4 ms·

Feynman had a nice story where he figured out a way to crack many of the safes in Los Alamos, then dutifully reported his method to some bigshot general. The ge

Feynman had a nice story where he figured out a way to crack many of the safes in Los Alamos, then dutifully reported his method to some bigshot general. The general said "hmm interesting, thank you very much", and banned Feynman from entering rooms with safes or something. The safes stayed as unsafe as ever. You remind me of that general. You should be hanging out on Catch The Hacker News, not Hacker News.

rhino369 11y ago
Testing on uninformed humans is unethical. Wasn't hackernews just all up in arms about the US military spreading germs to test bioweapons? Isn't this the same exactly thing?
- yellowapple 11y ago
  There's a pretty significant difference in scale between two nerds putting maybe 2 or 3 vehicles in danger of a dented bumper v. the world's largest military conducting live-fire bioweapon testing.
- danielweber 11y ago
  There can be times when it is okay to test on uninformed humans. For example, I have relatives who do fire safety. How people do (or don't!) evacuate from buildings when fire alarms go off is a big area of research. The ideal way to test this is to set off the fire alarm in a building where people do not know it is happening, along with some smoke and pyrotechnics. HOWEVER, there are ethical concerns, and a review board would ask questions like: 1. Has anyone else done this study before? If not, why not? How sure are you that no one has done it before? 2. What does the previous research with similar protocols say? What key question are we trying to answer? 3. What is the harm that will be present to people? Are we doing everything we can do to reduce that harm? 4. What more could we do to reduce harm but that might impact the reliability of the research? 5. Quantify how much of a benefit this research would be so we can compare to the risk you are presenting. 6. Demonstrate that you have done all the preliminary work that is necessary to achieve good results, so that we can make sure that the research is used. It would be foolish to put humans at risk and then be unable to use the research because we forgot something we could have taken care of upfront. These researchers would bomb most of these questions. The reason for an INDEPENDENT review board is that researchers tend to follow this flow chart: Have idea. ----> Wait, should I do this? ----> Yes, of course!

rhino369 11y ago

Testing on uninformed humans is unethical. Wasn't hackernews just all up in arms about the US military spreading germs to test bioweapons? Isn't this the same exactly thing?

yellowapple 11y ago
There's a pretty significant difference in scale between two nerds putting maybe 2 or 3 vehicles in danger of a dented bumper v. the world's largest military conducting live-fire bioweapon testing.
danielweber 11y ago
There can be times when it is okay to test on uninformed humans. For example, I have relatives who do fire safety. How people do (or don't!) evacuate from buildings when fire alarms go off is a big area of research. The ideal way to test this is to set off the fire alarm in a building where people do not know it is happening, along with some smoke and pyrotechnics. HOWEVER, there are ethical concerns, and a review board would ask questions like: 1. Has anyone else done this study before? If not, why not? How sure are you that no one has done it before? 2. What does the previous research with similar protocols say? What key question are we trying to answer? 3. What is the harm that will be present to people? Are we doing everything we can do to reduce that harm? 4. What more could we do to reduce harm but that might impact the reliability of the research? 5. Quantify how much of a benefit this research would be so we can compare to the risk you are presenting. 6. Demonstrate that you have done all the preliminary work that is necessary to achieve good results, so that we can make sure that the research is used. It would be foolish to put humans at risk and then be unable to use the research because we forgot something we could have taken care of upfront. These researchers would bomb most of these questions. The reason for an INDEPENDENT review board is that researchers tend to follow this flow chart: Have idea. ----> Wait, should I do this? ----> Yes, of course!