November 25, 2025
WiFi vibes or WiFi lies?
Nearby peer discovery without GPS using environmental fingerprints
Phones find nearby 'friends' by matching WiFi vibes—privacy skeptics and patent police pounce
TLDR: Shimmer tries to find nearby peers by matching the same WiFi and beacon signals, promising privacy without GPS. The comments erupted over weak privacy, possible patent overlap, and jokes that it’s “just SSIDs,” while some proposed speed‑of‑light latency as a better test—making this a hot debate on proximity without location.
Forget GPS: Shimmer says your phone can spot nearby people by matching the “vibe” of your surroundings—same WiFi names, Bluetooth beacons, and cell towers—using cryptographic fingerprints instead of raw lists. The dev insists we never reveal location, only that we see the same one, and a demo turns overlap into a secret handshake. Commenters asked: will this work off the whiteboard? Subway fan figmert loves it for underground spots, picturing transit apps that sync by shared station WiFi.
But the crowd’s hottest take is privacy vs usefulness. Security pro vessenes warns, “I don’t think it’s privacy preserving,” arguing better matching demands more exposure. Patent drama popped when tomgag said, “Watch out, possibly similar to this patent,” prompting armchair IP debates. The snark award goes to meindnoch’s “TL;DR: it’s just using the WiFi networks’ names,” while arcbyte pitched speed‑of‑light latency as a cleaner test. Memes landed—“Match my SSID, not my soul”—plus jokes about accelerometers measuring spacetime. Others asked if tuning would fix misses; skeptics saw fingerprinting bait for trackers. Love it or loathe it, the thread turned a clever demo into a street fight about privacy, patents, and whether your environment is a fingerprint or just noise.
Key Points
- •Shimmer enables nearby peer discovery without GPS by comparing cryptographic fingerprints of local signals instead of sharing raw observations.
- •MinHash creates compact similarity fingerprints of observed sets; LSH bands produce collision buckets for probabilistic matching.
- •Encrypted peer information is announced to a rendezvous server using public tags, enabling discovery and decryption by peers with similar environments.
- •The approach supports multiple categorical modalities beyond WiFi, including Bluetooth beacons and cell towers.
- •Detection reliability depends on overlap and probabilistic parameters; tuning MinHash k-values and LSH bands may improve matching, with further testing planned.