Difference between revisions of "Talk:2581: Health Stats"

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
Line 8: Line 8:
  
 
I see this as mainly a joke on how consumer devices often provide more precision than is actually needed, and users don't understand that the extra precision is usually not significant. The only reason "normal" human temperature is 98.6F, to the 10th of a degree, is because average temperature was first measured in Celsius then this was convered to Fahrenheit. But 37C was originally rounded off from an average, so it wasn't precise enough to warrant using an extra decimal place in the conversion. [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 19:23, 15 February 2022 (UTC)
 
I see this as mainly a joke on how consumer devices often provide more precision than is actually needed, and users don't understand that the extra precision is usually not significant. The only reason "normal" human temperature is 98.6F, to the 10th of a degree, is because average temperature was first measured in Celsius then this was convered to Fahrenheit. But 37C was originally rounded off from an average, so it wasn't precise enough to warrant using an extra decimal place in the conversion. [[User:Barmar|Barmar]] ([[User talk:Barmar|talk]]) 19:23, 15 February 2022 (UTC)
 +
:Also a saultory lesson in the misunderstanding of the relationship (i.e. that there really isnt one, at least reliably) between precision and accuracy. Without proper calibration, even moment-to-moment consistency of measurement can be sullied by it being (consistently) wrong, or wrongly read out. Adding more decimals may seem to give a more persuasive estimate, but doesn't do a thing to stop inaccuracy. [[Special:Contributions/141.101.98.225|141.101.98.225]] 19:51, 15 February 2022 (UTC)

Revision as of 19:51, 15 February 2022

Pretty late comic! GcGYSF(asterisk)P(vertical line)e (talk) 06:26, 15 February 2022 (UTC)

It's positive feedback, not negative. Negative feedback is when the response is in the opposite direction of the stimulus (e.g. Cueball becoming more relaxed as blood pressure goes up) and often results in an equilibrium state. A vicious circle (positive feedback) by contrast is when the response ends up increasing the stimulus further, as is the case in this comic.162.158.233.29 11:19, 15 February 2022 (UTC) https://www.albert.io/blog/positive-negative-feedback-loops-biology/#:~:text=Positive%20feedback%20occurs%20to%20increase,back%20to%20a%20stable%20state.172.70.126.221 11:50, 15 February 2022 (UTC)

I feel Cueball's pain. My employer has a checklist where we are supposed to take our temperature every day before coming to work. My problem is, I run hot. My "normal" temperature is usually 99.6ish, not 98.6ish. I knew this for years prior to the pandemic - I used to be a frequent blood donor and would get turned away about a third of the time because they won't take anyone above 99.5. Even though I knew all this, the paranoia induced by daily monitoring and a value that would be abnormal for others but totally typical for me got so bad that I don't do it. No one is enforcing it at the door - it is basically the honor system, and it was causing me more anxiety than actually solving anything. 172.70.114.69 18:54, 15 February 2022 (UTC)

I see this as mainly a joke on how consumer devices often provide more precision than is actually needed, and users don't understand that the extra precision is usually not significant. The only reason "normal" human temperature is 98.6F, to the 10th of a degree, is because average temperature was first measured in Celsius then this was convered to Fahrenheit. But 37C was originally rounded off from an average, so it wasn't precise enough to warrant using an extra decimal place in the conversion. Barmar (talk) 19:23, 15 February 2022 (UTC)

Also a saultory lesson in the misunderstanding of the relationship (i.e. that there really isnt one, at least reliably) between precision and accuracy. Without proper calibration, even moment-to-moment consistency of measurement can be sullied by it being (consistently) wrong, or wrongly read out. Adding more decimals may seem to give a more persuasive estimate, but doesn't do a thing to stop inaccuracy. 141.101.98.225 19:51, 15 February 2022 (UTC)