Tommy Huynh , May 13, 2001; 03:15 a.m.
"Changing the system from light signals to radio signals is problematic in several ways: first, it starts getting involved with FCC and other various regulating organizations."
RF emmissions are regulated under the FCC's part 15 rules just as much as IR emmissions. Canon could have just as easily(regulations-wise) used the 900MHz or 2.4 GHz band that are public spectrums requiring no licensing (but does require certification as does the IR E-TTL and just about all electronic products nowadays). There could potentially be interference with cordless phones, home networking devices, etc.. but this could be easily overcome(for the most part) with conventional coding, proprietary protocols, and an employment of FHSS(Frequency Hopping Spread Spectrum) on the airlink (Spread spectrum is required as part of the rules anyways). *However*, this would be much more costly than the IR setup and here is the real answer why this is not done.
"This could potentially reduce the effective range of the system even more than the ST-E2 does already"
Also not true. Using RF signals in the public bands, you could have a range of almost 2 miles(!) depending on if you're line of sight, your surroundings, and how much modulation or bandwidth you need(not much). Just look at the range of your digital cordless phones.
One thing that disturbed me as a Canon user in the article was how much exposure compensation that guy needed to use to properly expose his subjects using the wireless E-TTL slave. Have others here found it necessary to add +1.5-2 stops on the slave to get proper exposure?