If some US drone manufacturer crosses the administration in some way, say in terms of backdoors or lack thereof, it's one less option for the consumer.

Think chat control in the EU but based on executive order in the US, and drones.

Not like a full-on conspiracy or anything but I wouldn't say it's beyond the pale for that kind of conversation control to be SOP for some groups somewhere.

Also, we already know that many foreign countries have intentionally hired people to do specifically that.

If the US does actually ban DJI drones the price of them will skyrocket (there's no good competition at the price point), DJI will do their little rebrand shuffle they did last time, and we'll have wasted a few news cycles (by design).

It's incredible that people will say this with a straight face, then export Instagram to the rest of the world and proceed to cry "Free Speech" when Meta bans come on the table.

It's incredibly hard for me to square if the concern is "ByteDance is the bigger threat to the US" or if the concern is "ByteDance is the bigger threat to my stock portfolio"

This sounds like a straw man.

I worked on the TikTok ban bill. I'm also a huge sceptic of our own social media companies.

"When researchers gave people willow bark extract corresponding to 240 mg of salicin, then looked at how much salicylic acid was present in their blood over time, it was the equivalent of taking 87 mg of aspirin (300 mg to 600 mg is recommended per dose, with up to 3600 mg allowed per day). Notably, 240 mg of salicin is the recommended daily dose specified by the European Scientific Cooperative on Phytotherapy...

If... each cup of tea provided 240 mg salicin (possible with a good steeping and a high salicin content in the bark), then one would need to drink 41 cups of tea to get a full, therapeutic aspirin dose of 3600 mg."

Wouldn't you only need around 4 cups to get a full dose? That seems not unreasonable to me. The 10L would be to get the maximum safe dose, which seems like a different thing.

It's relevant because it's a primary argument the author uses to dismiss willow use in older times (even as they point to similar use later as eventually motivating the discovery of aspirin even later).

And 240mg is right under the lower end of the recommended dose.

So, two cups?

Or more likely, “drink this until you start to feel better”.

This depends entirely on how bitter it is. There are certainly root bark teas you can brew that will induce vomiting before completing 4 cups.

Someone chewing on or otherwise consuming willow bark extracts isn't taking aspirin.

There's a lot of disputes between state and federal authorities about this. The 9B figure is being disputed, and there's a lot of claims that the state has been kept in the dark about it intentionally, with claims that MN GOP House members on fraud investigation committees have been purposely withholding information from the state, forwarding their information instead to federal prosecutors. These House members don't seem to deny this:

https://kstp.com/kstp-news/top-news/house-fraud-committee-wi...

https://www.startribune.com/walz-says-theres-no-evidence-of-...

I'm not sure people are really disputing that the fraud happened. I think there are disagreements about the scope of it and about the way in which the investigation has been happening, and the motives for the way in which it has happened.

Here's the whistleblower portal they set up earlier this year for example:

https://mnhouserepublicans.com/whistleblower-portal/

They then forwarded tips to the federal DOJ and not the state.

So are they picking on MN? I don't think the fraud, to the extent it is there, is them picking on them. The way in which this has unfolded might be.

In any event, thank you for the response and links.

I think the idea of Amazon buying Roomba causing monopoly concerns is kinda weird. But I also think it's absurd for anyone to blame the FTC for this. Amazon had enough leverage and money to take on the FTC over this, and probably could have made the FTC look laughable in the process, if they pushed antitrust. The fact that they didn't says everything about the actual value of Roomba.

"Area for future improvement: developers continue to improve the ensemble’s ability to create a range of forecast outcomes."

Someone else noted the models are fairly simple.

My question is "what happens if you scale up to attain the same levels of accuracy throughout? Will it still be as efficient?"

My reading is that these models work well in other regions but I reserve a certain skepticism because I think it's healthy in science, and also because I think those ultimately in charge have yet to prove reliable judges of anything scientific.

I've done some work in this area, and the answer is probably 'more efficient, but not quite as spectacularly efficient.'

In a crude, back-of-the-envelope sense, AI-NWP models run about three orders of magnitude faster than notionally equivalent physics based NWP models. Those three orders of magnitude divide approximately evenly between three factors:

1. AI-NWP models produce much sparser outputs compared to physics-based models. That means fewer variables and levels, but also coarser timesteps. If a model needs to run 10x as often to produce an output every 30m rather than every 6h, that's an order of magnitude right there.

2. AI-NWP models are "GPU native," while physics-based models emphatically aren't. Hypothetically running physics-based models on GPUs would gain most of an order of magnitude back.

3. AI-NWP models have fantastic levels of numerical intensity compared to physics-based NWP models since the former are "matrix-matrix multiplications all the way down." Traditional NWP models perform relatively little work per grid point in comparison, which puts them on the wrong (badly memory-bandwidth limited) side of the roofline plots.

I'd expect a full-throated AI-NWP model to give up most of the gains from #1 (to have dense outputs), and dedicated work on physics-based NWP might close the gap on #2. However, that last point seems much more durable to me.

Looking at the last 10 years gives a different perspective (not great for Firefox but maybe underscores something is different recently in general):

https://gs.statcounter.com/browser-market-share/desktop/worl...

A few: Developer tools are quite slow; Airline websites often break during checkout; JS games and video players sometimes stutter or use a lot of CPU

And when people ask them what they meant, they revealed that they used some package from the arch user repository that apparently required manual compiling for every update.

And instead of thinking that this wasn't the unusual behavior of a particular package, they insisted that this was the normal Linux packaging experience, which was why Linux as a whole was a terrible operating system.

A bunch of commenters chimed in emphasizing that the whole package distribution system in Linux is designed to among other things, handle dependencies and avoid manual compiling (though it's available as an option), and they were all dismissed as just being fanboy apologists.

And I'm able to access my bank, my credit cards, my utility bills, in Firefox without issue. So I'm not sure what people are talking about.

One thing I am familiar with though in the aftermath of gamergate was a bunch of motivated reasoning to complain about games and insist that they had design flaws or bugs, when really? The bugs weren't real but were kind of just a different way of saying We Don't Like This Game. And so reports of perceived bugs in some cases are as much a social phenomenon as they are a sincere representation of software functionality.

I don't want to say there's no bugs but for every one person's unsubstantiated anecdote, I seem to be able to find two people able to reproduce a functional version of the experience without issue. And just to zoom in on the bank login issue in particular, I use a credit union with an old decrepit HTTP site that was recently updated to a slightly less old and decrepit HTTP site. Plaid is unable to successfully log in, but the web interface works perfectly fine on Firefox mobile.

It happens so rarely, I don’t keep Chrome installed and have to download a new version of Ungoogled Chromium when I need to see if something only works in Chrome, which I can only remember doing about twice in the last year!

What you run into the most is the website saying, hey, it looks like you are not using a browser we have tested against, so we are not going to let you log in. Please come back when you have Chrome, edge, or Safari.

Generally I use a beeswax and mineral oil finish, sometimes this other product I can't remember the name of made from flax oil.

I've been wondering why jojoba oil doesn't get mentioned more in these discussions, either in combination with something else or on its own? It's a wax but liquid at room temperature, and seems to be stable for a long long time, long enough at least that it would probably need some refinishing before it might go bad.

But if you put wood treated with non-polymerized oil in a hot soup or if you pour hot tea into a cup finished with jojoba oil, the oil will get out of the fibers and into your hot liquid, the fiber will raise and the wood will start to feel rough after a few uses and start to get stained from your food and beverage.

I understand why you'd use a polymerizing oil for certain things; I guess I wonder about it as a substitute for mineral oil.