> I understand this is instead of the 2% cash back that Apple Card pays on full payments.
That would reduce the appeal a little. And that's different than their own-store policy on their card -- if you buy from Apple with an Apple Card on the installment plan, you get the 0% for 12 months or whatever, and you get the 3% cash back too.
Thanks for the 2% comment, by the way. Made me go look, since I only recall it being 1% (for most things). Turns out it's 2% whenever I use my Apple Watch to do the transaction. Works for me!
No, Average is any measure of central tendency, which include all of the median, mode, and the various means (harmonic, geometric, arithmetic,...), as well as some others.
The mode is the most broadly applicable, since (at least if you accept multiple values) it is well-defined on any, even merely categorical, data. The median requires ordinal data, and the various means tend to require at least interval-level measures (some of them require ratio-level measures.)
The arithmetic mean seems to be the most common grade school mathematics “average”, possibly because its the one that does the most to exercise basic arithmetic skills. But its rarely, when distinct from the median and mode, the most useful, and often doesn’t match the intuitive understanding of “average”.
> In descriptive statistics, the mean may be confused with the median, mode or mid-range, as any of these may be called an "average" (more formally, a measure of central tendency). The mean of a set of observations is the arithmetic average of the values...
“arithmetic mean” is well defined (“the average as we know it” is not a particular sensible phrase to use in a subthread where the point in debate is specifically whether the average refers to one specific thing.)
That doesn’t make the word salad sentence used later in the article— “the mean is the arithmetic average”(emphasis added)—make any sense. Were it “the average is the arithmetic mean” it would merely be wrong, but at least make sense (and it would be making the same incorrect point it was being cited to support), but flipping “mean” and “average” so it claims that the former is the “arithmetic” version of the latter goes beyond wrong into being a pile of words that are individually sensible but don’t mean anything, right or wrong, as assembled into a sentence.
The mean is what is often meant by the term but not always. "How to lie with statistics" is a great book and I highly recommend it to anyone still confused by such terms -- or anyone who ever sees, say, ads.
"There are three kinds of lies: Lies, damn lies and statistics" -- Mark Twain
There are plenty of things that you cannot do in an apartment. If it's a building with multiple tenants, you'll be forbidden from doing stuff that's an unnecessary disturbance to other tenants. But even apart from that, there are physical limitations. For example, many apartment buildings forbid large aquariums because the building is not designed to handle the uneven load distribution.
Look, this argument basically just boils down to "businesses can't fire customers for whom it doesn't make financial, operational, or reputational sense to serve." And like sure that sounds good except that we currently take the opposite stance by having protected classes. They are the single carve-out exception for the general idea that you can otherwise fire your customers for any reason you see fit.
All compute classes have never been treated equally though. If I open an AWS account and launch dozens of instances and DDoS other customers / locations on the internet AWS will terminate my account and shut everything down. The provider always has discretion on whatever it really wants to provide service to you or not.
Folks need to wake up to the fact that if we all end up relying on 3 cloud providers, we also end up having to live with their discretion regarding what is acceptable / moral use.
Also, this isn’t really the same as a DDOS, this is localised computation effectively. Running a hash function.
> Tbf I wouldn’t like to think they’d even know what I’m doing. Privacy should be ok.
that would be my biggest concern. how are they deciding what i'm doing, and what happens when my usage pattern triggers their magical detection?
am i booted immediately? do i get a useful number of business days to say "nope, i'm actually doing something else entirely!" and they leave it at that? or do i have to somehow "prove" what computations i'm going to run before i do so?
it's well and good to let them have whatever ToS they want, and that's certainly the direction i lean in, but enforcement of some ToS can imply concerning things.
You do get abuse notices that you have to respond to before your server gets isolated from the Internet. If you don't, you can request unblocking later from support staff.
Maybe the key would be to keep the artificial diamond ring at a similar price point. Larger, designer brand, more manual labor details, I don’t know. Not saving money, but get a superior product.
I don't think Google is quite as neutral here. Google suggests adding "cheater" to the search of the victim's names and ranks those websites.
What if the manual of the hammer included a section "try hitting someone with it".
Now, Google's recommendation clearly came from seeing those words appear together, but to me Google is somewhere between completely neutral like grep, a hammer, or a car and fully editorialized like a blog or newspaper.
You’ll end up main connections and smaller ones to reach smaller destinations. This doesn’t mean it needs a single center like Paris. Compare the map to Germany where there are roughly 3 parallel fast North-South lines not focused on a single center.
Population in France is much more concentrated around Paris than population in Germany is. The Paris metro area accounts for 20% of France's population, while you need the biggest three metro areas in Germany to get to that percentage. Also Germany has double the population density, so even the less dense parts have more people.
It's actually not just population density. A big part of it is economic and bureaucratic activity. In France everything is centred around Paris, in Germany that's very different, e.g. Berlin is not a very big economic hub (although that's changing somewhat).
> Also Germany has double the population density, so even the less dense parts have more people.
That is not causal.
Some countries have a higher population density, but the population is mostly concentrated in one area and the low density areas are really low density.
I presume there is a standard way to compare density distributions between countries.
This reminds me of Sol LeWitt. Some of his art is just a simple set of instructions that can be followed by other people to "install" by strictly following them and drawing on walls [1].
There are a lot more examples in art that many people could replicate easily, but only the artist came up with the idea.
I assume the merchant fees are the same either way.