A bus is for sure more efficient per mile than a conveyor.

Wear and tear.

The technology has improved, so the device is more complicated than the early prototypes. That's why I included the diagram above.

how hard could it be?

Well, how much mass does it need to move, if it's 100m long, and is fully loaded with people? To keep it simple, we'll be generous and assume an average 80kg human, spaced evenly every 1m, so that's 8000kg of mass that the motor needs to be able to move, constantly. We don't want the system to change speed under load, but keep a constant speed, so we'll actually want to be able to handle more than the average load, with room to spare, so call it 12000kg. That's... a lot of torque.

We'll also need a compensation mechanism to keep the speed steady as the load on the motor changes, and we'll need an emergency stop braking system. We'll need shock absorption to handle the impact of people stepping onto the walkway and potentially dropping any objects they're carrying, so that the constant load/unload doesn't cause additional wear on the drive system. We need the mechanism to move the hand railing along with the walkway. We'll need to run electrical wiring to each unit. We'll need to dig a hole in the ground to fit the mechanism for each unit. We'll need to safety check each unit post-install.

This simple-in-theory device is pretty complicated in practice. We haven't even talked about weatherproofing.