A remote control works by having a remote unit transmit a signal to a receiver. The receiver then decodes the signal, and if it is valid takes some action.
A typical example could be like this:
A TV remote has many buttons. Pressing one of them sends a signal to either a custom encoder, or a microcontroller. This device then creates a stream of pulses that is sent to an infra-red LED. The LED produced invisible flashes of light that are picked up by the receiver in the TV. It first uses a photodiode, an amplifier, and filters to recover the stream of pulses (this is all typically done in a single device) which are sent to a microcontroller which determines if the stream of pulses encodes a valid function. If it does, the function is determines and the microcontroller does what is required (possibly sends a signal to power up the TV, or to turn it off, or to increase the volume, brightness or contrast, or to change the channel)
IR is not the only method of sending a signal. You could use an audio signal (I once had a TV that used ultrasonic signals), a cable (my grandmother had one of those), or a radio signal (I have some -- not TV remotes -- that use this).
With some minor variations, the method you use to get the signal from the transmitter to the receiver is largely immaterial.
There would be very few such systems that are not software controlled to some extent. Some of the simpler systems (with up to (say) 4 on/off type controls -- like a garage remote) may use custom chips that encode and decode without the need for a microcontroller.
Why don't you like IR?
