3 ms·

Are you suggesting that they take phone video of all their experiments over the course of multiple years? Or just the one they know will succeed?

by mhb 3y ago

lambdasquirrel 3y ago
A lot of science takes a mind-numbing amount of time at the lab desk. Something like 40 hours a week just dedicated to picking apart fruit flies with tweezers under a microscope is pretty common. I think I read somewhere that with MatSci, mortars and pestles and blisters are involved. It would be a lot of recording that would have to happen, and then you have to record every setting on every instrument you touch. How do we know that that's actually that reagent in that bottle? Yeah this just doesn't hold up for actual science. It's not something you can just take a Zoom recording of.
- fullspectrumdev 3y ago
  > mortars and pestles and blisters are involved Can confirm from my life prior to working with computers. “Is this mortar and pestle actually clean?” “How much grinding is the right amount?” “Is the window open or closed?” “Did someone fuck with the settings on this machine since I last used it?” And so on…
DannyBee 3y ago
I'm sorry, you really think it's that hard to have a camera running in a lab that records footage continuously, and you save the clips where you succeed? What percent of the cost of running these experiments do you think that is to buy a nest or a unifi camera? Heck, Do you believe the university doesn’t already have a surveillance system almost certainly in the hallway to the lab. Plenty of them even have cameras in every lab already. In the past two weeks, I was in three material science labs in Georgia, and South Carolina and all of them had cameras in every lab. This is not that hard. If you wanna argue that it’s not helpful, that’s one thing. But arguing that it is difficult seems very silly.
- mhb 3y ago
  Ah, another example of how something is so easy when you just imagine other people doing it. Your aspiration, I suppose, is that this (e.g., surveillance?) video will document the experimental activities in enough detail so that they can be replicated and/or convincingly proven. As a practical matter, it's ludicrous to think that any of the examples that you suggest as obvious and trivial would do either of these things much less survive a cursory consideration of their cost/benefit. Maybe you should prototype your system using a Nest camera in the corner of a lab and demonstrate reading the label on a critical reagent from the video. Further, this is a classic example of premature optimization. Should only labs that are confident that they will be contesting for a Nobel prize be monitoring all their experiments or do you also envision this as an activity for labs with lesser aspirations or self-esteem?
  - DannyBee 3y ago
    What makes you believe I haven't done this before? If you want to know, maybe, you know, ask ("Have you ever tried to do this before in a lab?" - this is also not hard, and even the people not contending for the Nobel prize can do it), rather than just assert that this is just something i only want to see others do in my imagination, which is not just rude, but actually, it turns out, wrong. But hey, don't let that stop you. Please, continue to assert things about me rather than actually ask or otherwise use means that might gain knowledge of me and what i have or haven't done. To answer the rest, so, actually, no, that's not my aspiration. My aspiration was "producing and testing", specifically targeted at "early stage believability". Which is what I said. You seem to have decided that means "be able to show all pieces in a convincing way" or "make a how to video". That would be silly, which is why i didn't say that. This is is mostly about showing a video of the end result in a meaningful way, as the literal top thread on this on HN now does, posting multiple reproduction and testing videos. If you can show production in some way at all, awesome. I'm not interested in the residents cynics shitting on every single way that it could be faked, nor are most other people. To most people, those videos are much more convincing than all the papers and anything else you produce. As for who monitors, as I mentioned elsewhere, lots of labs already have cameras, because universities now put them in almost every room. As I mentioned, I was even in multiple material science labs in the past week, and all had cameras. But please, feel free to continue to pretend that this is some super complicated thing to do when it's already being done and being used to good effect, convincingly, on HN, right now. I guess they all just contending for a nobel prize, because otherwise, this is clearly impossible, despite it literally having been done.

Are you suggesting that they take phone video of all their experiments over the course of multiple years? Or just the one they know will succeed?

by mhb 3y ago

Are you suggesting that they take phone video of all their experiments over the course of multiple years? Or just the one they know will succeed?

lambdasquirrel 3y ago

A lot of science takes a mind-numbing amount of time at the lab desk. Something like 40 hours a week just dedicated to picking apart fruit flies with tweezers under a microscope is pretty common. I think I read somewhere that with MatSci, mortars and pestles and blisters are involved. It would be a lot of recording that would have to happen, and then you have to record every setting on every instrument you touch. How do we know that that's actually that reagent in that bottle? Yeah this just doesn't hold up for actual science. It's not something you can just take a Zoom recording of.

fullspectrumdev 3y ago
> mortars and pestles and blisters are involved Can confirm from my life prior to working with computers. “Is this mortar and pestle actually clean?” “How much grinding is the right amount?” “Is the window open or closed?” “Did someone fuck with the settings on this machine since I last used it?” And so on…

DannyBee 3y ago

I'm sorry, you really think it's that hard to have a camera running in a lab that records footage continuously, and you save the clips where you succeed? What percent of the cost of running these experiments do you think that is to buy a nest or a unifi camera? Heck, Do you believe the university doesn’t already have a surveillance system almost certainly in the hallway to the lab. Plenty of them even have cameras in every lab already. In the past two weeks, I was in three material science labs in Georgia, and South Carolina and all of them had cameras in every lab. This is not that hard. If you wanna argue that it’s not helpful, that’s one thing. But arguing that it is difficult seems very silly.

mhb 3y ago
Ah, another example of how something is so easy when you just imagine other people doing it. Your aspiration, I suppose, is that this (e.g., surveillance?) video will document the experimental activities in enough detail so that they can be replicated and/or convincingly proven. As a practical matter, it's ludicrous to think that any of the examples that you suggest as obvious and trivial would do either of these things much less survive a cursory consideration of their cost/benefit. Maybe you should prototype your system using a Nest camera in the corner of a lab and demonstrate reading the label on a critical reagent from the video. Further, this is a classic example of premature optimization. Should only labs that are confident that they will be contesting for a Nobel prize be monitoring all their experiments or do you also envision this as an activity for labs with lesser aspirations or self-esteem?
- DannyBee 3y ago
  What makes you believe I haven't done this before? If you want to know, maybe, you know, ask ("Have you ever tried to do this before in a lab?" - this is also not hard, and even the people not contending for the Nobel prize can do it), rather than just assert that this is just something i only want to see others do in my imagination, which is not just rude, but actually, it turns out, wrong. But hey, don't let that stop you. Please, continue to assert things about me rather than actually ask or otherwise use means that might gain knowledge of me and what i have or haven't done. To answer the rest, so, actually, no, that's not my aspiration. My aspiration was "producing and testing", specifically targeted at "early stage believability". Which is what I said. You seem to have decided that means "be able to show all pieces in a convincing way" or "make a how to video". That would be silly, which is why i didn't say that. This is is mostly about showing a video of the end result in a meaningful way, as the literal top thread on this on HN now does, posting multiple reproduction and testing videos. If you can show production in some way at all, awesome. I'm not interested in the residents cynics shitting on every single way that it could be faked, nor are most other people. To most people, those videos are much more convincing than all the papers and anything else you produce. As for who monitors, as I mentioned elsewhere, lots of labs already have cameras, because universities now put them in almost every room. As I mentioned, I was even in multiple material science labs in the past week, and all had cameras. But please, feel free to continue to pretend that this is some super complicated thing to do when it's already being done and being used to good effect, convincingly, on HN, right now. I guess they all just contending for a nobel prize, because otherwise, this is clearly impossible, despite it literally having been done.