> If you don’t want to kill flies, wasps, bees, or other useful pollinators while eradicating the tiny little bloodsuckers that are the drone’s target, you need to be able to not only locate bugs, but discriminate mosquitoes from the others.
> For this, he uses the micro-doppler signatures that the different wing beats of the various insects put out. Wasps have a very wide-band doppler echo – their relatively long and thin wings are moving slower at the roots than at the tips. Flies, on the other hand, have stubbier wings, and emit a tighter echo signal. The mosquito signal is even tighter.
Fascinating engineering! Doesn't seem like it would be possible but it apparently is. There's also more visuals at about 17 minutes in the video embedded in that article, the signatures seem fairly distinct.
Imagine the sound a mosquito makes when it flies near your ear; it's quite distinct. I'm sure it's possible to distinguish mosquitos based on that (which is a factor related to the doppler signature mentioned).
I wonder how distinct it is, really. It sticks out to us, but “mosquitos is enemy #1” is one of the strongest evolutionary pressure we’ve got, right? And one of the few that persists to this day.
Our brains probably have a dedicated cluster of neurons in there somewhere specifically looking for the Mosquito noise.
Related PSA that drowning doesn't look like drowning [0] — people don't yell, splash, and wave like you might expect and is "almost always a deceptively quiet event".
1 more PSA, Drowning is the #1 cause of death for people with autism. It generally happens when the person is being supervised by someone not their parents.
If you are taking care of someone with autism around water, be super aware of that.
A child with autism is 160x more likely to die from drowning vs the general population [1]
Key drowning indicators include: head low in water with mouth at water level, head tilted back, eyes glassy/empty/closed, vertical position without leg movement, hyperventilating/gasping, and attempting to swim but making no forward progress.
As supplemental reading see "bulk positive randomness" as discussed here [0]. It is tempting, but fallacious, to imagine "luck" as a zero-sum, EV=0 (or worse) process across every domain of life, where all upside or alpha has long been arbitraged away. Yes, there are some domains where you want to minimize variance, like the distribution of transit times to the airport when you have a flight to catch, or the odds of being robbed in a bad neighborhood. But there are also many situations where meeting new people, trying new things, etc. have many upsides and few drawbacks, and for most people these are (1) common and (2) recognizable.
Apple's easing function (like others) is parameterized by physical characteristics (e.g. spring force, damping) that are easier to model from first principles, but there are other parameters (overshoot distance, anticipation size, animation time) that are more useful for animation. A closed-form parametrization with the latter might be tricky to derive, but some kind of iterative solver (plug in desired animation parameters, get values for the physical parameters) shouldn't be too difficult?
I have thought of this as knowing _about_ things, as opposed to knowing things, and not having the self-awareness to be able to differentiate between the two.
I've seen this most notably in a former coworker who enjoyed watching YouTube videos (especially when the rest of the team was hard at work, but that was another point of contention entirely). He thought himself very knowledgeable on various topics because he could readily regurgitate talking points, but if you asked him about second-order effects, or implicit simplifying assumptions, or how X from the video would be different if Y and Z were different, it was obvious how surface-level his "understanding" was.
> He thought himself very knowledgeable on various topics because he could readily regurgitate talking points
I do think people have that bias - when someone is able to regurgitate talking points or answers about whatever topic in no time, said person is perceived as intelligent.
Because it takes time to actually know if someone is smart, because intelligence is complex and multifaceted – and we don’t have time anymore.
So, if within 10s you have one guy regurgitating something and the other one shrugging, it does make some sense to assume the first one is smarter.
What would make even more sense is to accept one cannot know this about someone in 10s. But uncertainty sucks, humans will always try to reduce it as much as possible, at all cost.
Not specifically about kings but — I recommend The Time Traveler's Guide to Medieval England by Ian Mortimer, for a very thorough look at daily life in this era.
More specifically about kings, ACoUP has a great writeup [0] on royal legitimacy and the various purposes of the king's court in the context of my favorite medieval shenanigans simulator, Crusader Kings III.
I generally agree with these points but I sometimes find "do X, not Y" suggestions too prescriptive to be helpful. Compelling phrasing should be context-aware and mindful of how users actually use the product, what their pain points are, and what they find useful.
Largely I take issue with the use of metrics. "Create events 10x faster" reminds me of the "put hard numbers in your resume" advice that people blindly follow — it's not a useful description to use here! When I go into XYZ Calendar and create a new event by copy-pasting fields, I don't think "hey I wish this was 10x faster" but rather "I wish I didn't have to do this". Announcing the feature as "Duplicate events instantly" cuts right to the heart of the improvement without splitting hairs of "I can do unnecessary work N times faster".
For "fixed a thread deadlock that froze the UI for up to two second when creating a new file" vs. "we sped up new file creation, so you can now create a new file in under 20ms, a 100x speedup from v1.2" — if the new file stutter is a common complaint, we can celebrate this by acknowledging the flaw and its fix. "Creating a new file is now instant after we fixed a thread deadlock bug" conveys that (1) a common user annoyance is now gone, (2) we recognized that it was an annoying bug, and (3) we put in the work to fix it. And no "100x" in there — we didn't speed up an annoyance, we got rid of it entirely.
Of course performance metrics aren't always bad — "Gleam JavaScript gets 30% faster" announcement is great! But compiled language speedups is something that users care about, and not something we want to abstract away (e.g. by merely saying "faster" or "better"). Here, putting a concrete number in the title and letting it speak for itself is absolutely the right choice.
Context is everything for making writing compelling. On that note — Style: Towards Clarity and Grace [0] is the best book I have read on writing clearly and effectively.
From the title I was expecting some hardware faults that were transmissible (as opposed to merely widespread), like the classic "hardware virus" story from The Daily WTF: https://thedailywtf.com/articles/the-hardware-virus
Yeah, the headline is using "epidemic" clickbaitly just to mean widespread, not transmissible.
The classic real example of actual transmissibility was the Zip drive click of death. A bad drive would damage disks, which would in turn damage another drive they were put in. The case was rarer than people thought but did happen. https://www.grc.com/tip/codfaq4.htm
I got an electric shock plugging in a zip drive once. They used to arc when you plugged the mains cord into the back of the drive or the power brick, I forget which.
Yes, the definition of "epidemic" literally refers to something being widespread (etymology derived from "upon the people"). It's not wrong to refer to e.g. an obesity epidemic despite obesity not being contagious.
My main gripe with fingerprint sensors on the back is that it's easy to inadvertently smudge the camera lens when unlocking the phone. Some phones have/had fingerprint unlock on the side power button which is similarly convenient, although I actually don't mind the underscreen sensors that are most common these days. I do appreciate being able to sneak a peek at my phone by discreetly unlocking it at very oblique angles that aren't possible with Face ID.
> If you don’t want to kill flies, wasps, bees, or other useful pollinators while eradicating the tiny little bloodsuckers that are the drone’s target, you need to be able to not only locate bugs, but discriminate mosquitoes from the others.
> For this, he uses the micro-doppler signatures that the different wing beats of the various insects put out. Wasps have a very wide-band doppler echo – their relatively long and thin wings are moving slower at the roots than at the tips. Flies, on the other hand, have stubbier wings, and emit a tighter echo signal. The mosquito signal is even tighter.
Fascinating engineering! Doesn't seem like it would be possible but it apparently is. There's also more visuals at about 17 minutes in the video embedded in that article, the signatures seem fairly distinct.