> The merchants think so, too, or they wouldn’t accept the processors that let their banks hand out these cards.
Visa / Mastercard / American Express all have lines of premium credit cards (Visa Infinite, World Elite Mastercard, Amex Platinum), and they're very much too big to ban. You'd just be left with one processor in the US (Discover, now owned by Capital One).
This is wrong. Merchants can elect to only accept debit cards.
In recent years, all of my utilities have added 3%+ credit card surcharges, so I pay most of my household expenses with debit cards/ACH now.
Tmobile, Comcast, Verizon, ATT, Target, grocery store, electric utility and water utility (government), annual vehicle tax (government), auto body shop, daycare, and any home repair contractors all charge 3%+ (or give a discount, same thing), so I basically only use credit cards for other retail stores and travel and restaurants.
Shockley might have been a better researcher/engineer than manager, still I'm sure a movie about his early days, including the alienation of some of the people he recruited later, would be an interesting watch :)
Hilariously, "os as g" adds one more byte than it saves, since os is only used 4 times but the alias takes 5 extra bytes to save 4. And "socket as s" comes out even.
If you wanted real savings, you'd use "d=bytes.fromhex" instead of defining a function -- 17 bytes!! And d('00') -> b'\0' for -2 bytes.
We could easily get the byte count down further by using base64.b85decode instead of bytes.fromhex (-70 or so), but ultimately we're optimizing a meaningless metric, as you mention.
Indeed; most personal banking customers can fall back on FDIC insurance ($250k should be more than enough to cover your emergency fund). This isn't the 1920s.
It sure isn’t the 1920s, it’s the 2020s so things like digital money are ephemeral and whimsical.
The bigger question is how much food and medicine is there in the supply chain buffers? If all production was to stop immediately — how many calories are on the continent? How many grams of insulin or penicillin?
In a crisis how will those things be distributed? Will it be based on immediate need or social class?
What’s keeping the system going anyways? Why do ships continue to come with consumer goods from China? Why do farmers send their grain to market?
It’s kind of neat to think about what will happen in this sort of scenario. I wonder how long the data centres will keep running, churning out models that don’t have a market an aren’t quite good enough for AGI.
Alas, for Silicon Valley Bank they went with 'too big to fail' and also covered uninsured deposits. That's moral hazard and endangers the core purpose of the insurance.
Agreed. That said, FDIC would have not been able to cover all $150 billion or so of uninsured SVB deposits directly from the insurance fund, so had that been the only available option for making depositors whole, then FDIC would have had to pass.
Well, insurance should only covered insured deposits.
> [...] so had that been the only available option for making depositors whole, [...]
On paper, FDIC might be independent and have its own balance sheets. But in practice and given politics, FDIC itself can't fail / isn't allowed to fail. It'll always be bailed out, and that's what the market expects.
For the stability of the economy, it would have been better not to make uninsured depositors whole.
https://en.wikipedia.org/wiki/Happy_Eyeballs is the usual name. It's not quite identical, since you often want to give your preferred transport a nominal headstart so it usually succeeds. But yes, there are some similarities -- you race during connection setup so that you don't have to wait for a connection timeout (on the order of seconds) if the preferred mechanism doesn't work for some reason.
Request hedging or backup requests are indeed the terms
I know for requests where you give the first request a bit of a headstart. I didn’t know about the term happy eyeball to signify that all requests fire at the same time.
> I didn’t know about the term happy eyeball to signify that all requests fire at the same time.
It's not quite the same. Usually with Happy Eyeballs, you want to try multiple protocols (e.g. QUIC vs TCP, or IPv6 vs IPv4), and you have a preference for one over the other. As such, you try to establish your connection via IPv6, wait something like 30ms, then try to establish via IPv4. Whichever mechanism completes channel setup first wins, and you can cancel the other one.
It's a mechanism used to drive adoption of newer protocols while limiting the impact on end users.
The actual explanation starts a couple minutes later, around https://youtu.be/KKbgulTp3FE?t=1553. The short explanation is performance (essentially load balancing against multiple RAM banks for large sequential RAM accesses), combined with a security-via-obscurity layer of defense against rowhammer.
Eh. It depends what your bottleneck is. If the bottleneck is now, say, CPU cache contention because you've doubled your thread count, it's entirely possible that FL1 running on the new server generation is operating in a different regime than on the previous generation. You can see some hints of that happening, since doubling thread count didn't result in a doubling of throughput.
In fact, I suspect based on the throughput doubling with FL2, we're back in the same regime as the baseline.
It would be useful to see what the latency is of FL2 on Gen12 compared to baseline (FL1 on Gen12), just to confirm.
Yes fair points. The think it’s also indicative of how important it is that code be optimized for the specific hardware it will run on. Systems need to be considered and optimized as a whole. Still an interesting post.
Visa / Mastercard / American Express all have lines of premium credit cards (Visa Infinite, World Elite Mastercard, Amex Platinum), and they're very much too big to ban. You'd just be left with one processor in the US (Discover, now owned by Capital One).