[adrian_b]

If you use a directive antenna to concentrate the radiated power into a small solid angle, to reach a distant receiver, you also increase in the same proportion the interference for another receiver that is located in the same direction as yours, but which does not want to receive your signal.

So limiting EIRP provides a limit for the interference suffered by a receiver that happens to be in the direction towards which you transmit, for which it does not matter at all which is the total power that you transmit in all directions.

[tyzoid]

True, but it dissuades folks from using directional signals, broadcasting RF energy in more directions and increasing the noise floor for everyone. I feel like there should be some sort of middle point here.

[adrian_b]

Directional signals are easy to use only when both the access points and the wireless stations are in fixed locations.

When this is true, it is trivial to use classic directive antennas to achieve very long range communications with standard WiFi. There is no need for expensive phased array antennas.

For mobile stations and/or access points, phased array antennas are not enough. See my other reply.

WiFi, Bluetooth and the other kinds of communication protocols standardized for use in the unlicensed bands are intended mainly for cheap mobile devices, and mobility at a modest price restricts the antennas to be omnidirectional.

[BenjiWiebe]

Directional antennas also are a benefit on the receiving side.