3 ms·

Someone just needs to create an institution that does nothing but verify the research/experiments of others.

by JakeAl 3y ago

golemotron 3y ago
The question is who would pay for that. If the research results look commercially viable it might be better to just try them out in the market. Medical and pharmaceutical might be the exception because of risk and regulation.
- kurthr 3y ago
  I question even this case, because Pharma seem perfectly happy to push through drugs with unproven effectiveness. See the Aduhelm scandal (which has expanded to the entire Beta-Amyloid research angle). It wouldn't be so bad (see the Vitamin Industrial Complex), if they weren't paid for by insurance (ie everyone).
- hospitalJail 3y ago
  Easy: Create and organization, crowd fund it. You can knock out a few studies for free/low cost. Hard: Can we form an intellectual union for this(professionals/10%ers)? Protects us from the mob and the ruling class + we can collect dues for things that are incredibly important like this.
kurthr 3y ago
I often remark that it's so easy to fail to replicate any moderately complex experiment, so you really have to do a great job to be sure that it's actually invalid. Whether it's the chemistry, equipment, or animals (even yeast) involved there are so many variables that need to be controlled only an expert can do it. If you think that the Methodology and experimental setup will (or even could) be sufficient, you're in for a rude surprise. Now there's now so much low hanging fruit, that it might be possible for a few labs to just look at the most outrageous published experimental claims for replication. Especially, anything published in a prestigious journal or groups with lots of influence could be carefully examined, but of course you have to do a great job to get your negative results published... it's also worth mentioning that the reputational risks of doing this are not small. If you're already being pushed out of academia and love working for peanuts though?

Someone just needs to create an institution that does nothing but verify the research/experiments of others.

by JakeAl 3y ago

Someone just needs to create an institution that does nothing but verify the research/experiments of others.

The question is who would pay for that. If the research results look commercially viable it might be better to just try them out in the market. Medical and pharmaceutical might be the exception because of risk and regulation.

kurthr 3y ago
I question even this case, because Pharma seem perfectly happy to push through drugs with unproven effectiveness. See the Aduhelm scandal (which has expanded to the entire Beta-Amyloid research angle). It wouldn't be so bad (see the Vitamin Industrial Complex), if they weren't paid for by insurance (ie everyone).
hospitalJail 3y ago
Easy: Create and organization, crowd fund it. You can knock out a few studies for free/low cost. Hard: Can we form an intellectual union for this(professionals/10%ers)? Protects us from the mob and the ruling class + we can collect dues for things that are incredibly important like this.

kurthr 3y ago

I often remark that it's so easy to fail to replicate any moderately complex experiment, so you really have to do a great job to be sure that it's actually invalid. Whether it's the chemistry, equipment, or animals (even yeast) involved there are so many variables that need to be controlled only an expert can do it. If you think that the Methodology and experimental setup will (or even could) be sufficient, you're in for a rude surprise. Now there's now so much low hanging fruit, that it might be possible for a few labs to just look at the most outrageous published experimental claims for replication. Especially, anything published in a prestigious journal or groups with lots of influence could be carefully examined, but of course you have to do a great job to get your negative results published... it's also worth mentioning that the reputational risks of doing this are not small. If you're already being pushed out of academia and love working for peanuts though?