17 ms·

It's difficult because ontologically you can never actually prove something is "safe". It's proving a negative (never any harmful effects). Some harms are subt

by zxcmx 7y ago

It's difficult because ontologically you can never actually prove something is "safe". It's proving a negative (never any harmful effects). Some harms are subtle or situation dependent, and small effects will only show up under widespread or frequent use. Perhaps deliberately slowed or phased rollout could help. It would also make materials science and the chemical industries "feel" a lot more like medicine. Companies would have to spend billions on studies to get new chemical formulations or materials to market which could have similar effects on innovation (and monopoly) to those we see in the drug market. Not necessarily bad depending on what you value but the effects would be very far reaching.

lukifer 7y ago
Granted. I mean, strictly speaking, we can't prove that the gravitational constant won't change tomorrow. :) But I would like a substantial enough body of evidence from competent and disinterested assessment(s) so that our societies can take the right long-terms risks on new materials (where "new" = less than 100 years). Obviously easier said than done; I like the idea of putting it in the same category as medical treatments (though let's be frank, even that process is hardly flawless).
- jchanimal 7y ago
  We only discovered recently that baby powder causes cancer, and it’s not “new”.
colechristensen 7y ago
Shrug. You can define "safe" as an absolute and then argue absolute is impossible, yes. It's a bit of a straw man argument though. Or you can drop the binary label and start developing statistics about known effects and outcome ranges. These things can be done; further understanding of bio-chemical systems and simulation makes discovery by widespread experiment less and less extant. The issue isn't that looking for and finding effects positive or negative is hard, the issue is that few are trying and few are interested in the effort. (i.e. most people either have the conservative, 'everything is safe' attitude or the luddite 'everything is bad for you' attitude, and neither involve much discovery)
- tolmasky 7y ago
  You can’t drop the binary label if your intent is to do something with the information. Ultimately there will be a “safety” threshold at which these materials would be allowed or not. If the only statement is “we should generally be studying everything more, in an undirected manner with no stated purpose aside from the inherent value of simply knowing more about everything and letting each individual draw their own conclusions”, then yeah sure, but there’s nothing stopping this from happening now, aside from the fact that such a mission statement doesn’t provide particularly strong incentives.
  - rcxdude 7y ago
    Generally when it comes to hazardous materials, there's not a binary 'banned/OK'. There can be restrictions on use for given applications, restrictions on purchasing, etc. Rarely are materials completely banned outright.
    - tolmasky 7y ago
      This feels like a bizarre straw man. Yes, you can use lead paint on road markings because no one will lick them and lead paint is amazingly resilient to water damage -- but I think we all have a shared understanding in a casual conversation about what we mean when we say "lead paint is banned", there's no need for someone to say "well actually -- that's not strictly true, current regulations still allow for non-consumer use and old buildings are still grandfathered in if they have not been repainted, and additionally its not an outright ban but like most substances it means it has to fall under a certain percent content". Either way, these are still ultimately binary restrictions in their respective domains. It's a cascade of if statements, not a smooth function. Just because there are twenty binary decisions, or the decisions are in separate categories, does not make them non-binary. A non-binary approach would be something like having a gradient of punishments for a smooth spectrum of allowed percentages. It's hard to properly describe, because again, it's not really often seen.

by zxcmx 7y ago

lukifer 7y ago

Granted. I mean, strictly speaking, we can't prove that the gravitational constant won't change tomorrow. :) But I would like a substantial enough body of evidence from competent and disinterested assessment(s) so that our societies can take the right long-terms risks on new materials (where "new" = less than 100 years). Obviously easier said than done; I like the idea of putting it in the same category as medical treatments (though let's be frank, even that process is hardly flawless).

jchanimal 7y ago
We only discovered recently that baby powder causes cancer, and it’s not “new”.

colechristensen 7y ago

Shrug. You can define "safe" as an absolute and then argue absolute is impossible, yes. It's a bit of a straw man argument though. Or you can drop the binary label and start developing statistics about known effects and outcome ranges. These things can be done; further understanding of bio-chemical systems and simulation makes discovery by widespread experiment less and less extant. The issue isn't that looking for and finding effects positive or negative is hard, the issue is that few are trying and few are interested in the effort. (i.e. most people either have the conservative, 'everything is safe' attitude or the luddite 'everything is bad for you' attitude, and neither involve much discovery)

tolmasky 7y ago
You can’t drop the binary label if your intent is to do something with the information. Ultimately there will be a “safety” threshold at which these materials would be allowed or not. If the only statement is “we should generally be studying everything more, in an undirected manner with no stated purpose aside from the inherent value of simply knowing more about everything and letting each individual draw their own conclusions”, then yeah sure, but there’s nothing stopping this from happening now, aside from the fact that such a mission statement doesn’t provide particularly strong incentives.
- rcxdude 7y ago
  Generally when it comes to hazardous materials, there's not a binary 'banned/OK'. There can be restrictions on use for given applications, restrictions on purchasing, etc. Rarely are materials completely banned outright.
  - tolmasky 7y ago
    This feels like a bizarre straw man. Yes, you can use lead paint on road markings because no one will lick them and lead paint is amazingly resilient to water damage -- but I think we all have a shared understanding in a casual conversation about what we mean when we say "lead paint is banned", there's no need for someone to say "well actually -- that's not strictly true, current regulations still allow for non-consumer use and old buildings are still grandfathered in if they have not been repainted, and additionally its not an outright ban but like most substances it means it has to fall under a certain percent content". Either way, these are still ultimately binary restrictions in their respective domains. It's a cascade of if statements, not a smooth function. Just because there are twenty binary decisions, or the decisions are in separate categories, does not make them non-binary. A non-binary approach would be something like having a gradient of punishments for a smooth spectrum of allowed percentages. It's hard to properly describe, because again, it's not really often seen.