"Just as such devices had done through all of human history, these located one another in geometrical space -- a simple exercise, nothing more than time-of-flight computation. [...] They made great spy devices [...] Localizers were by their nature a type of computer network, in fact a type of distributed processor."
Interesting. Does anyone know how this technology compares to passive antennas in RFIDs?
There was an article the other day posted on this. It operates at 24 and 60 ghz whereas most RFID is in the UHF (900 MHz band). This radio harvests energy to run an oscillator and actively transmits data whereas RFID only reflects back power (no active rf transmission). Also, this device require a fairly massive (42 dBm / ~20 watts rf) for a few cm operating distance compared to RFID which achieved several meters of distance using 36 dBm / 4 watt transmitters. The biggest achievement is the size reduction. In the previous HN thread, someone poured out that this is academically very interesting, but probably practical.
Did you mean "probably impractical"?
I had the impression these could "process" something. Can passive RFID do that?
Passive RFID chips usually just respond with a serial number, passive smart cards such as found in Oyster and credit cards can have a processor and even run Java. They both basically have a chip with some wire coils attached. I think.
>and even run Java
A very limited subset of Java, like Java Card. No garbage collection, no arbitrary objects, limited numeric types, etc.