It's funny how these comparisons all depend on how you measure things. Yes, China has lower emissions per capita, but if you factor in economic output, the scales tip again. China has higher kg per PPP $ of GDP. On that measure, China's CO2 emissions are less efficient now than the US's in 1990. And the US's total emissions peaked around 2005

And even if you look at emissions per capita you can see a clear trend of the US getting vastly more efficient over the last few decades. while China's emissions per capita are increasing rapidly. Given China's extreme demographics, I wouldn't be surprised if their emissions per capita exceeded the US's.

I'm not defending whatever GP was on about. We're all totally fucked if we don't get emissions down (and probably even if we do...). But I think there are some useful ways to compare economies and measure progress on these issues.

The American right has absolutely blown a circuit in its reasoning about this. They can't talk about out-competing China in this area because they'd have to admit the left had a few good points about clean energy. The argument appears to be:

1. 1. Why bother building things more efficiently if China's not going to
2. 2. We need to be competitive with China on trade!
3. 3. EVs suck!
4. 4. clean energy sucks, don't install it!
5. 5. Drill baby, drill
6. 6. Why are China's solar panels so cheap? We need tariffs!

If only being able to prove it in court were a sufficient defense. You may be bankrupt from legal fees before it gets that far...

Maybe. I have heard that many gas stations break even on gas sales, and make their profits from their convenience stores. If so, they might just find other ways to draw in customers: EV chargers / battery swaps, Vision Pro headset cleanings, slot machines, or even more addictive snacks.

FWIW, most of the gamers I know are like me: we do most of our serious work from a Mac, and we have a separate PC for gaming. We don't trust the Windows systems with anything sensitive.

I would absolutely love to replace my four-way KVM switch setup (work mac, personal mac, personal PC) with a two-way.

I don't have much hope for ever being able to play all the games I want to play on Apple hardware, but it would be nice.

...Most of those productivity benefits are going straight to executive compensation and bonuses, with a piece going to the shareholders (but only through speculative increases in stock price -- not dividends)...

This definitely is not my area of expertise, but I wonder whether increased interest rates will put pressure on companies to pay out more in dividends. A lot of dividend stocks (esp. utilities) have been getting crushed since their yields are < the 5-6% bond yields.

And since this is a story about AI, I asked its opinion:

Yes, as a hypothetical senior stock market analyst with a background in economics and finance, I would say that an increase in the federal funds rate can indeed put pressure on companies in various ways, affecting their capital allocation decisions such as share buybacks and dividend payments. Here's a brief explanation:

1. Cost of Debt: As the federal funds rate increases, the general cost of borrowing also tends to rise. Companies that rely heavily on debt for financing will find it more expensive to borrow. If a firm has a significant portion of its capital structure in variable-rate debt, the interest payments on that debt will increase, potentially reducing available profits.
2. Opportunity Cost: Higher interest rates mean that the risk-free rate (like the rate on a government bond) is more attractive relative to stocks. For income-seeking investors, if bonds start offering better yields, they might shift some of their allocations from dividend-paying stocks to bonds. This can put pressure on dividend-paying companies to either maintain or increase their dividends to retain these investors.
3. Capital Allocation Decisions: Companies often choose between reinvesting in the business, buying back shares, or paying dividends. If the cost of borrowing increases, share buybacks funded by debt might become less attractive. Similarly, if companies perceive that their stock is undervalued and they have sufficient cash reserves, they might prioritize buybacks over dividends. On the other hand, if they want to keep or attract income-seeking investors, they might prioritize dividends.
4. Economic Slowdown Concerns: If the Federal Reserve is increasing the federal funds rate to combat inflation or preemptively cool down an overheated economy, it might be seen as a signal of potential economic slowdown or tougher times ahead. In such scenarios, companies might become more conservative with their capital allocation, potentially prioritizing dividends to reward shareholders or holding onto cash for future uncertainties.
5. Market Perception and Shareholder Expectations: The reaction of the broader stock market and individual shareholders can influence a company's decision. If the market values dividends more in a higher interest rate environment, companies might adjust their policies to meet these expectations.

In summary, an increase in the federal funds rate can indeed influence companies' decisions about share buybacks and dividends. However, it's essential to remember that numerous other factors, both internal (like a company's specific financial situation and strategy) and external (like overall market conditions and industry trends), also play a critical role in these decisions.

and I asked ChatGPT (Sept 24) if it agreed with my response to you:

Your observation about the potential impact of rising interest rates on dividend-paying stocks is astute. As interest rates rise, especially if they rise rapidly, the yields on safer assets such as government and corporate bonds become more attractive. Consequently, income-seeking investors might shift some of their capital from dividend-paying stocks to bonds, especially if the yield on these bonds surpasses the dividends from stocks.

I absolutely get over $100/mo of value from ChatGPT. At least my employer does. Just in the past 12 hours it saved me at least an hour of time.

I was dealing with a [poorly written] script someone had written a few years ago. It was causing issues in something I was working on, so I needed to debug it and add+update some documentation along the way. After fixing some of the obvious issues, I realized I was wasting my own time. I had ChatGPT generate documentation for each function and the whole script, explicitly outline areas where the script would fail in odd ways, and write up a list of potential bugs/problems. Then I vetted those, made a few more changes, etc. That took about 10 minutes end to end. And it did a much better job than the average software engineer would do if I had given them such half-assed input.

That said, the hype is overblown and promoters aren't worth listening to.

You misspelled "web 3".

Seriously, though, I am one of millions of paying customers for several [narrow] AI products. They have dramatically increased my productivity. If you haven't already, I'd like to encourage you to consider that it's being hyped and invested in because millions of people are finding it to be a really useful, innovative tool.

To counteract my own point, here's a completely useless (but highly amusing to me) use case for ChatGPT. I asked it to write a sonnet about its usefulness, citing some of the things I use it for:

100% the same experience here. ChatGPT is also an extremely useful rubber ducky. It can't do my job for me, but it's definitely made me much more efficient. It's more useful to me right now than a mediocre junior engineer/intern.

LLMs remind me of the early days of configuration management. Some sysadmins were slow to adapt, especially since early tools like Cfengine 2 were clunky at best. But the productivity of admins who learned the new tools skyrocketed. Configuration management tools didn't replace systems administrators--Ops people are still in high demand. But due to the productivity gap, admins who won't or can't use configuration management tools are far less employable.

I agree on principle, if only so we can explicitly shift all spending from "tax cut" to "spending" and have slightly less stupid debates about fiscal policy. Especially on things like the home/property owner and child-raising tax incentives.

But I do want to raise the issue of localized cost of living. $14,580/yr goes a lot further in Mobile, AL (~$25k/yr median income) than San Francisco (~$55k/yr), Berkeley, New York, etc.

I've always found it a little bothersome that $100k/yr is solidly middle class in some areas, fairly well off in others, yet the two have the same federal tax rates.

But then the stock price would go down!

Funny, though, Intel did announce a 5.2% dividend yield for March right after this story broke. And their share price has been up since.

I was going to ask how you knew that, and if you knew what her exercise price was... then I realized I could just find the disclosure online: https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.sec.gov%2FArchives%2Fe...

It looks like she just (last week?) exercised options for 187,607 shares at $2/share when the stock price was ~$28.30/share. That's $4,934,064.10 net, right? But hey, this crap's complicated. A lot of those shares were from a 2015 grant when the company was probably worth a bit less.

Do we know what kind of stock she got (and tied to what strings) in 2022 and 2021? The most recent I found (in a quick search) was $7.43/share for her 2018 options. Salary.com shows her getting stock award of ~$12m in 2022 and $5.6m in 2021, compared to options in 2020 (and presumably earlier). If she got straight RSUs, those are still worth something. A whole lot of something.

You're not really wrong--this is total comp, not salary. But it's fair to call RSUs at a publicly traded company a hell of a lot closer to salary than ISOs at a private one.

That level of incompetence managing a pension fund should also be criminal.

Oh really? How do you define "live better"?

I really dislike this common refrain. It usually centers around the argument that we live in a time where modern necessities would have been marvels in the past (refrigerators, TVs, medical advances). As if Rockefeller would trade his vast empire for a refrigerator and three jobs!

It ignores the facts that less privileged modern humans have less financial and social security, less free time, suffer more from stress and food insecurity, eat less healthy foods, and often have to choose between work and spending significantly quality time with their own children.

By my metrics, American slaves did not live better than Roman emperors. The modern middle class does not live better than the wealthiest from a century ago. The privileged live better because they can meet higher order needs, pursue their own goals, and pay others to do all the labor that would otherwise occupy their time (cooking, cleaning, growing food, even making money by managing their capital). Have you seen what a family office does?

I have to agree with the rest of your defense of California, but what parts of SF are you going that you're not seeing human feces, or is it just the act itself you haven't seen?

I have seen human feces every time I have been to the city in the last few years, especially in BART, SoMa, and within a mile of the Tenderloin.

The rest of your post is spot-on. And it's worth asking how many of the homeless folks in California have been bused over from other states vs. how many are CA native.

FWIW, I'm reading this article on Arc. I've been using it for... six months? It's a real product and I love a lot of things about it. Easels and notes built into the browser are great. Split view is clutch. "Spaces" instead of new windows + tab groups is a much better organizational system. Allowing use of different (or any combination of) profiles in every space is really awesome. Their UI also makes bookmarks usable for me again.

From the UI and product side, it's great. A really big improvement over Chrome for me. It's still just Chromium rendering, plugins, etc. under the hood. But the overall UI is such a big improvement.

On the tech choices side, I have serious questions. I'd love to talk to their CTO about why they chose to write it in Swift (that's why it's Mac only right now).

Instead of porting it and maintaining it in two languages, or re-writing in Rust, they're taking on the much harder job of trying to get Swift running on Windows. No idea what they want to do about Linux.

My best guess is that they started with Mac because that's what they use, it's what many early adopters use, and their team knew Swift.

From my experience with the browser this year, I think they have a great product. I was an early adopter of Netscape, Firefox, and Chrome. I remember my feeling with each jump, that "this is so much better, this is what I want to use going forward" feeling. I feel the same way about Arc. I don't know if we'll all be using Arc in three years, but Chrome and Firefox should be eyeing Arc's user experience if they want to remain relevant.