At San Francisco electricity prices of ~$0.50/kWh, using an old gaming PC/workstation instead of a lower power platform will cost you hundreds of dollars per year in electricity. The cost of an N100-based NAS gets dwarfed by the electricity cost of reusing old hardware.
Automaking is a capital intensive, operationally complex, and fiercely competitive (read: low profit) industry. Apple's gross margin is nearly 3x higher than Tesla's, and Tesla will face increased margin pressure as more of the industry electrifies. Even if Apple were to match Porsche as the leader in volume+profit luxury automotive, it wouldn't be nearly as lucrative as their current businesses. There are other areas in which Apple could expand to more profitably leverage its core competencies.
Aside from all the difficulties that come with self driving, I suspect Apple cancelled its car effort because they couldn't figure out how differentiate its offering at a price low enough to drive volume and a cost low enough to drive comparable profit to its other businesses.
I hate the Google AI Overview. More of my knowledge-seeking searches than not are things that have a consequential, singular correct answer. It's hard to break the habit of reading the search AI response first, it feeling not quite right, remembering that I can't actually trust it, then skipping down to pull up a page with the actual answer. Involuntary injection of needless confusion and mental effort with every query. If I wanted a vibe-answer, I'd ask ChatGPT with my plus subscription instead of Google, because at least then I get a proper model instead of whatever junk is cheap enough for Google to auto-run on every query without a subscription.
And of course there's no way to disable it without also losing calculator, unit conversions, and other useful functionality.
In Hong Kong and Mainland China, the sidewalks are railed off everywhere except the crosswalk, presumably to prevent the anarchy that occurs when pedestrians are allowed to freely cross the road anywhere.
A little jaywalking is good, a lot of jaywalking renders the road unusable to cars. You don't have to be pro-car or anti-transit to recognize the inefficiency in having roads that are uselessly congested with erratic foot traffic.
Based on my experience taking 10+ APs 10+ years ago, the scores were already inflated. A 4 seemed like it indicated the student might understand the material well enough to build on it with subsequent study in that area, and a 5 raised the odds to probably. Never understood how schools could give credit for 3s, besides language where higher/lower scores could inform placement level in college.
Due to generous financial aid policies, Stanford ends up being free or cost-to-family less than flagship state schools for all but (depending on definition) the professional class and wealthy...
8oz of Starbucks’ bestselling Pike’s Place Roast is 150mg, and between different roasts and sizes, you can buy a drip coffee at Starbucks that has nearly 500mg of caffeine.
Isn’t the stat about STEM grads not working in STEM largely people with degrees in fields like biology (with all of its vocationally limited subfields like evolutionary and population) or chemistry (where you might only be qualified to work as a low paid technician without graduate study)? It seems to me that the difference in employability between a BS in Biology and a BSEE is enormous, both in ease of finding a job as well as career earnings.
This could easily be skewed by self-reporting bias. Perhaps wealthier students feel they have less to potentially lose by answering the survey honestly.
In my experience rubbing shoulders in these circles, kids going to the top tier prep schools are often just really impressive--the culture at these schools normalizes an expectation of high achievement across multiple domains. Not uncommon to see the same kid be a starting quarterback on the football team, an award winning painter, and a nationally ranked debater. I got into an Ivy equivalent college by enthusiastically pursuing my interests at a less competitive high school, and frankly, I think I might have cracked under the pressure at one of these top tier prep schools.
I completely believe that there are many majors/programs at Harvard no more rigorous/educational than those at decent large state universities, but make no mistake, these top tier prep schools are often leaps and bounds more difficult/rigorous than well ranked suburban public high schools, outside of the nationally exceptional magnets like Thomas Jefferson or Bronx Science.
It doesn't even have to be large sums, but the existence of legacy admissions creates goodwill between the university and its alums that broadly motivates consistent, modest donations.
Legacy admissions are also a way to increase yield (percentage of students enrolled versus accepted) which is one of the many ranking-driven stat games.
I'm way too ADHD to hold an engineering job without medication. I'm also quite sensitive to amphetamines. Vyvanse gave me terrible insomnia, and the extended time release worsened the psychiatric side effects (depression, social anxiety, paranoia) when taken in large enough doses to aid focus (which were always pretty low in absolute terms).
Vyvanse basically ruined my life in high school in part because my psychiatrist was afraid to prescribe those pesky abuse-prone IR amphetamines, but I felt I had to keep taking them because it was my only way to succeed academically. Having also been a heavy caffeine user over the years, I think Vyvanse is in an entirely different ballpark.
I'm now managing my ADHD effectively by taking low doses of Adderall on an as-needed basis (average of probably 90 days per year, <10mg per day). Minimal insomnia, minimal depression, no paranoia.
This article is specifically limited to people with employer-sponsored healthcare plans. $20k is a reasonable estimate for a family trying to get decent health insurance independently.
One example of a precaution taken: Spread-spectrum clocking.
Perfectly clocked signals have really high EM spectral peaks (noise radiated at particular frequencies). You can intentionally make the clock a little worse, spreading that radiation over more of the spectrum, and it will reduce the chance that those peaks break other things.
Overpriced relative to the COGS of the hardware, and yet a bargain compared to what you get from any other test equipment company for the same money. The magic is in the software, and I’d much rather pay for that excellent software via expensive hardware than go back to a pre-Saleae world.
There’s a distinction in the US between fully-funded PhD graduate students (who receive a stipend contingent upon research/TA work in an employment relationship) and professional/masters students who pay tuition, even though the latter group is also referred to as graduate students.