Ideal maybe, but if they want to maximize revenue they need to optimize around the marginal customer. Marginal customers most likely exhibit strong correlation between hours watched and retention.
For new technologies, adoption curves can vary wildly, and for different reasons. When it comes to clean tech, PV and wind specifically, the adoption is dominated by sunsetting investments over a long horizon.
Other types of tech have much faster adoption cycles due to the lower cost of substitution. For each peak-hour kWh from PV comes at an increased cost of non-peak-hour kWh from non-PV (whether storage or generation).
In my case, I returned three relatively expensive items in a year and I got a warning email from them. Googled the text of the mail and it turns out it's the first step to getting your Prime account banned, and based on the contract they probably wouldn't have to pro-rate the subscription fees.
They know all your purchases, home location, work location (most likely as people deliver to work all the time), and all your returns, as starters. How do they actually make use of it to upsell, etc.? Their recommendation system.
VIX attempts to measure volatility but it is not volatility itself. The lowness of VIX has been driven by the relative increase of the denominator (total asset value, which has surged in the last few years) not by a decrease of the numerator.
In general, it's usefulness as an indicator of volatility has decreased lately.
VIX measures the risk premium on options on the S&P 500 index. I don't see how total asset value ends up in the denominator?
Another explanation is that VIX is being actively dampened by the growing popularity of the 'short volatility' trade. E.g., the SVXY ETF among others. We're seeing today just how volatile volatility can be!
It's actually worse than that. Cities have different water districts and often wealthier areas will receive better water than no-wealth areas. Instead of different prices, everyone pays the same but gets a different product.
The same is generally true for police, fire, and emergency services ("911 is a joke in your town").
I had a stewardess dump a tray full of water on me and my laptop about an hour into a 5 hour flight. She couldn't give me more than a $50 credit, and I had to go to the customer service rep desk after the flight.
First thing he said after I explained to him the situation was "Do you have twitter?". Confused, he explained to me that it was the first thing he had to tick off on the customer resolution checklist. I get it, but a bit of insult to injury.
DHH posted about his bad airline experience[1] in 2013. An app for addressing both your experience and his experience is feasible for the HN crowd. Please steal this business plan[2]. tl;dr Today, getting issues resolved places the burden of effort is the wronged. This can be shifted. "Reversing" things does wonders.
I've almost had hot coffee spilled on my laptop. I've learned to keep an eye out for anyone walking by with anything spillable and hide my laptop. I get that sometimes they can't avoid it with turbulence/people bumping into them/whatever.
I haven't used Facebook in a long time, deactivated my account. I had 2 factor auth on with my phone and they just pinged me twice, back to back, yesterday from their 2 factor auth number asking me to reactivate.
Spamming me from a 2 factor auth number to use your product is a new low I've never encountered elsewhere.
With the 1980 chart: if we should be ensuring high income growth to people according to their income percentile alone then gradually you'd assume they'd move up along the curve (since its income, and not wealth, and income*growth over time would move to the right then).
In this world, everyone would slowly move up the curve according to their tenure/age.
However, if we believe that people should have income growth more dependent on their willingness to work (rather than just income percentile) then over time, with high income growth on the low end, those most willing to work would move up.
In this world, then the income growth is a dependent variable of 'willingness to work' and logically would move up the income percentile over time as labor re-sorts itself into higher income growth percentiles based on that willingness to work.
One simple chart doesn't tell you much. It'd be best to see cohort effects across this time period to see if the bottom percentile is a bunch of 18 year old kids now (millennials), versus a bunch of prime age baby boomers itching to work hard.
Having said that, the cohort effects I've seen are reassuring, but there are still big issues. The main point is that this chart isn't that great at saying anything conclusive.
