Yeah. This makes perfect sense if you consider it to be an interpreter for natural language that has a really good macro facility.
2,862 karma · joined May 14, 2021
> High levels of community masking, including both source patients and exposed people, have been associated with reductions in infections ranging from 10% to 80%, with more protection when there is consistent mask-wearing in high-risk areas such as households.
Not to mention the link in that paragraph, which references 18 studies supporting the effectiveness of masks: https://www.cdc.gov/coronavirus/2019-ncov/science/science-br...
> Many commentators have claimed that a recently-updated Cochrane Review shows that 'masks don't work', which is an inaccurate and misleading interpretation.
> It would be accurate to say that the review examined whether interventions to promote mask wearing help to slow the spread of respiratory viruses, and that the results were inconclusive. Given the limitations in the primary evidence, the review is not able to address the question of whether mask-wearing itself reduces people's risk of contracting or spreading respiratory viruses.
https://www.cochrane.org/news/statement-physical-interventio...
And that's all a good story, except that it mostly didn't happen.
Not everybody was typing in BASIC listings, then spending hours debugging where the typo was. There was more than enough premade software, including games, out there for the major computers on the market (C64, Apple II, TI-99) that you could have a lot of fun without ever seeing a BASIC prompt.
And, while the manuals were much lower level then than they would be just a few short years later, and you could certainly learn a lot about how to control the machine by PEEKing and POKEing memory locations, the fact that if you POKE some address and the screen changes color doesn't tell you anything about how it all happened. It's just as mysterious as how moving the mouse on a modern computer moves a pointer on the screen.
BASIC is too high level to teach you anything meaningful about what's going on under the hood. But, fortunately, the machine languages of the home computers of this era were generally pretty well documented. However, even then, while you'd know there's this thing called a CPU and that it has things called registers and that you can give it instructions to read and write memory locations, those things are still pretty big abstractions if you want to claim to understand "how it all works."
Not if you buy the stuff that literally says "pure acetone," e.g. https://www.amazon.com/Onyx-Professional-Acetone-Remover-Fas...
I was going to suggest that the only "adulterant" likely to be in something labeled "pure acetone" would be water from the air, but Stack Exchange suggests differently: https://chemistry.stackexchange.com/questions/54341/acetone-...
I'll say. Oak Ridge National Lab has this to say about the toxicity of acetone:
> Acetone (CAS No. 67-64-1) is a clear, colorless, highly flammable liquid with a vapor pressure of 182 mm Hg at 20C (Morgott, 1993). It is completely miscible in water and soluble in organics such as benzene and ethanol (ATSDR, 1994). Its log Kow has been estimated to be -0.24 (ATSDR, 1994). Acetone is used primarily as a solvent and chemical intermediate, and it is also found in some consumer products such as nail polish remover (Inoue, 1983; Kumai et al., 1983).
https://rais.ornl.gov/tox/profiles/acetone.html
Given a choice between the chlorinated stuff, and something one can drink half a shot glass of and not suffer any toxic effects, I'll take the acetone any day.
I generally buy my acetone online, BTW. It's relatively easy to buy small quantities of pure acetone, because it's sold as nail polish remover.
> GPT-4 performed at the 90th percentile on a simulated bar exam, the 93rd percentile on an SAT reading exam, and the 89th percentile on the SAT Math exam, OpenAI claimed.
https://www.cnbc.com/2023/03/14/openai-announces-gpt-4-says-...
So, clearly, it can do math problems, but maybe it can only do "standard" math and logic problems? That might indicate more of a memorization-based approach than a reasoning approach is what's happening here.
The followup question might be: what if we pair GPT-4 with an actual reasoning engine? What do we get then?
[0]: I'm not a materials scientist or any kind of physicist, so I'm leaning a bit on Wikipedia here. Please correct me if I'm wrong. https://en.wikipedia.org/wiki/High-temperature_superconducti...
I care about stability in my employment situation as much as the next canine. I've worked at 4 startups, each with under 150 employees apiece. I've asked questions about funding, client base, growth plans, etc. All the normal "hard" questions you have to ask as a prospective startup employee. Not once have I ever asked where they banked. Not once has anyone I know asked where their employer banked.
I would submit that if the banking system becomes fragile enough that asking about such a thing is actually a good idea, we have bigger problems as a whole, which would make a prospective employer's answer to any such question irrelevant. And if that's the case, why ask?
Another way to put it: it's not employees and consumers putting "all eggs in one basket." It's the economy as a whole. If there's one thing that the 2008 financial crisis proved, it's that if the banking system gets gummed up, second and third order effects very quickly begin to set in and ruin things for everybody.
Now, if you know how to reconstruct the banking system to avoid this, I'd like to hear it. But as long as my employer's payroll funds aren't stored under the CEO's mattress, I think I'm good with that.
I don't disagree that doing things like addressing medical debt are worthy ways to promote the general welfare, but let's not pretend it's an either/or, thing here, either. You don't seem to want to acknowledge the full context of the situation, which seems at least on the edge of disengenuousness. This as well, after you try to frame it as a "bank bailout" then backpedal when called on it.
On that point, the government does have an obligation to "provide for the common defense and the general welfare of the United States," and that is clearly one of the overarching purposes of the Constitution itself. I find it hard to argue that saving tens of thousands of jobs[0] but by making the depositors whole, when the assets of SVB, illiquid though they may be, can cover 60-90% of the cost, is at all the wrong thing to do. This is literally part of why we have a government, and why markets are regulated at all.
[0]: I couldn't find a good source on the number of jobs, but that seems like the correct order of magnitude, anyway.
Oh, man, this made me recall the "borrow, then die" strategy mega rich people use to avoid capital gains taxes:
* Step 1 is to have a large investment portfolio.
* Step 2 is to get a loan from a bank against that portfolio with a sweetheart interest rate, very long term, and interest only payments with a final balloon payment.
* Step 3 is to let the dividends and appreciation on the portfolio pay back the loan.
* Step 4 is to die, leaving the entire portfolio to your heirs. This gives them the ability to take advantage of the step up basis, so they pay no capital gains should they sell, which also gives them the ability to immediately sell a portion of the assets to pay off the loans. They can then go on to implement this same strategy themselves.
There's not a direct mapping between this and the "held-to-mortality" book, but the spirit is certainly very similar. While it's losses being hidden by banks using the "held-to-mortality" book strategy, it's gains being hidden by people using "borrow, then die." In both cases, it's essentially done by delaying realizing those losses/gains until as late as possible.
NO PIRATES, ONLY ROBOTS, WHO ARE CLEARLY NOT HUMAN AS WE ARE.
But, no comment about Jim Cramer would be complete without mentioning that there's now literally an "inverse Cramer" ETF: SJIM.