HN user
Because it comes in at a different frequency versus when it goes out. Light, including UV and visible light, hits the ground, then the ground gets warm and radiates in the IR, which can be blocked by clouds.
They may be referring to the meningitis vaccine: https://www.bmj.com/content/377/bmj.o997
Perovskites break down when exposed to moisture, so while I do think you could DIY a solar cell pretty easily, it wouldn't last long. There are ways around the moisture problem, but they're not so easily DIYable. "In particular, water promotes fast decomposition, leading to a drastic decrease in device performance." [0]
Sorry for the paywall, but I was reminded of this article in the Economist about the popularity of the Toyota Prius in Mongolia. One reason is low prices, but another is that they will start in extreme cold when ICE engines can't: https://www.economist.com/asia/2018/12/22/everyone-in-mongol...
"By working with the police department in Chandler, AZ, for example, Waymo has been able to train its cars to yield, pull over, or stop when it hears or sees sirens." [1]
[1] https://www.recode.net/2017/10/15/16472896/alphabet-waymo-se...
They will pull over.
I think you make a good point, but there's a big difference between the amount of waste produced and the amount of waste that makes it into the natural environment, as opposed to a recycling center or dedicated landfill. Just because developed countries are producing more waste does not mean that they are contributing more waste to the bottom of the ocean. (I am not suggesting that we should not all aim to produce less waste and handle what waste we do produce better.)
If your flight time is half, you can run twice as many flights per day with the same plane, which halves your amortized capital cost per flight. That doesn't get you all the way there, but it's a significant cost reduction, which combined with others, could conceivably get you there.
Actually, we have some very advanced materials which I know from personal experience can hold concentrated sulphuric acid for years! Here is an example of a container made from this amazing material: https://www.restauro-online.com/Sulphuric-acid-95-97-pa-Reag...
Do we know yet if CloudFlare has similar rate limits?
SendGrid has a newsletter/marketing product called "Marketing Campaigns" that's worth a look, it satisfies those three requirements.
I'm talking about the raw reads, which is important if you want to try a different alignment or base-calling method. You can debate how important it is to be able to do that, but I'm not trying to argue that the data should be kept, I was just explaining why the total size of publicly available RNA-seq data (the sum total of which the parent is attempting to organize) runs in the petabytes.
It's not generally a significant effect, unsubscribe rates tend to be well under 1%. Most people just ignore emails they don't want.
An RNA sequencing run generates on the order of 10GB of data, a typical study requires many runs (treatments, controls, replication of results, etc), and posting the raw data is required by most biology journals. I'm not surprised that there is over 1PB of data available to curate.
The author is making a distinction between whether Uber was legally at fault (as stated in the article, likely not) versus whether the accident was avoidable. I agree with the author's position that the accident was likely avoidable.
Not arguing that you shouldn't pay close attention to safety considerations, but many of those optimizations are necessary for Tesla because they have constraints that would be easy to relax in a DIY scenario: very low volume/high density, the possibility of a collision, consistent proximity to humans, etc. I don't think it's fair to suggest that you need to be individually liquid cooling every cell to ensure safety.
The woman cannot go from being "not in path" to "in path" instantaneously unless you believe in teleportation. Supposing the woman got about 1 foot into the path of the car before being struck, the car would have had a minimum of 1 ft / <woman's speed in ft/s> seconds to react, assuming total blindness until the woman was "in path" (I can't think of a real world scenario where this assumption would be strictly true). Suppose a speed of 12 ft/s (~8mph), the car would have had a minimum of 8/100ths of a second to react. Supposing 3 ft of visibility before in path (about the distance from edge of car to lane), the car would have a minimum of ~1/3 second to react. That's assuming the woman was biking along at a good clip already, which is unlikely given that she was crossing a road. So, in all likelihood, the car had over 1/3 of a second to do something. That's just above the typical reaction time of a human (~1/4 second), but I don't think it's an unreasonable expectation for an autonomous vehicle.
I wonder if it's because Uber algorithms are trained using Uber driver data (taxi drivers tend to be very aggressive drivers).
Lots of comments pointing out how obvious it is that the paper was not typeset with real LaTeX, but I think that's a little beside the point. Scott isn't arguing that it looks 100% as good as LaTeX. The article cites research that suggests that writing in LaTeX hurts productivity, even for expert users:
We show that LaTeX users were slower than Word users, wrote less text in the same amount of time, and produced more typesetting, orthographical, grammatical, and formatting errors. On most measures, expert LaTeX users performed even worse than novice Word users
You’re setting up a bit of a false dichotomy, in my opinion. Treating something like a black box doesn’t mean you have to “throw up our hands and claim to know nothing whatsoever.” Rather, it means you have to change your approach to understanding the system.
People who are concerned about “black boxes” seem to think that we need a first principles or causal-mechanistic explanation for what’s going on in a machine learning system to have any confidence in it. That couldn’t be further from the case. By interrogating the inputs and outputs of a “black box” you can learn all you need about how it works. Much (if not most) of our understanding of the physical world comes from carefully probing black box systems—systems for which we have no a priori knowledge of mechanism. So, the alternative is not to “throw up your hands,” it is to take a considered, scientific approach to understanding the relationship between inputs and outputs in your model: in what situations it succeeds, in what situations it fails, how changing a single variable affects the output, etc. Yes, that can be difficult for a complex model, but why should anyone expect it to be simple?
Anecdotally, I've found the opposite. People who are smart tend to be good at lots of things.
I'll propose a theory: very loud, low frequency sounds actually create small vibrations in your eyes, and most brains have learned to pipe this sensory input to the auditory cortex. So, the shake in the GIF triggers a "sound" through your eyes.
I know you're already getting a torrent of anecdotes right now, but I really think it's important not to sweep this under the rug. I have a co-worker who lives in a nice suburban neighborhood and had CPS called on him because his child was allowed to play in the park across the street from his house unsupervised. His child was interviewed by a social worker, his house was searched for drugs, the contents of his fridge were evaluated, and he had to go to court. In my state, it only takes one person to call, and once that person calls, an investigation is legally required, regardless of how spurious.
I had a slightly different issue (non-traditional major for the field I was trying to get into), but I found that it significantly improved my response rates to put my education section at the bottom, instead of the top, of my resume. It made it less likely that a recruiter would dismiss me out of hand based on my education before forming an opinion based on my skills and experience.
Not sure about the brain connectome project, but for the human genome project (and many other genome projects) whole genomes are available for browsing and download here: https://genome.ucsc.edu/cgi-bin/hgGateway.
If you want actionable, I highly recommend Jean-Luc Doumont's "Trees Maps and Theorems." It's like a way more practical, actionable version of Tufte's work. Probably my favorite book on communicating with data. He's a great speaker as well, do take the opportunity to see one of his talks if you can (I believe some are on YouTube).
I disagree with the analogy to speech recognition. What makes speech recognition difficult is that it's highly dependent on context (sounds the same, but means something different depending on surrounding words, what the conversation is about, or even who you're speaking with). With driving, you should be able to make a good decision with just the instantaneous state, given enough information about that state (objects, velocities, etc). You can argue about what constitutes "enough information," but it seems plausible to me that given enough sensors we could meet or exceed the amount of information taken in by a pair of human eyes on a swivel.
You might look into Secor: https://github.com/pinterest/secor
I think "particle-wave duality" is a bit of an old-fashioned idea that needs to go away. It is true in the sense that we can choose to describe the behavior of elementary particles from both a quantum view point and a Newtonian viewpoint, but that doesn't mean that things (e.g., electrons) are literally particles and waves at the same time or depending on the situation. It's waves all the way down--waves with momentum and kinetic energy and attractive and repulsive forces, just like what we think of as "particles."