You can select on minimum guest rating, and anyone with special requests (eg an extra guest, an infant) also cannot instant book. Not age or race if that’s what you were thinking :-)
Unlike others, I have only had good experiences with Airbnb, both as a host and as a guest. The platform has improved a lot in the past couple of years -- many small quality-of-life improvements for non-professional hosts to make things easier to manage, such as automatic pricing, automatic blocking of days to avoid back-to-back bookings (if you need it), instant booking but only for guests that match your criteria, templated instruction messages to guests, etc. etc. Customer support has been decent too, though I have also heard the horror stories. It's too bad they feel the need to turn it into an everything app, I can only imagine how great the platform could be if they just decided to be the very best short rental platform. For one, they could ditch Jony Ive's reworking of the review process in favor of something that actually solves the issue with rating inflation and differences in how people rate.
Lack of a viable business model is precisely what Zach points to in the article, which doesn't preclude him from feeling a bit sad about how it was stripped for parts.
I would also expect the estimated magnitude of the effect to go down over time, but that's just my general attitude to these kinds of things, the fact is that the discontinuity design that they use already accounts for variations between classes, teachers, schools, years. The way it works is that some unexpected event that applies to some people but not others is taken to represent a natural experiment, and then variation between groups before the event is compared to variation between groups after the event. The comparison is never against no variation.
The smoking gun is really in Table 3 and Table 4, where you can see that the effects that were observed are compatible with a population effect of 0, or alternatively you can look at Figure 2 and note that you could draw a straight line (no effect) within the confidence bands. Doesn't mean the effect is not there, but that there's insufficient evidence that it is, and that we should indeed be very careful about taking the estimates at face value.
Around the time when CoffeeScript was popular, there was also https://moonscript.org/ for Lua. Made by the very same prolific Leaf who authored this DSL tutorial.
It's probably getting better but the amber-colored LEDs used to be rather inefficient. I've also heard that white lighting can slightly improve reaction times of those in traffic and leads to slightly clearer captures for security cameras. I personally think these benefits do not outweigh how extremely ugly and unwelcoming they are, but "city officials just don't care" is not what led to the adoption of white LED street lighting at all.
I wonder if actually the fact that "cargo culting" is used almost exclusively to refer to the follies of highly intelligent people in the West has made it less offensive over time. Gives it more of a flavor of "look at how stupid those (historical) cargo cults are, they're almost as ridiculous as we are."
I am really happy for you if that has been your experience, and I fully acknowledge that the fact that the extremely neurotic style of modern parenting so many people seem to practice is incompatible with pretty much anything does not imply that all kids and all kinds of parenting leave no room for anything else... still, please, recognize that there is an enormous amount of variance in what kids are like and that some things just cannot be "metabolized" that easily. I might have written something like what you wrote if we had only had our second kid, also given my experience with how wildly effective certain interventions (e.g. sleep training) can be. Knowing what our first is like, who by the way has no medical issues but just happens to be a pain in the ass, your remarks instead sound completely ridiculous.
This is mentioned in the article, but it's important to know that this is not just some Italian novelty, there has been a worldwide effort to figure out how to tax digital services provided by multinational companies in a way that makes sense and avoids funny business with transfer pricing and other tax loopholes, see https://en.wikipedia.org/wiki/Base_erosion_and_profit_shifti.... The news here is primarily that Italy is not waiting on a big multilateral agreement but has started to implement some of these ideas on its own.
In Europe it is to some extent, with the Energy Performance of Buildings Directive. Shades or awnings aren't required, but for every new building or renovation the overheating potential must be calculated and gets added to the building's "energy score", and buildings with an energy score that is too high either don't get a permit or the owners get fined.
Multiple imputation simply means you impute multiple times and run the analysis on each complete (imputed) dataset so you can incorporate the uncertainty that comes from guessing at missing values into your final confidence intervals and such. How you actually do the imputation will depend on the type of variable, the amount of missingness etc. A draw from the predictive distribution of a linear model of other variables without missing data is definitely a common method, but in a state-of-the-art multiple imputation package like mi in R you can choose from dozens.
I'm not a physicist so I can't answer that question, though personally I trust in her expertise and really loved her book Lost in Math, but many of her most recent videos and tweets are not about physics at all but instead about nuclear power, capitalism, climate change, not having children, trans athletes, AI and so forth. The lure of punditry...
Ah, I guess things are a bit different in Europe with 230V. My corded chainsaw is 2400W (10A) and my corded wood chipper is 2800W (12A). I've also heard that our extension cords are typically thinner (1.5mm2 gauge) and easier to handle for the same reason.
Corded chainsaws, hedge cutters, strimmers, wood chippers etc. are not really feasible for rural parcels or for professional gardeners, but in my opinion for a larger suburban plot they really hit the sweet spot and are often overlooked: not nearly as noisy as gas powered (work on Sundays without upsetting the neighbors!), lighter in weight, no maintenance, much more powerful than tools with batteries, you never run out of power, half the price of a cordless tool. Yeah, sure, rolling up a 50m extension cord is a pain in the ass, but in exchange you get tools that actually work.
It does look like the California law also applies to professionals (though see jakewins' comment), which probably sucks, but I'm seeing more and more 36V gardening tools pop up, so maybe the transition won't be as hard as it looks.
I'd add to those the necessity to have some distance from the material to avoid the recency effect. You can have that distance either because you'll be talking about something you haven't worked on for at least a couple of weeks, or because you've developed an ability to retain a birds eye view and adopt an outsider's perspective even when you're in the weeds. A lot of academic talks go wrong because, although the speaker deeply understands their topic, that teeny tiny detail that they were fiddling with yesterday is really not the thing they should try to communicate today, but they can't help themselves. Yet ask that same person to talk about their previous work, and you'll get a high quality impromptu introduction to the field.
"Good enough" makes it sound like barely a step up from a CSV file. I'd say its support for various encodings [1] including a great default (dictionary + run length encoding on the indices) and compression algorithms that can be set for each individual column, columnar access, partitioning, a parallelized reader out of the box, in-memory filtering and other ops concurrently with loading in the data (thanks to Arrow) etc. etc. are all really wonderful when working with medium-sized data.
Or the other way around: as a tax with a Georgist flavor. If your land has a value of $x/m^2 when free of zoning restrictions, but your neighborhood really wants to keep it free of noise / traffic / high buildings or high density, and as a consequence land value drops to $y/m^2, that is perfectly fine but it will cost you $x - $y each year.
Yeah, when I was looking into this a couple of years ago, thermal panels were about 4 times as efficient as photovoltaic panels, but they were also 4 times as expensive. The ratio has probably shifted in favor of photovoltaics since then. If you have very limited space on your roof solar thermal can still be a good idea, but otherwise why prefer low grade energy (heat) over high grade energy (electricity) if you can get the same capacity at the same price?
Going off on a tangent here, but there's an inherent tension to "make no claims without support" when support typically refers to citations. The academic literature has become so large that you find support for almost any claim whatsoever, which makes it possible to string together an entire spiel that supposedly builds on existing science but is actually completely unsupported and void of any argumentation. Whenever you read "Method A has been shown to be a great way to investigate B" with no arguments, no summary, only citations, and especially when those citations don't mention page numbers but are meant to support hyperspecific claims, be suspicious.
> A spokeswoman for the Association for Home Appliance Manufacturers says [...] data last updated in 2019 shows that the average life of an appliance has “not substantially shifted over the past two decades.”
So, unless she is lying, there is actually no story?
That's one way to put it. Another would be to say that an appliance is (or should be) designed to last for approximately the amount of time it is likely to be in use, as people often upgrade old machines that haven't failed because the new ones are more energy efficient, faster, quieter, more convenient or clean better. There is no point in designing a washing machine to last 50 years, it only adds to the cost and nobody will use it for that long.
That's a good point and a legitimate reason to get rid of the pool, but I guess I was talking more about often-voiced worries that big artists are getting richer and indie artists are getting poorer due to Spotify's algorithm -- and those worries long predate the weird ambient stuff.
Well, sure, but whether an artist gets the full amount or a pittance from their label is not relevant to the question of whether Spotify's pooled numbers are unfair to some artists. The ultimate payout is not something Spotify has a say in.
This has always seemed like fear-mongering to me, because for the pooled proportion of plays to be completely different from the mean of the per-user proportion, there would need to be an extreme discrepancy in amount of songs listened to per month and/or in the kinds of songs played by those who pay for premium vs. those who opt for ad-supported. No, if I only play the same artist for an entire month, they likely won't get $10 (-$2 in expenses), maybe it's $7 or maybe it's $12, but I don't see how you can go from there to "Beyoncé is getting all your money!?!?!" Are there any specific examples of artists that get much more or much less than you'd expect, or is this all just a thought experiment?
A linear relationship between any transformation of the outcome and any transformation of the predictor variables — so the function is linear but the relationship between predictors and outcome can take on almost any shape.
How cheaply do you think you can run a clinical trial with human patients in the absence of bureaucracy? I imagine increased insurance and recruitment costs would eat up the difference.