@herf

I make f.lux, models, and data to understand circadian and sleep behavior. Previously: Picasa co-Founder/CTO, Google, MetaCreations, CMU.

 California, USA
justgetflux.com/
Joined July 2006
  • Tweets7.9K
  • Following2.3K
  • Followers4.5K
345Media
I just published "Why Standard Time is better" link.medium.com/tgYAui3tdV

Why Standard Time is better

We have to stop changing the clocks twice a year, but it’s also important to pick the right schedule when we do it.

herf.medium.com
18
58
28
126
Didn't we give up general-purpose computing so our phones wouldn't overheat?
3
518
iOS measures "time in daylight" for myopia, though advice is strange. Myopia intervention studies adjust "time for recess". 120 minutes outdoors at school is not the same as 80-120 minutes total & also the dose-response goes past 2hrs/day. See ncbi.nlm.nih.gov/pmc/article…
3
538
Lots of good arguments here. But mainly - CPI is getting overused as "affordability" when it's not... doesn't consider the cost of debt, and the economy recently is based on debt. e.g., car prices can go down while monthly payments go higher. open.substack.com/pub/mattst…
237
I’m a fan of normal portrait lengths, so feels like something’s missing from the iPhone 15 pro max—an optic between 24mm and 120mm. That means you’re shooting 3x zoom using 10% of the sensor (1-4 megapixel depending how you count AI/quad Bayer stuff, but it’s not much!)
1
657
There should really be a kind of contract law for "building on someone else's platform" - by now there are enough Unity/Google Maps/etc. rug pulls that you could imagine what platform owners might reasonably owe people who put years of work into them. Otherwise it's only serfdom.
2
4
471
Also, 72% increase before lifestyle corrections (since chronotype has a large impact on lifestyle).
American Academy of Sleep Medicine
@AASMorg14 Sep 2023
Investigators from @BrighamWomens found evening #chronotype, or going to bed late and waking up late, was associated with a 19% increased risk of #diabetes after accounting for lifestyle factors. ow.ly/uMhS50PLBzu
261
Unity: "we are going to track reinstalls and not charge you"... bizarrely ignoring last 6 years of Apple privacy changes. theverge.com/2017/4/23/15399…

Uber tried to fool Apple and got caught

Uber geofenced Apple’s Cupertino headquarters to hide that it was tracking iPhones

theverge.com
203
Limiting iPhone 15 to USB 2.0 speeds takes me back 20 years... when we had 180nm CPUs (rather than 3nm) and read Compact Flash cards from our DSLRs at 30MB/sec, the *exact same speed* you can use to backup your iPhone 15 next week.🤨20 years, it's amazing.
1
2
323
The only thing is that you might get by with a 512MB CF card back then. And now your iPhone can be 512GB. So it's about a thousand times slower.
181
Locking software crop to the Pro model, I see what you did there.
1
1
284
In case this is vague: The iPhone 15 pro features 28mm and 35mm modes, but these are crops of the 48MP sensor. "Cropping" was never an advanced technology, and the same feature would work just as well on 14 pro and the basic iPhone 15, which all use a 48MP sensor.
2
314
There's basically no amount of money you can pay a developer (outside Apple) to say "set my iPhone alarm to wake me up on weekdays but please *skip* national holidays".
1
287
The mistake in this Cochrane review is letting marketing dictate what scientific terms mean. Here "blue light filtering" means "UV filtering plus a few percent of blue". The plain-language meaning of "blue light filtering" is not met.
Health Nerd@GidMK21 Aug 2023
Really interesting new Cochrane review out looking at blue-light blocking glasses and health There were over a dozen RCTs into the question, but most of them were terrible The one decent trial found no benefit cochranelibrary.com/cdsr/doi…
1
3
2
966
If you are tempted to say "oh but they filter 10%" (which is still nonsense) please consider that most lights in the studies are LED, and the effect is truly minimal for these sources.
140
Want to reduce C0VID transmission in #schools this fall? Based on the new #ASHRAE Standards, this is the chart I'm recommending families and collective bargaining units use to lock in clean air. 1/15
Simplified Recommendations for Classroom Air Cleaning to Reduce the Risk of Far-Field Airborne Infectious Disease Transmission, Derived from ASHRAE Standards 241 & 62.1 TABLE. Educational Facility Type, followed by cfm/person, then ACH Libraries 40 2.7 Art classroom 40 5.3 Wood and metal shop 40 5.3 Computer lab 40 6.7 Media center 40 6.7 Science labs 40 6.7 University and college labs 40 6.7 Daycare, ages 4 and under 40 6.7 Classroom, ages 5-8 40 6.7 Classroom, ages 9+ 40 9.3 Music, theater, and dance 40 9.3 Lecture classroom 40 17.3 Multiuse assembly 40 26.7 Lecture hall, fixed seats 50 50.0 Note. cfm/person based on ASHRAE Standard 241. ACH derived from typical full-capacity person density estimates from ASHRAE Standard 62.1. Use the most comparable educational facility. If half or one-quarter capacity, prorate the ACH estimate accordingly. Calculations focus on reducing risk primarily of far-field transmission in rooms with well-mixed air, and masks remain needed...

ALT Simplified Recommendations for Classroom Air Cleaning to Reduce the Risk of Far-Field Airborne Infectious Disease Transmission, Derived from ASHRAE Standards 241 & 62.1 TABLE. Educational Facility Type, followed by cfm/person, then ACH Libraries 40 2.7 Art classroom 40 5.3 Wood and metal shop 40 5.3 Computer lab 40 6.7 Media center 40 6.7 Science labs 40 6.7 University and college labs 40 6.7 Daycare, ages 4 and under 40 6.7 Classroom, ages 5-8 40 6.7 Classroom, ages 9+ 40 9.3 Music, theater, and dance 40 9.3 Lecture classroom 40 17.3 Multiuse assembly 40 26.7 Lecture hall, fixed seats 50 50.0 Note. cfm/person based on ASHRAE Standard 241. ACH derived from typical full-capacity person density estimates from ASHRAE Standard 62.1. Use the most comparable educational facility. If half or one-quarter capacity, prorate the ACH estimate accordingly. Calculations focus on reducing risk primarily of far-field transmission in rooms with well-mixed air, and masks remain needed...

42
615
68
1,429
164,725
Making people wake up before the sun continues to be a bad idea. Here it increases suicide.
htw@heniek_htw20 Jul 2023
Longitude-based time zone partitions and rates of suicide Across the U.S., western partitions had statistically significantly higher rates of #suicide compared to eastern partitions and averaged up to two additional yearly deaths per 100,000 people. sciencedirect.com/science/ar…
3
417
A decade ago, Facebook let companies advertise to people who liked a competitor's page or group, and it was very cheap to do this. I'm sure it was profitable enough in the short term, but *many* brands stopped making pages & groups. I wonder what the long term cost was.
2
223
Interesting - *on the night shift*, early birds get c19 at a much greater rate than night owls. Wonder if it's the reverse for daytime jobs? pesquisa.bvsalud.org/global-…
1
1
186
It's important that Parental Controls has a "disable feeds and recommendations" setting, which doesn't currently exist. 1/n
2
1
3
587
more replies
This "feeds" rule clearly applies to most social apps, Youtube/Netflix, but the details are hard to map. Would it apply to "top ten lists" or automatic music playlists? I think an important line is algorithmic suggestions vs. human-generated ones. 4/n
1
186
Recent laws (e.g., in California) that make companies liable for harm to kids have no specifics, but "no feeds or recommendations" is an important one. Let parents turn this stuff off, for the whole device at once. 5/n
166
Load more