But does it need to break to create a risk?
"the risk posed was far less alarming than both the union and the Department for Transport had claimed. Instead of penetrating cockpit windows, rigorous tests of drones launched against"
I'm not so sure I'd support that theory. Because even if the windows only get cracked instead of broken there's still plenty of damage being done. Not necessarily physical damage (as demonstrated in the study) but the pilot(s) are still at risk for getting exposed to some severe distraction.
Sure, that doesn't have to immediately result in a major crash, but it's still a risk factor which I think should not be taken too casually, as seems to be done here.
Of course I still think a general drone registration seems a bit off and only diverts the attention away from the real problem. I mean... Do you really think that a regular drone used somewhere in-land (say 100km away from the nearest airfield) could pose a risk for any airplanes? I somewhat doubt that.
Instead of requiring people to register, why don't they uphold better security measures around airfields and actually enforce those? So: if you spot someone operating a drone near an airfield then you fine him for endangering air safety. Surely it should be doable with todays technical standards to pick up any signals which are used to operate a drone and then take according action?