Hacker News new | ask | show | jobs
by gruez 1916 days ago
1. people will eventually tune them out, like with prop 65 warnings or the existing nutrition facts/calorie labeling

2. while it's easy to calculate what's the nutritional content in a food, estimating future failure rates isn't trivial and there's a lot of subjectivity involved. Companies will definitely be fudging the reliability numbers to get an edge. See for instance, the failure rates for hard drives. The annual failure rate on the spec sheets are around 0.3%, but empirical data by backblaze puts them anywhere from 0.3% to 12%. Therefore I'd expect these nutritional fact labels to be totally useless at best, and a waste of time/resources at worst.
2 comments
jfim 1916 days ago
Prop 65 warnings are pretty useless though, since they have very limited information that does not allow one to evaluate the risk incurred.

Case in point, my first internship in California was in a building with a sign that said "This building contains chemicals known to the state of California to cause cancer and birth defects or other reproductive harm." What's in the building? Who knows. Could be really bad chemicals, or just someone who has a beer on their desk [0].

It would be much better to have some information about the chemicals contained, how bad are the chemicals, and what is the expected effect of the chemicals at the concentration at which they're encountered.

[0] https://oehha.ca.gov/chemicals/alcoholic-beverages-0

link
ficklepickle 1916 days ago
Yes, they need some actionable information. I remember seeing my first one as 15 year old Canadian on vacation. My first thought was good thing I'm in Hawaii. My second thought was I can't do anything with this vague information.

It was some green slime you put in your bike tires to prevent puncture leaks. I had never seen that before. I bought it and took it home with me, skillfully avoiding California so it didn't become carcinogenic.

link
guerrilla 1916 days ago
So it was better when it was impossible to know what was in things and count calories? Not perfect doesn't mean not better.

> The annual failure rate on the spec sheets are around 0.3%, but empirical data by backblaze puts them anywhere from 0.3% to 12%.

I'd suggest dealing with that as fraud, not giving up.

link