - HTTP/2 and QUIC run concurrent requests on one connection, so many boundaries have no gap at all. - Keepalives and background traffic keep connections busy, hiding the gaps that do exist. - Any gap threshold is a guess: too short and it splits mid-request, too long and it misses most boundaries. - Even after detecting a gap, the VPN can't move a live connection to another exit. It has to reset the connection to force a reconnect, which costs a full handshake through the mixnet. - The reconnect then usually resumes the old TLS session, which relinks it on the server side anyway. The good news is that the VPN doesn't need to guess. Split tunnelling isolates the apps that matter, like your zkAPI CLI, from the rest of your traffic (like a long running Jitsi call with totally different traffic pattern and trust assumptions). On top of that we can expose a local SOCKS endpoint where each isolation key maps to its own exit from a pool of pre-warmed sessions. The app knows what a request is, so it uses a fresh key and a fresh connection per request, and you get request-by-request delinking for exactly the traffic that needs it.

We're thinking about this at Gnosis VPN. A mixnet like HOPR can in principle route every packet over a different path to a different exit. The hard part is using that in practice. Gnosis VPN establishes prepaid sessions to an exit. Relays are paid per packet and the exit can only answer through SURBs (single-use reply blocks) the client sends ahead, so it can reply without learning who you are. In theory sessions could be tiny and switch constantly, e.g. keep 100 open and cycle them per target IP. But: 1. Every session has setup overhead through the mixnet 2. Sizing prepayment for tiny sessions is hard: Prepay too little and the session runs dry mid-transfer and stalls while it tops up. Prepay enough to be safe and most of it expires unused. 3. Each session needs its own SURBs sent upfront, so uplink bandwidth scales with the number of sessions rather than with traffic The deeper problem is that the VPN's network interface only sees packets. It can't tell where one request to a server ends and the next begins, and each TCP/QUIC connection is pinned to one exit anyway. HTTP/2 puts many requests on one connection, many sites and apps sit behind the same CDN IPs, and TLS resumption and cookies relink whatever you manage to split. The VPN could try to guess request boundaries from timing, treating an idle gap on a connection as the end of a request. That isn't great either: - HTTP/2 and QUIC run concurrent requests on one connection, so many boundaries have no gap at all. - Keepalives and background traffic keep connections busy, hiding the gaps that do exist. - Any gap threshold is a guess: too short and it splits mid-request, too long and it misses most boundaries. - Even after detecting a gap, the VPN can't move a live connection to another exit. It has to reset the connection to force a reconnect, which costs a full handshake through the mixnet. - The reconnect then usually resumes the old TLS session, which relinks it on the server side anyway. The good news is that the VPN doesn't need to guess. Split tunnelling isolates the apps that matter, like your zkAPI CLI, from the rest of your traffic (like a long running Jitsi call with totally different traffic pattern and trust assumptions). On top of that we can expose a local SOCKS endpoint where each isolation key maps to its own exit from a pool of pre-warmed sessions. The app knows what a request is, so it uses a fresh key and a fresh connection per request, and you get request-by-request delinking for exactly the traffic that needs it.