HTTP/2

HTTP/2 is a major upgrade to the web's core protocol. Where HTTP/1.1 effectively handled one request at a time per connection and opened many connections to compensate, HTTP/2 multiplexes many concurrent requests and responses over a single connection using a binary framing layer. It adds header compression and lets the server prioritise streams, cutting the round trips needed to load a modern page. It still runs on TCP, which means a single lost packet can still stall all its streams — the head-of-line problem that HTTP/3 later solves.

Because HTTP/2 changes how a client frames requests, its low-level behaviour has become another fingerprinting surface: the specifics of how a client uses HTTP/2 can reveal whether it is a real browser or a script, complementing the TLS fingerprint. For general use it just makes pages faster, and it works transparently through the s4m VPN and, over its CONNECT tunnel, the HTTP proxy. For automation, be aware that speaking HTTP/2 like a genuine browser is part of looking like one to a strict target.

Because a client's low-level HTTP/2 behaviour has become a fingerprinting surface, strict targets check that it matches the browser you claim to be, alongside your TLS fingerprint. For everyday use HTTP/2 simply makes pages faster and passes cleanly through the s4m VPN and, over CONNECT, the HTTP proxy. For automation, speaking HTTP/2 like a genuine browser — not a bare scripting client — is part of looking like one to a demanding site.