I was confused about why my average sleep duration over six months was so high, so I decided to investigate. I found out that, for some reason, the six-month calculation doesn’t use the actual number of days but instead uses the magic number 147 to calculate the average.
"This is a little app that showcases how a multi-trillion company like Apple Inc lies to you about your sleep average over six months to make you feel better"
I think a different tone and maybe assume there is good intent instead of immediately claiming it is something malicious (lies by a trillion dollar company) would help with the messaging.
I am curious. But how you present this doesn't really help.
My 6 month data is definitely wrong but it is hilariously more wrong for the time periods I was traveling 6-8 time zones away from home where it's telling me I averaged from 6:05pm to 6:00pm (a cool 24 hours) while showing a graph that has an 8 hour window where I'm not sleeping.
Other than that it's somehow giving me an average of waking up around an hour later than I'd actually wake up.
Apple's sleep data is fucking suspect to start with.
I went through a stage of getting insomnia (or I was woken up by something). I would wake up, get out of bed for 30 minutes, then get back. it would be registered as light sleep stage
I once got the running shits, I was on the toilet shitting my life force out for three hours. It claimed I was only awake for 20 minutes that night.
I saw a lecture years ago where, among other things, the professor demonstrated the iPhone calculator app doing calculations differently depending on if you held it portrait or landscape. I think that was poor design rather than intent to mess up your calculations.
There are so many odd decisions and/or bugs with it..
for example, logging the weight in Kilograms using shortcuts and asking it to read it back while sometimes run into float rounding 75.0000000000001... why not store in grams as an integer!!
Sometimes time in bed is shorter than time asleep!
Can we stop with the "lies" business? This is a very old bug that is inexplicably not fixed, but I highly doubt an Apple employee did this maliciously. For what purpose?
I just checked my own data and there is clearly a bug. The 6 months average is almost a full hour longer than any of the individual preceding 6 months…
Seems like a stretch to assume it’s intentional though.
I have a single night of sleep logged to Apple Health in the last 6-months, and the 6 month average exactly matches that single night. It seems that the author of this app is assuming that Apple is using the wrong denominator when calculating the 6-month average, but I think there could be other explanations for the difference in calculated values. For example how the two approaches might deal with data from different sleep data sources within Apple Health from the same date, how multiple sleep sessions on the same calendar date are merged prior to averaging, etc.
11 comments
[ 4.1 ms ] story [ 24.4 ms ] threadI think a different tone and maybe assume there is good intent instead of immediately claiming it is something malicious (lies by a trillion dollar company) would help with the messaging.
I am curious. But how you present this doesn't really help.
Other than that it's somehow giving me an average of waking up around an hour later than I'd actually wake up.
I went through a stage of getting insomnia (or I was woken up by something). I would wake up, get out of bed for 30 minutes, then get back. it would be registered as light sleep stage
I once got the running shits, I was on the toilet shitting my life force out for three hours. It claimed I was only awake for 20 minutes that night.
for example, logging the weight in Kilograms using shortcuts and asking it to read it back while sometimes run into float rounding 75.0000000000001... why not store in grams as an integer!!
Sometimes time in bed is shorter than time asleep!
Seems like a stretch to assume it’s intentional though.