the mechanics of cooperation probably scale better than those of defection, by their nature. Defectors need to pay higher costs guarding themselves against the other defectors, and always trying to figure out how they are going to pick the right defections for themselves to win.
That's not known for sure. The game theory simulations go one way or the other depending on what assumptions you use. I'm not sure you can say "probably" there.
I'm thinking of slightly more esoteric concerns, in recent simulations I've seen some evidence that something like the All Defector from The Fractal Prince, a book in a series recommended in this thread, might be possible.
If you're confused by a widespread practice, consider that it doesn't have the purpose you assume it has. In the case of tech interviews, it's plausible they're selecting for willingness to be abused by managers.
There was the zenphone 9/10 in the same size, which also sold poorly enough that the 11 is a regular huge phone. Other than the same thousands (not even millions) of us complaining online year after year there's just no demand.
Sometimes I look at a product from a major company and wonder if I'm taking crazy pills. Why am I as some rando able to instantly notice stupid design? Don't they have teams of people examining obvious failures? Butterfly keyboard just for the apple example.
Put the damn battery on the back of the head to prevent neck problems you lunatics.
it's partially driven by scale. Take a good product and have it get accolades on the review sites and on reddit. The flood of sales triggers the manufacturer to find ways to cut costs. Saving ten cents doesn't matter when you sell ten thousand of something. But when you sell ten million, well you can hire people whose whole job it is to find ten cent cuts. Each cut itself isn't really noticeable, but eventually they add up. I learned about this from running a food company. Our competitors make their protein bars for, I'm told by a long time industry insider, 17 cents in ingredients.
That also involves insurance companies. I wonder how many of them have "exposure" on both sides of the medical business. So long as they can raise premiums/limit coverage they can seek rents. US medicine needs first and foremost transparency at every level. Transparency is in many ways obstructed by HIPAA. Sure, I wouldn't want even my largely unremarkable medical history plastered all over the Internet along with my more or less personally identifiable information or mined to target/discriminate against me, but there has to be a better way where we can use the information to lower costs and improve outcomes.
I highly recommend low profile switches for those who haven't tried them. They weren't really available in many models until recently and they're a big step up from standard switches if you're more used to laptops.
> they're a big step up from standard switches if you're more used to laptops
In what way? I was contemplating a keyboard with mechanical low profile switches. One of the reasons was that I could easily remap the keys without having to buy new keycaps, since they all have the same profile, and for some reason, every manufacturer insists on their own home / en / pg up / pg dn layout (I'm looking exclusively at "75%" and similar layouts, which put those keys in a column).
But I'm hesitant because I've trained myself to not bottom out when I type with my regular mechanical keyboards, and my fingers enjoy the lack of abrupt stop at the bottom.
One switch I've not seen mentioned here yet is the Kailh Sunset switch [0], developed in collaboration with Darryl from lowprokb.ca. It is my favourite low profile switch, and the closest in feel to my X220-and-previous-generation Thinkpad keyboards.
Low profile switches are not compatible with MX keyboards, because all the dimensions are different. There are “mid profile” switches made by GMTX and perhaps others, which can be swapped into an MX board. (The Kailh mid-profile switches are not MX compatible.) You might have to be careful about keycap compatibility, to avoid longer caps that might bottom out on the switch plate.
Strangely, if you use an alt keyboard layout, dropping most of the vowels actually makes it a little less comfortable to type. Coleman and (particularly) Dvorak both put a lot of effort into making common words either alternate hands or roll off your fingers, as well as making the most common letters and vowels very accessible to your fingers. Since this method of typing makes those convenient letters less common, it seems to actually make tpng lk ths a ltl + uncmftbl.
