6 ms·

I can't, but here's the kicker...neither can you. In this business, you're not allowed to presume safety. Unless you've been trained to understand the ramificat

by madrox 8y ago

I can't, but here's the kicker...neither can you. In this business, you're not allowed to presume safety. Unless you've been trained to understand the ramifications on airspace, you really don't know. That's not how a safety first culture works. You don't make a change to a business critical production system without going through change control...why is it so absurd to consider registering your activity in airspace before you do it? The only reason I can think of is that it's inconvenient for you, which sucks but that's not good enough.

myrryr 8y ago
Stats is on his side. He CAN assume it is safe, since given how many are in the air, and how few problems we have, in his scenario we can assume it is safe unless shown otherwise. I am for the regulation, but, I am also against people not using basic stats. Riding a pair of rollerskates can be dangerous to the people around you, but, you wouldn't force people to licence to do that right? You can hurt someone if they get a paper aircraft in the eye, but you don't licence them either. It has to be on the basis of how likely you are to get problems, and how dangerous those problems are going to be. Saying "I can (or can't in this case) conceive of a situation where this could be a problem" isn't a high enough bar. For larger drones, hell yeah you will need licencing, and flying near aircraft, then again, yes. But this case? In this case, he has stats on his side.
- madrox 8y ago
  As luck would have it, I have a degree in statistics and I'm a trained statistician! So let's talk about stats. Let's assume he's 99.99%[0] safe to fly his drone in this situation. That means 1 in 10,000 drone flights are not. That's actually not a great safety margin when you consider how often this is done a day. There are 2.5 million drones in operation at the moment. You're overemphasizing the individual case here and not the population at large. There's actually a formal name for this type of statistical fallacy, though I've forgotten it. There's TED talks about it and everything. But you can probably get better than 99.99% assurance of safety here...and you do that by registering your intent with people who assess safety for a living. [0] this is, of course, a ludicrous statistic. It's illustrating the absurdity of thinking that just because something is unlikely doesn't mean it won't happen. In this case, when it happens people can get seriously or fatally hurt.
  - gbacon 8y ago
    You’re thinking of the fallacy of composition. https://en.wikipedia.org/wiki/Fallacy_of_composition https://en.wikipedia.org/wiki/Fallacy_of_composition
  - kempbellt 8y ago
    I'm curious where you are getting your 99.99% figure, and why you thought that is a would be a safe assumption to make your point out of. Do "trained statisticians" frequently assume with data? If you can provide evidence that drones are 99.99% safe in the aforementioned "this situation", I will be happy to recant my retort. Otherwise, please avoid strawman arguments. As for @myrryrs argument. I believe the sentiment is along the lines of "don't let perfect be the enemy of good" - which I am in favor of. Blanket rules that disallow amateur drone usage in relatively safe situations are overkill. General safety rules that urge drone users to stay away from airports - probably a good idea. If my intent is to fly my drone around in the woods while I go mountain biking, why should I be required to tell anyone? Let us add regulation where regulation is needed - not throw it around like a panacea whenever something unpleasant happens.
    - madrox 8y ago
      99.99% is a wild assumption merely to illustrate the fallacy in thinking that just because something is unlikely doesn't mean it won't happen. And I'm all for "don't let perfect be the enemy of good" in many situations...but none of them are where people die if mistakes happen.
      - myrryr 8y ago
        And yet, you eat food prepared by others, take buses, play sports, swim, step foot on a boat, cross streets. (I am only guessing you do these things, most people do) ALL of these are where people die if mistakes happen. It just comes down to risks. Again, I actually think that a licencing situation for larger drones would be a good idea, I think bans near airports is good. I just hate some of the arguments people use to get there.
jsight 8y ago
As a pilot of both full-sized aircraft and quadcopters, I think I can explain how 100 feet could potentially be a danger. Directly underneath the approach path of a runway that is less than a few hundred feet away might be bad. Otherwise, there isn't a material impact. Noone should be flying their airplane within 100 feet of his house (by regulation). I don't mean to call you out specifically, but this is really the problem with our regulatory environment. Lots of people involved with regulation actually have remarkably little understanding of the cost and benefits associated with the regulations that they advocate. I wish that I could go into more detail, but I've seen some truly crazy things being decided by people who literally had no idea of the consequences of their actions or how people in their field actually behaved.

by madrox 8y ago

myrryr 8y ago

Stats is on his side. He CAN assume it is safe, since given how many are in the air, and how few problems we have, in his scenario we can assume it is safe unless shown otherwise. I am for the regulation, but, I am also against people not using basic stats. Riding a pair of rollerskates can be dangerous to the people around you, but, you wouldn't force people to licence to do that right? You can hurt someone if they get a paper aircraft in the eye, but you don't licence them either. It has to be on the basis of how likely you are to get problems, and how dangerous those problems are going to be. Saying "I can (or can't in this case) conceive of a situation where this could be a problem" isn't a high enough bar. For larger drones, hell yeah you will need licencing, and flying near aircraft, then again, yes. But this case? In this case, he has stats on his side.

madrox 8y ago
As luck would have it, I have a degree in statistics and I'm a trained statistician! So let's talk about stats. Let's assume he's 99.99%[0] safe to fly his drone in this situation. That means 1 in 10,000 drone flights are not. That's actually not a great safety margin when you consider how often this is done a day. There are 2.5 million drones in operation at the moment. You're overemphasizing the individual case here and not the population at large. There's actually a formal name for this type of statistical fallacy, though I've forgotten it. There's TED talks about it and everything. But you can probably get better than 99.99% assurance of safety here...and you do that by registering your intent with people who assess safety for a living. [0] this is, of course, a ludicrous statistic. It's illustrating the absurdity of thinking that just because something is unlikely doesn't mean it won't happen. In this case, when it happens people can get seriously or fatally hurt.
- gbacon 8y ago
  You’re thinking of the fallacy of composition. https://en.wikipedia.org/wiki/Fallacy_of_composition https://en.wikipedia.org/wiki/Fallacy_of_composition
- kempbellt 8y ago
  I'm curious where you are getting your 99.99% figure, and why you thought that is a would be a safe assumption to make your point out of. Do "trained statisticians" frequently assume with data? If you can provide evidence that drones are 99.99% safe in the aforementioned "this situation", I will be happy to recant my retort. Otherwise, please avoid strawman arguments. As for @myrryrs argument. I believe the sentiment is along the lines of "don't let perfect be the enemy of good" - which I am in favor of. Blanket rules that disallow amateur drone usage in relatively safe situations are overkill. General safety rules that urge drone users to stay away from airports - probably a good idea. If my intent is to fly my drone around in the woods while I go mountain biking, why should I be required to tell anyone? Let us add regulation where regulation is needed - not throw it around like a panacea whenever something unpleasant happens.
  - madrox 8y ago
    99.99% is a wild assumption merely to illustrate the fallacy in thinking that just because something is unlikely doesn't mean it won't happen. And I'm all for "don't let perfect be the enemy of good" in many situations...but none of them are where people die if mistakes happen.
    - myrryr 8y ago
      And yet, you eat food prepared by others, take buses, play sports, swim, step foot on a boat, cross streets. (I am only guessing you do these things, most people do) ALL of these are where people die if mistakes happen. It just comes down to risks. Again, I actually think that a licencing situation for larger drones would be a good idea, I think bans near airports is good. I just hate some of the arguments people use to get there.

jsight 8y ago

As a pilot of both full-sized aircraft and quadcopters, I think I can explain how 100 feet could potentially be a danger. Directly underneath the approach path of a runway that is less than a few hundred feet away might be bad. Otherwise, there isn't a material impact. Noone should be flying their airplane within 100 feet of his house (by regulation). I don't mean to call you out specifically, but this is really the problem with our regulatory environment. Lots of people involved with regulation actually have remarkably little understanding of the cost and benefits associated with the regulations that they advocate. I wish that I could go into more detail, but I've seen some truly crazy things being decided by people who literally had no idea of the consequences of their actions or how people in their field actually behaved.