DanAndersen

That's true -- though the figure you're linking focuses specifically on measles cases rather than overall mortality, which is what the OP article is focusing on.

(as an aside, it's really weird that the (first few) comments here on HN seem a little adversarial; I wonder if maybe some people looked at the headline only and assumed it was attempting some sort of anti-vax argument, which the article isn't doing at all.)

Very good link. I think a lot of people get intimidated by the idea of emergency preparedness because it feels like such a big, overwhelming thing to deal with. One cause of this is perhaps the Hollywood/media image of "disaster" as a singular extreme event that obliterates the average person except for a few plucky heroes. In contrast, there's a full spectrum of severity, and any small amount of preparedness can help.

In case anyone here hasn't seen it before, I'd like to share the article "Doomsday planning for less crazy folk" [0]. It's a good, sober-headed step-by-step guide to evaluating one's own threat model and making actionable steps toward improving one's emergency preparedness -- both small-scale and large-scale. I have a routine to-do item to re-read one section of this article every few weeks and think of ways I can make myself more resilient.

[0] http://lcamtuf.coredump.cx/prep/

Agreed, the GPS priors are very helpful -- the ability to specify prior known positions/orientations of photos, to some known error tolerances, is one of the power features in Agisoft Metashape Pro, which I wish was available in more photogrammetry software. Being able to know the actual nearby photos a priori and avoid the costly O(N^2) pairwise comparison is a great benefit.

Some research I did a few months ago was about trying to find ways to effectively bring in that advantage of outdoor photogrammetry into indoor environments, by piggybacking onto AR HMDs with built-in SLAM: https://www.youtube.com/watch?v=ldrGpGrOaZc

Perhaps an AI-assisted version of Ben Folds' "composing a symphony on-stage"? [0]

[0] https://www.youtube.com/watch?v=BytUY_AwTUs

What should be the exact dollar amount to avoid "looking like he's using it to drum up publicity"? You presumably have some idea of what the "right answer" is since your gut is telling you $1bn is too much.

Additionally, what I find particularly frightening and unsettling is the possibility that we'll never be able to know just how harmful microplastics are, due to the rapid loss of anything we could call a "control group" on the planet. Rapid industrialization and globalization has led to land, water, and sky all being infused with this stuff; are there populated areas anymore where we could point to and say "this place is untouched by microplastics" and compare the health of the people and ecosystem there to the affected areas?

This is true, which is why machine learning has long since learned to not even think of what you describe as a meaningful measure of accuracy. If you look at the linked paper [0], you'll find that the author uses the "ROC AUC" metric [1]:

>The ROC AUC score represents the probability that when given one randomly chosen positive instance and one randomly chosen negative instance, the classifier will correctly identify the positive instance

[0] https://arxiv.org/pdf/1902.10739.pdf [1] https://en.wikipedia.org/wiki/Receiver_operating_characteris...

A question for anyone who does collecting of stamps/coins/baseball cards, etc.: How do you deal with / feel about the trend of the creators of the collected items making things specifically to be collected?

A while back I picked up a few cheap pre-WW1 German coins -- fun to have a bit of history for not that much money. I haven't pursued coin collecting because I could imagine going down a rabbit hole of spending way too much money on it. But when I see, say, the US Mint releasing special commemorative coins specifically for collectors, my whole interest in the field dries up. It takes a hobby that, in my opinion, would be about discovery and the individual search for something and turns it into merely another consumer product to consume. Same with so many stamps nowadays, being designed and released specifically for collectors to gobble up. And then there's baseball cards and Magic cards and "colllector's editions" of video game boxes that are all designed as these mass-market products that satisfy the collecting urge in a cynical way.

If I did end up really getting into collecting, I'd think it would be more enjoyable to collect interesting rocks found during travel, or historical items like coins/stamps that were all made before the trend of marketing to collectors.

If you list job requirements, make them actual requirements and don't try to upsell the talent you get by exaggerating the requirements.

Some applicants are naturally more haggle-friendly and intuitively "know" they should apply to jobs for which they don't technically have all the listed requirements. Other applicants are more self-doubtful and will not apply, even if given their experience they would be well-suited for the job. Minimize the amount of white-lying both the applicant and the company have to do.

To any Google employees who read HN: please raise this general issue in your internal groups and try to push to change this culture of faceless algorithmic decision making. Companies should not have to rely on getting to the HN front page to have a human being actually look at these issues. You have a voice internally and if you don't use it, in some way you are complicit.

>Is this research somehow tied into the relationship between UI and imaging AR head mounted displays or are you focussing on tablet/phone AR?

It's focused on having virtual/AR content that is rendered by an AR HMD like the HoloLens, but which is rigidly anchored to a handheld object. So, for example, having extra windows or panels on a phone that float beyond the rectangle that defines the physical smartphone shape.

>This is actually why I think Magic Leap made a smart decision by making their device so 'enclosed' (as opposed to MSHL that is extremely open and unoccluded), which effectively artificially narrows the peripheral vision.

I think this was an issue that the HoloLens 1 had/has in terms of the plastic "eyeglasses" region -- because it looks like it wraps around the full eye FOV when looking at the headset before putting it on, users are disappointed when they wear it and it's revealed just how (relatively) low the FOV is. Setting proper expectations is important for users.

As someone doing AR research, I can definitely echo the importance of higher FOV for getting closer to the "holograms." I am currently doing some investigations involving tracked hand-held devices (like smartphones/tablets) with the AR HMD, and I've found that users tend to unnaturally hold the device really far out in an attempt to see any AR content that is placed relative to the tracked object. Hopefully even a modest improvement in FOV can overcome this subconscious tendency and open up a wider range of interactions.

Far be it from me to fall into the "everything I don't like is a Russian/Chinese covert action" conspiracy theorizing that is increasingly common these days -- but if I were an adversarial state actor, I would definitely look into ways of encouraging these sorts of anti-US-military "employee rebellions" in the tech industry. Geopolitics never rests, and an important asset of any nation is its tech industry, both during wartime itself and also in any long-term leadup to conflict. I wonder if the US government has internally analyzed the question of whether, in an increasingly globalized and politically agitating environment, its companies can be counted on to be supportive and/or loyal in the event of conflict.

"Build your own X" has its limits. The payment processors are the ones that will be and are the censors. Enough social-media shrieking and Mastercard and Visa willingly bend to the loudest voices. We don't yet live in a world of ubiquitous/easy crypto, and mailing cash or money orders around is not feasible at scale. Only government regulation will heal the imbalance.

You're correct that there's often a lot of interpretation and historical research that has to be done when looking through such cookbooks. However, they can offer quite an interesting glimpse into another era. Specifically, I'd recommend watching some of the Townsends "18th Century Cooking" YouTube videos [0] which show how some unusual (or familiar!) recipes from colonial America may have been made.

[0] https://www.youtube.com/user/jastownsendandson

OK. Some people react to external criticism with seething resentment. Others end up finding it a useful source of energy to push themselves into a new state even if they are hating it in the moment. The continued "tough love" approaches in traditionally masculine spaces like military boot camps suggests it works for many.

This is the danger when our lives become increasingly dependent on digital middlemen who are not regulated as utilities, where legislation has not caught up, who can proudly declare themselves "just a private company you have no say" and use nebulous ToS violations to shut people down.

When people violate actual law, we know which specific sentence in the written law was violated. We have a chance to defend ourselves. We have (supposedly) impartial judges and juries to help determine the truth.

Tech companies are building parallel institutions for judgment and execution, but building it for themselves and they have no plans to let their users have any say or transparency. It's a digital version of "rule of men" rather than "rule of law" -- "you're condemned because I, the King, say so."

This will continue until external powers (regulation) force the quasi-digital-fiefdoms to open up. Hemming and hawing about "gee, this is kinda harsh" won't have an effect on them.

>at the whims of a foreign corporation that will stop at nothing if it means 2 cents more on their bottom line, or gets them on the good side of a government entity.

Or if it gets them out of a flash-in-the-pan PR issue fomented by the latest political outrage mob, or a couple 'woke points' to score with their ideological allies.

The tech community had its chance to stand firm and demand openness and liberty of use when it comes to accessing communication services provided by mega-corps. They instead let themselves be seduced by "it's a private company they can do whatever they want go build your own platform" when it was being wielded against less sympathetic targets.

There's consistently a lot of shock and outrage about foreign interference in American elections, however there seems to be no feasible solutions to the problem. If, as people seem to be doing, "election interference" doesn't mean hacked voting machines but 'hacked voters' (exposure to subversive foreign propaganda), then there is no defense against it without going to what we would consider a repressive state. Low-cost social-media memes, groups, and bot-comments are a super-weapon that can be deployed against any government that purports to be 'democratic' in its methods.

At a minimum, to actually combat 'foreign interference in elections,' the US would need to: - rigorously require anyone paying for political ads on social media to be US citizens, - require anyone posting comments on US-visible political social media topics to also be US citizens, - limit the exposure of US citizens to foreign media, both state-sponsored and 'private' - place additional surveillance or suspicion on first/second-generation immigrants who may be spreading latent ideological memes from foreign governments

In short, a China model rather than a traditionally Western model. In a globalized society with such free flow of information, is it even possible to have a national democracy that doesn't implicitly give the rest of a world voting access through rival-state mechanisms?

The article doesn't seem to give a direct link to ElementAI's findings.

Summary: https://www.elementai.com/fr/news/2018/recherches-sur-twitte...

Detailed findings: https://decoders.amnesty.org/projects/troll-patrol/findings

The study was exclusively focused on women and a men-vs-women comparative analysis was outside their scope.

Some interesting results:

- While left-leaning female politicians received slightly more abuse than right-leaning female politicians, there is a much stronger and opposite disparity for journalists. Right-leaning female journalists receive more abuse than left-leaning female journalists -- the gap is triple that of the left/right politician gap. Does anyone know what might be the cause?

(as an aside, the mouseover for seeing the exact percentages was really annoying; the percentage for female right-leaning journalists kept getting covered up by the text interpretation).

The text summaries of the data also seem to have slightly non-synonymous interpretations:

>6.57% of all mentions of journalists working for left leaning media groups were problematic or abusive

>Journalists working for right leaning media groups were mentioned in 64% more problematic and abusive tweets than journalists working at left leaning organisations

>Left leaning politicians receive 23% more problematic and abusive mentions that their counterparts in right-leaning parties

>6.18% of all mentions of right leaning politicians were problematic or abusive

It almost sounds like making a distinction between "receiving abusive mentions" and "being mentioned in an abusive tweet."

- Regarding the breakdown of abusive mentions for different races, Latinas seem significantly lower in receiving abusive tweets than the mean (White, Asian, Mixed-race) -- with Black women significantly higher, as reported in the article. Does anyone know if there are social explanations for why that would be the case?

For now, at least.

Some companies have massive annotated databases of people's faces from different angles, and the ability to do plausible 3D reconstruction of the faces.

Does anyone know if there's anything about the unlocking method that requires the model to be 3D-printed?

It seems like you could render the 3D model of the head on a monitor and point the phone's camera at the monitor. If the phone needs to see some sort of parallax change relative to its own motion (from internal sensors), then one could put a 6-DOF tracker onto the phone and use it to update the rendered viewpoint of the head (a form of user perspective rendering).

Such an approach would be quite useful for law enforcement to gain access with much less time waiting for printing.

Agreed! My favorite Christmas carols are the ones that aren't just marking a jolly merry festival, but have that "foreboding" melancholy potency you describe. I think G. K. Chesteron did a good job of capturing the sense of it, in his "The Everlasting Man" [0]:

>Christmas for us in Christendom has become one thing, and in one sense even a simple thing. But like all the truths of that tradition, it is in another sense a very complex thing. Its unique note is the simultaneous striking of many notes; of humility, of gaiety, of gratitude, of mystical fear, but also of vigilance and of drama. [...] There is something defiant in it also; something that makes the abrupt bells at midnight sound like the great guns of a battle that has just been won. All this indescribable thing that we call the Christmas atmosphere only bangs in the air as something like a lingering fragrance or fading vapor from the exultant, explosion of that one hour in the Judean hills nearly two thousand years ago. But the savor is still unmistakable, and it is something too subtle or too solitary to be covered by our use of the word peace. By the very nature of the story the rejoicings in the cavern were rejoicings in a fortress or an outlaws den; properly understood it is not unduly flippant to say they were rejoicing in a dug-out. It is not only true that such a subterranean chamber was a hiding-place from enemies; and that the enemies were already scouring the stony plain that lay above it like a sky. It is not only that the very horse-hoofs of Herod might in that sense have passed like thunder over the sunken head of Christ. It is also that there is in that image a true idea of an outpost, of a piercing through the rock and an entrance into an enemy territory. There is in this buried divinity an idea of undermining the world; of shaking the towers and palaces from below; even as Herod the great king felt that earthquake under him and swayed with his swaying palace.

[0] http://www.worldinvisible.com/library/chesterton/everlasting...

I think the person you replied to was looking to make a "Great Leap Forward" wordplay.

Also potentially relevant to the tech community here and anyone who works on currently-non-dominant social media platforms: politicians are already demanding that Gab be held criminally liable for the speech and actions of their users.

From Howard Dean [0]:

>You are a facilitator of neo nazis and other haters. You should be tried for being an accomplice to murder. You’re lucky shutting you down is all you get.

[0] https://twitter.com/GovHowardDean/status/1056710782342311938

Related: Gab had also posted a public statement [0] on Medium after the shooting in Pittsburgh.

However, Medium has now suspended Gab from their platform, so that the only way to read the statement is to use an archive link [1].

Hopefully the tech community learns from this about the dangers of using centralized external services for your public presence. If you find yourself under distributed political attack your voice will not make its way out there without being filtered by intermediary media outlets unless you own the full stack of your system from top to bottom.

[0] https://medium.com/@getongab/gab-com-statement-on-the-tree-o...

[1] http://archive.is/vysbL

I think it's actually the reverse. Your critique is that the NPC meme ignores that lots of liberals/progressives exist.

The rightist promoters of the NPC meme would agree with that statement -- but they would say that it's because most are just acting as ideologically/memetically programmed automatons. Crucial to understanding this meme is that its promoters see "libs" not as literal bots, but as humans-with-bot-thinking, who have grown up on a diet of platitudes and mantras without deep analysis.

On the contrary, the left/libs/progressives (whatever you want to call them) on Twitter have had a habit of actually doing what you're saying, when they pull out the "oh look another Russian bot" troll. That meme actually asserts that the number of actual conservatives/right-wingers is literally small and being amplified by actual programmed chatbots.

Is it really that hard to understand? At its simplest level, the NPC meme is an extension of "I'm a member of [insert outgroup here], look how dumb I sound" trolling.

Mark Hofmann is my (ex-)uncle (by marriage, not by blood). Needless to say, the whole affair isn't something we really bring up at family gatherings. From a historical/detached standpoint the events are all certainly interesting and scandalous and fascinating, I'm sure. My main reaction to it all is just sadness at the pain he put his wife and kids through because of it, and how they've been able to show strength by moving forward in their own lives. It's times like this I feel extremely blessed that our own extended family is close-knit enough that we can support each other.

You don't need academic knowledge of anatomy if you grow up in a culture (that of medieval peasants) where hunting, cleaning, and cooking animals was likely common. Much can be inferred from analogy, in the way the non-human dissections were (and still are) used for training surgeons.