The way k-means is constructed is not based on distances.
K-means minimizes within-cluster variance. If you look at the definition of variance, it is identical to the sum of squared Euclidean distances from the centroid
Its UI is similar to Worldle (https://worldle.teuteuf.fr/), it might be implicitly assuming you're familiar with it already, in which case the "how close you are geographically" mechanism makes sense.
What you're ignoring is that a large proportion of the money printed (trillions) went towards "corporate welfare", inflating the profits of the very companies who then turned around and told their employees "sit tight we're tightening the purse strings due to the pandemic, no raises for you we can't afford it"
From the video, it appears what Disney is doing is much simpler compared to the techniques in the article. It seems children are simply "filling in" parts of existing figures (e.g. the Mickey Mouse ears shown in the video), which are then animated. This seems to me like a less challenging problem, as all the rigging/joint/etc stuff that Meta is doing can be done by-hand, beforehand, in the Disney case, it does not need any ML. The cool thing with Meta's system is that it (hypothetically) works on arbitrary human-like figures.
The US is still far more culturally homogenous than the EU, so this comparison falls kind of flat. Even if the US began as you say, with large cultural differences, it isn't really the case anymore. You could make the argument that there is a cultural divide between cities and the countryside for sure (in the sense that someone from Austin and someone from Boston are more similar than someone from Austin and someone from the Texas countryside), but this divide exists within individual EU countries as well, and is not limited to the US.
The landing page is quite noisy, what with the marquee text and the top and bottom text banners, etc. I would recommend simplifying things a bit and going with a cleaner look, right now it kinda reminds me of the style of websites in the early 2000s. Of course if that's what you were going for carry on lol. The actual editor itself is quite nice, good job!
For instantaneous time you would say ώρα, not χρόνος. "Τι ώρα είναι;" = what time is it, lit. "what is the hour". Χρόνος for me implies a span of time, not a single moment
It's time for lunch would idiomatically be something like "ήρθε η ώρα να φάμε μεσημεριανό" (it came the hour for us to eat lunch). For instantaneous time you use hour (ώρα) or στιγμή (moment). So there's actually three "temporal terms" in Greek, ώρα, χρόνος and καιρός. In my head καιρός is the "longest" of the three (describing a period rather than an instant), and ώρα and χρόνος are sort of depending on context. Ώρα is used in contexts like "δεν είναι η κατάλληλη ώρα για αυτό" (it's not the right time for that), where you could replace it with "moment" in english (στιγμή). Χρόνος is the least frequently used in the modern language I would say, and is mostly associated with abstract notions of time as a concept (e.g. "the passage of time") but also in specific terms like "I don't have time for that" (δεν έχω χρόνο για αυτό).
All of these words overlap somewhat so it gets somewhat confusing to describe to a non-native speaker, other than saying that generally a native speaker will be able to tell if the use of one of the three "feels right" in a certain context or not. Also several of these terms have additional meanings! For example, ώρα is also literally hour (είναι τρεις η ώρα = it's three o'clock, literally "3 is the hour"), χρόνος also can mean "year" (για έναν χρόνο = for one chronos = for one year), etc. etc.
Honestly I think this is the most exciting thing, and I'm very glad to see support for it from the DE itself. Currently I think as someone else mentioned it's only GNOME that does it, otherwise you have to use hacky third party programs like touchegg which essentially just emulate certain keyboard shortcuts. So don't even think about finger movement tracking
If all the companies that are willing to hire you equally treat you like crap, you aren't really given much choice. In terms of entrepreneurship, it carries a ton of risk. People living paycheck-to-paycheck typically aren't able to carry such a risk with no safety net (family, etc.).
This is a fantasy. Voluntary contracts imply that workers have the option of not working for starvation wages, which in many cases is simply not true. If your options are to work a dead-end minwage job or, well, die on the street, is it really a voluntary contract that you're entering? I would say not.
This is interesting, but the paper still notes that in most "real-life" applications people will likely still prefer gradient-boosted trees, just because you need to allocate significant computation to hyperparameter tuning even in the case of the MLP-with-regularization-cocktail. For just getting something off the ground quickly based on tabular data, GBDTs are still unbeatable.
Out of those three examples you gave, Western Europe still wins on 2/3 (higher life expectancy, higher quality of life). Granted the second is subjective somewhat, but I believe most would agree. The first is objectively lower in the US compared to Western Europe, you can check the statistics. Per capita GDP is only high in the US due to massive wealth inequality and a ridiculous concentration of billionaires. It's not "ignoring every contrary indicator", it's that most of the indicators that are used to support the idea that the US is "better" (e.g. per capita GDP) are fundamentally flawed in different ways, and don't amount to a proper comparison of the quality of life.
a) how are these regards cherry-picked?
b) it doesn't no, you are right. However the "third world country with a gucci belt" phrase is intentionally tongue-in-cheek and used to call attention to the fact that the US is behind other first-world countries in many different important ways, which it is. Personally I consider healthcare, worker rights, and infrastructure pretty key issues of extremely high importance, and I'm not sure how it's possible to see it otherwise.
I never understood what FLoC offers to users directly (rather than to advertisers) -- FLoC requires a user to opt into it, right? Why would I do this as a user, what incentive does Google give me?
Comparing all of these points against Western Europe, the US is objectively behind. There's your point of comparison. Western Europe has far more expansive worker protections (just look at "right-to-work" states, you can get fired for any reason whatsoever), far more encompassing free healthcare (not just for the poor and older individuals, Medicare and Medicaid is what you're referring to I presume right? There are no other programs for free healthcare in the US), far freer press (according to the World Press Freedom Index, the US is 44th, most Western European states are higher), infrastructure is not crumbling in Western Europe in the same way it is in the US, far lower inequality (this is obvious, check any inequality index), less corruption, on what point exactly of these is the US better than Western Europe?