I mean, sure, I get your argument, but the general state of the art these days is that turning on your console and TV and soundbar and everything is a many-second series of pauses of waiting for each device to come up and sync with the other devices, even if you have a one-button solution. A big part of the problem is that developers of devices either don't care about startup time or simply aren't competent to improve startup time.

Microsoft IS competent to get this stuff right, but even if they're perfect they're stuck with a Samsung or Vizio TV and some random soundbar.

Well, I mean, it's deeper than that. The robot doesn't understand that humans ARE. It doesn't understand what breakable is. It doesn't understand that it is interacting with humans. It doesn't even understand that it is playing chess in the way that a human does. It doesn't understand at all, it's just following patterns. So there's not even a starting place from which to teach it how to make these kinds of value judgements.

This reeks of "I didn't want to come to your stupid party anyhow!"

Recycling isn't a problem because the aggregate amounts aren't sufficient, recycling is a problem because the cost per unit doesn't pay off. It's the same with plastics, it doesn't matter if a trillion dollars of plastics are discarded every year, what matters is if it's cheaper to recycle plastics or make new plastics. The paper mentions no new techniques or technologies which will make the problem easier or cheaper, so there's still the big "Step 2: ????" problem before the profit.

A few years back, I donated to a politician's campaign. Now I get a shit-ton of emails from random politicians I cannot vote for anyhow, begging for money. As far as I can tell, the emails are written by coked-up 8th graders who just discovered that you can change fonts in text. Though given the grammar and spelling used, maybe 8th graders is giving them the benefit of the doubt. The emails often are addressed to my email address, but with someone else's name, so obviously someone is making money off of lightly permuting email lists and reselling them.

The combination of poor list management and atrocious writing ability literally makes me embarrassed to be associated with the party in question.

So let the streamers pay to get winning matchups, and give a cut of the revenue to the people they are crushing. That way streamers don't even have to be very good at the game to win!

Or, better yet, just pair them with bots.

[Is it too obvious to pair them with other streamers?]

I'm sure someone had a pretty stable revenue stream off of this. It just wasn't any of us.

There are already plenty of adequate CPUs available for Chromebooks. Which are then paired with 4GB of memory to make sure they don't get too useful.

The fine developers you reference are listed on the NTFS Wikipedia entry, having developed a filesystem driver that runs in Windows NT.

It's a different driver from the NTFS3g drivers which run under Linux, and give access to the filesystems written by NTFS under Windows.

I mean, sure, but I also don't really have any need for that $5 latte I buy every morning. Life is too short to quibble about whether you should put 10 hours into a project on a $35 device versus putting 12 hours into a version that will run on a $3.50 device. I just think it's neat that so many people are interested in buying these things for themselves or their kids or whatever that I can piggyback on their volume to get cheap toys for myself.

I mean, don't get me wrong, I have projects I've spun up on an ESP8266 specifically BECAUSE it wasn't obviously reasonable to do so, and at least one had me wondering if I should acquire some ATtiny85 boards, just to see (I moved on). But I see the code people post on their instructables and similar, and, well, many people aren't even hardcore enough to even properly understand what they are doing, much less really squeeze much out of their devices. If you're just stringing together gleanings from Stack Overflow and YouTube comments, then, well, by all means pick your device based on price point and/or which device your favorite YouTube channel seems to like. If you eventually figure out that it was the wrong device, that's GREAT, you've learned something! Keep at it!

The problem with the weeder courses isn't that they shouldn't be teaching Calculus, it's that students aren't prepared. If you've spent the past six years being taught that math is too hard to understand, so just memorize some stuff to get through it, you aren't going to have a good time in Calculus.

It depends on whether the values of those constants are measured as a consequence of the size of the proton, or whether the size of the proton is being measured as a consequence of the value of those constants. If the measurement itself involved application of those constants, then the changes are already baked in.

If you pump your NFT from $1 to $100,000 by trading amongst your sock puppets, then you'll owe capital-gains taxes on the gains between where you first bought in and where you last sold out (and bought in, obviously). Then, since the value provided is more artificial than normal, you risk having the NFT drop back to $1. If you happen to figure out in January of 2023 that you owe taxes on $99,999 in gains, you can sell your NFT for $1, and ... you're in some trouble.

Of course, having sold for $1, now you have a $99,999 loss which you can carry forward! Assuming you don't generate some capital gains to net it against, you can deduct up to $3k of income every year until you use it up!

This basic scenario happened to people in the dotcom bubble who early-exercised their stock before IPO, and then had the price collapse under them before they could sell because of the post-IPO lockup. Hopefully people realize this before the end of the year, because if they round-trip from $1 back to $1 within the same tax year, in all nets out fine.

Agreed. The key point is that there needs to be back-pressure in the system to force adaptation. For instance, recently many products have made their packaging more composite (like a plastic container with a different kind of plastic shrink-wrap around it, which you're supposed to remove before recycling). In a correctly-working system, they'd reduce net costs by designing the containers to be MORE recyclable, not less recyclable. They'd work to find a way to use FEWER different kinds of materials, rather than more. Just because it's cheaper to use a particular plastic in a particular use doesn't mean it's inherently expensive to use a different kind of plastic.

To be fair, this question comes up because we've answered incorrectly to "Do teenagers need to know CS?"

We have an unfortunate tendency to imagine that we can force-feed high-level knowledge to younger kids and accelerate success. It upsets me that they have "AP Calculus" and the like in high schools. If you want to go to college in a math-heavy major, take Calculus at your college, where the instructors relied on to teach the Calculus classes will have history with the instructors relying on the Calculus prereqs. Where you'll be in Calculus class with other students who will be in your classes with a Calculus prereq. Where they have scheduled the classes so that you feed more directly from the lower-level classes to the higher-level classes, so you can apply your knowledge while it's still in your brain.

Or, you can take an AP Calculus class, skip right through freshman Calculus, and then find yourself retaking Calculus as a sophomore because you find out you don't know your shit. YMMV, I guess.