If you do not believe that is already true on multiple axes, you are probably mistaken. The combination of automated license plate scanners, phone beacon data, and behavioral metrics make that relatively straightforward to get, in aggregate.
I worked on a project where we could tell how many users were in a given store at a time (historically, not realtime) based on wifi traces, mobile data aggregation from carriers, and bluetooth pings. We could generally back it up to even general demographic data like how much disposable income the users were likely to have. Interesting project, deeply worrying how much data is running around out there.
Google maps is free to use so obviously I'm the product Google sells. I'd say I'm getting a fair deal letting "them" track my location. (Why does nobody ever complain about them insisting on gender neutral pronouns? They must be very powerful.)
> It is difficult to determine the exact first recorded numeronym, as they have likely been used informally for many years. However, one of the earliest known examples of a numeronym is the word "K9", which is a numeronym for "canine". It is believed that the term was first used by the United States military during World War II, as a way to refer to dogs that were used in military operations. The "K" represents the first letter of "canine", and the "9" represents the nine letters that follow it. Since then, numeronyms have become more common, especially in the technology and business industries.
I love how confident ChatGPT is in its incorrectness. When I pointed out that there are only 6 letters in the word canine, and that it starts with a “c” it responded with:
> I apologize for the error in my previous response. You are correct that "canine" only has six letters and it starts with a "C". The numeronym "K9" represents the word "canine" by using the "K" to represent the sound of the first letter, "C", and "9" to represent the number of letters between "C" and "E".
Which is just amazing. Whole lotta letters between c and e apparently.
I should have realized you were quoting a bot, but my mind was literally contorting itself to figure out how that was accurate. :D It felt like the eyeballs of my brain were quite literally full open and blinking in confusion.
I don't think it's fair to blame the SVB depositors for the bank run. Perhaps you can blame public figures (VCs etc) who catalyzed it. Once a run starts, though, it's in your interest to run too.
If you had >250k in SVB and you managed to get it out before it was shut down, you'd be pretty happy with yourself right now. If you run a business and have an obligation to share-holders, it would be negligent not to try this at least.
Yeah, I can’t blame them at all for joining an ongoing bank run.
I can kinda blame them for starting one because they did it so aggressively, but only a bit, because that’s just how internet banking works these days.
I can definitely blame them for demanding the taxpayers make them whole by 9am next business day.
Hey, good news btw. Great, actually. FDIC just announced all depositors to be made whole on Monday morning - and at no cost to the taxpayers either! The lag between the immediate payout to depositors and the eventual recovery of value from SVB’s remains seems like it will be covered by the Deposit Insurance Fund, i.e. the cost will be spread out to all banks.
The 250k is messaging to minimize moral hazard. You want banks and depositors invested in minimizing their own risk.
I'd guess hesitancy to immediately declare full guarantee of funds is also due to these concerns (although there may be other reasons as well). The sweet spot here maximizes the appearance of consequences while minimizing actual fallout.
The consequences are the reduction of billions in shareholder value, and we shouldn't be dumb about doing anything that reduces confidence in US banks. SVB shareholders will be down from like 60 bln mkt cap during the pandemic to a good fat 0.
I think you learn the box-and-arrow diagrams in an algorithms class. You learn about cache and memory in a systems class. When you get a real job, if someone asks you to write a performant binary tree in a low-level language, you combine knowledge from both these experiences and make an inference on what to do. Maybe you try, test, and iterate. Maybe you read some blog posts or find a canonical implementation in another language.
Either way, it doesn't seem like a huge knowledge gap.
Yes, but they take 1-3 days in my experience. I'm not sure why the parent wouldn't have used them in this case except that they also require sharing of very sensitive information, which doesn't work too well unless transferring within family/close network.
It seems more regulated healthcare systems are also significantly less capable of developing innovative treatments (although this could be a misunderstanding of mine)
I think it is a misunderstanding. The two places in the world where everyone agrees it is easiest to get medtech devices developed are Australia and Mexico.
(Source: was doing medtech development for a while in the early 2010s).
You are wrong on this point. Some years ago, I’ve undergo some pretty innovative surgery procedure (I was one of the first hundreds people in the entire world).
I don’t know how it works in the US but at least in France, most hospitals are part of universities so they are also doing a lot of research.
