geeB

Nice project!

Can something like this in practice be used with HAM radio via some loophole, or it's basically impossible to use a large fraction of commonly used protocols/payloads (https, ssh, any secured protocols, VPNs, etc) due to encryption not being allowed? I suppose you could argue that the intention is not to obscure the communication, but it seems pretty flimsy.

This fits MIT's historical pattern of buckling under external pressure, like what happened to Aaron Swartz or Star Simpson. Different this time because of the cancel mob aspect, but not too surprising.

My issue was not with stopping the assault, but more with the fact that the story as told seemed quite embellished, ran on a number of outlets, and it seemed specifically tailored to garner votes.

To be specific, on one side the "bad guy" was a staten island guy wearing a cowboy hat in the midst of a hate crime epidemic soon after the capitol riots. On the other side, with pictures, a mentally ill person of color. The assaulted photographer was there shooting Yang, with many other campaign people and other photographers. All the accounts and pictures were released by the campaign, nobody pressed charges, and as far as I know nobody helped the guy get some help.

I hope that's a bit clearer. I wasn't there and don't really know what happened for a fact, but after looking at multiple sources my best guess was between cheesy embellishment and distasteful stunt.

Now I don't know what to think anymore... I thought it was widely discussed at the time, but I'm having a hard time finding the original coverage of the event on youtube, and the written articles I find seem more balanced than what I remember hearing. So maybe fake news got me on this one.

That said, the Israel comments and the CNN job had a lot more weight in changing my opinion (which to reiterate, is not "Yang is the worst", just "Yang is like his peers").

My opinion is not informed, but from my point of view the opposite is true. It's been 10 years since the first specs and silicon, according to https://riscv.org/about/history/. And it seems that one has to go way out of their way to find RISC-V systems in the wild.

I do not have any personal experience, but sadly it does not sound too surprising from what's been out publicly. The moment he dropped out and immediately got the CNN job really changed my perception of him. Then after some events from his mayoral campaign, such as his Israel comments or the Staten Island ferry cowboy incident, it's pretty clear he is a politician like many others.

It would be interesting to see the breakdown of total dollar value by type of construction. For example where I live there seem to be mostly (by total number of units) large multi-family buildings, and then a smaller number of new custom homes and additions/large remodels.

My experience goes against what the article says about hours for the custom/remodeling jobs. It seems that here GCs give an estimate for a project, but based on an estimate of hours, and the final bill will come in based on the hours. Similarly, subcontractors will give an estimate of the total, but then will have labor in their invoices to the GCs, which then get passed on to the customer.

If anything, there seems to be a tendency of being very generous to say the least with what counts as a work hour. Especially GCs dropping people off in the morning and picking them up in the afternoon and "project managing" the rest of the day, but in some cases stuff like commutes, lunch, breaks, etc seems to make it into the count.

That said, it might be specific to my area, or this type of construction is minimal in total dollar value compared to the industry, so I'm not arguing against the main point of the article. Even if we halve the real hours, construction's productivity gains are nowhere near the gains of agriculture or manufacturing, probably just enough to get back to the peak of the late 60s. And in any case, as a customer I pay for the final project so the inflated hours still count for my view of productivity.

Would be interesting if someone who knows can explain. I'm interpreting it as there is a core/trusted team working on search that uses version control and does code reviews, but these are not visible to everyone else in the company. Given the constant battle against SEO, with 140k employees it makes sense that they would try to delay such knowledge from leaking. Odd that it sounds like the code itself is visible though.

Question for people that understand mRNA vaccines: why does the third dose supposedly have the same exact formulation? The narrative in the beginning was that this new technology would allow for vaccines to be developed and manufactured in ~1 month, tailored to the variants (even claiming that cancer would be next). Why can't we have a vaccine that's more effective against the new variants?

The hulu number was quite surprising to me, given that the ad supported tier is $6/mo vs $12/mo for ad free, and if someone watches enough the extra $6/mo would seem worth it. It might have a lot to do with their $1.99 for one year promotion they do on Black Fridays though, which would change things considerably ($2 vs $144 per year), especially for households without enough disposable income. Easy enough to switch to a different person every year and keep it going for a while, and also at the time of that article it had been less than 3 years since the free tier with ads was removed. I don't think privacy has much to do with it though. People that don't have any idea about these topics just wouldn't care, and at least in my case I care but I'm under no illusion that paying for ad-free changes anything other than the convenience of not having ads. iirc hulu's ads cannot be skipped anyway, so they pretty much can collect the same exact information from both tiers.

Cool idea! Small UI feedback: it took me a while to figure out the play button and that the URL is much longer and it encodes the whole script for copying and pasting/sharing.

> I don't think anyone argues that atmospheric lead _decreases_ crime rates.

Although in principle I don't see why it couldn't be the case, as leas poisoning also has symptoms such as depression and fatigue, which could marginally sway people towards inaction.

> Mark Cuban has called for stablecoin regulation in the wake of the TITAN collapse, telling Bloomberg in an email: “There should be regulation to define what a stable coin is and what collateralization is acceptable.”

is kind of ironic given this:

> One place that these organizations are VERY DIFFERENT is that they are not based in the USA and they are not corporations. They are foundations. They are Decentralized in their governance. NO ONE owns majority control (although the founders certainly have significant influence). This is not only because of the ethos of Decentralized Autonomous Organizations (DAOS), but also because of the ABSOLUTE STUPIDITY of our regulators forcing some of the most impactful and innovative entrepreneurs of this generation to foreign countries to run their businesses.

From the original blog post at https://blogmaverick.com/2021/06/13/the-brilliance-of-yield-...

In fairness, I'll grant him that it's possible that regulating stable coins in that way is not the same kind of regulations that drive entrepreneurs to foreign countries.

GP is correct for Bitcoin. The crackdown on mining seem to applies to all cryptocurrencies though, just based on power usage, so GPU based operations will be affected as well.

I used to think like this, but I've come to realize it's highly dependent on what your dream is.

For some things it's just not possible (or at least way harder than following your dreams in the first place)... the kind of things that require being part of something bigger and having a certain reputation and seniority among your peers. E.g. flying fighter jets or do underwater welding after a career in investment banking, or trying to enter a decade later an industry that was at its infancy in the first place.

There are lots of dreams (probably the majority of them) for which I fully agree with you though.

Being just a prototype it might just have been enough to test, like the test Starships these days. Once you can cruise you can trade passengers/cargo capacity for fuel. But probably all the numbers were quite aspirational.

Also it probably would have followed a flight path similar to the Curtiss NC-4, with refueling stops, leaving the longest leg to be between the Azores and Newfoundland. Still much farther than the range though.

I think even if they could make it work though it would have been awful for a passenger. The cruise speed listed is incredibly slow. NY to London would take almost 2 days of flight time.

That would be a local hidden variable. See https://en.wikipedia.org/wiki/Bell%27s_theorem and https://en.wikipedia.org/wiki/Bell_test

I think from their perspective it was very different though. The article specifically talks about compensation to a woman's family if her husband kills her. It sounds like if a slave or a woman was killed, it was considered similar to someone's cattle being killed or a property crime, as opposed to a homicide of a free man, and thus something to be payed for. It seems that the remarkable thing about this story (aside from the emperor's involvement) was that usually a husband wouldn't be encouraged to kill himself for murdering his wife.

Whereas in the barbarian kingdoms and later on, all murders were initially a private matter and people retained the right to kill someone back (not necessarily the killer) unless they opted for the compensation. If this specific case happened after the fall of the empire, probably the father would have just killed the husband back instead of seeking government involvement. And probably the husband's family would have not thought it fair and either seek compensation or kill another member of the wife's family.

Either way, I'm glad things have changed since then!

In western history, this was definitely something (re?)introduced by various germanic people after the western roman empire collapse, and already they were encouraging monetary reparations.

In roman times, if someone killed your cousin you could go to court and it would mostly end there. After the collapse, if someone killed your cousin your family had the right to kill someone of "equal value" from the family of the killer. Aside from the brutality and senseless violence (as seen from us today), the biggest problem was that people disagreed on "equal value", resulting in a feud, i.e. an escalation of revenge because each party felt that the other side's killing was way more than warranted to make things even.

So the states (starting from the early barbarian kingdoms) would encourage the family of the killed to accept a monetary payment to avoid the murders to spiral out of control. See https://en.wikipedia.org/wiki/Weregild. I think the concepts of feud and weregild did not even exist in the roman world.

But you are right that for a long time you still had the right for revenge, until eventually it was fully phased out. A bit like the transition between the gold standard and fiat currency.

I think since the early roman kingdom days, murders within a family (parricide) were punished by the state, at least some of them (e.g. a father might have had right of death over his offspring, but maybe that ended once they reached adulthood?). https://en.wikipedia.org/wiki/Poena_cullei on the punishment for parricide at a later time has something on it.

Eventually it would be extended to homicide outside one's family (Lex Cornelia de sicariis et veneficis), but as the GP says the article confuses things a bit between the state not being involved at all, and a lack of investigation.

I think murder was just like anything else in that regard, meaning that the state would not prosecute any crimes on its own without someone going to court in the first place. But that was also the case for the jewellery example, or anything else: if nobody complained about it, the state did not have a police force looking out for these things and the crime would go unpunished.

Could you explain this a bit for those of us that either never worked for a large enough organization, or that did not pay enough attention to the politics while there?

I get that a manager wouldn't want to put down their team since that puts them down too, but something this egregious seems so extreme that would not have those consequences, not to speak of the effect on the other employees' morale of having someone allowed to not work and get paid.

I don't see how someone can threaten a manager's career without having something over them, either being part of organized crime, be related/friends with a much higher up person in the organization, or have something on the manager for blackmail.

As per the siblings, the encoding for that would be 02 xx (also figure 1 in the paper).

As for delimiters, it's not so much that to use COBS you need the delimiter, but the causality usually goes the other way, i.e. you want to send packets over an unreliable stream where every symbol is valid data, so you use 00 as the packet delimiter, and encode the actual packets with COBS (so the encoded packets won't have zeros).

So in general you won't need to "flush" partial data, because the natural way to interface to this type of channel is to send whole packets. If you don't know in what state the channel is, you can always send out a 00 before the start of packet to make sure the previous packet gets terminated.

And usually the packets themselves will have some additional validation like CRC, so the previous incomplete packet will be ignored.

I think a lot depends on what your goal is. Is it to better understand the methods section in the papers you are reading, or do you want to do some experimental work, or you have a job you are going after, or...

This is extremely broad and ambitious. Younger me would have said go for it as I loved to learn everything, but older me has forgotten much of the stuff that I so much loved to learn, so I moved to the camp of learning what you need.

Unfortunately I don't know too much about brain-computer interfaces, especially if it's cutting edge research.

At a high level, these are my recommendations:

The basic ideas about how circuits work is presented in any introductory book, the E&M book (Purcell) would mostly be useful for device physics and transmission lines plus other RF topics (mostly EMI, crosstalk, and other things that can go wrong). Some purists might argue on which side of the equation an inductor voltage should be, but it has zero practical effect. Also, this is a book for usually the second physics course in college, so you might have done that already and just need a refresher.

Similarly, unless you expect to be either developing novel devices, or be involved in fabricating existing devices in new nodes/conditions, you can skip anything about devices (types, structures, fabrication, materials, electron bands, doping concentrations, diffusion, drift, etc) and just the voltage/current behavior between pins should be plenty (these are covered in any introductory book). The chemistry book is mostly irrelevant for EE, although in the neuroscience case it's more applicable if we are talking about invasive electrodes (but still, probably too general and broad).

Books on integrated circuits depend a bit on whether you need to learn about some other topics that are not usually presented on their own, such as fast amplifiers, mixers, oscillators, etc with CMOS technology. I'd say though that RF/MW integrated circuits differs considerably from discrete RF/MW work, so again most likely you'll get away with treating various parts as opaque building blocks, connected by transmission lines. And I'm going to guess that for BCIs the frequencies involved are quite low, so this whole branch might be irrelevant.

Probably you'll need to learn the basics of data converters to digitize the brain signals, but again I'm not sure this warrants going through a course versus just the wikipedia page and a datasheet of a specific part you want to use. As with the other things above, courses are usually designed for people making converters, not people using them.

Signals, systems, feedback, control systems are very fundamental "mathy" engineering tools that apply to more than just EE, so probably a good tool to have in general.

I see your questions about wireless systems. Again as above. Usually these books are designed for people wanting to develop these things professionally, and if you just want to communicate wirelessly it's mostly learning the "API" that some chip has to do what you want. Not to mention the compliance nightmare to roll your own if it's beyond a handful of prototypes.

I think you get the theme. Sadly EE outside the companies making ICs has become very similar to software where you are basically plumbing black boxes together. And if you don't have a standard application, with lots of time spent on figuring out hacks to use existing parts in non-standard ways, because if you can't find the perfect part the barrier to rolling your own is much steeper than not in software.

So in a way, The Art of Electronics is very applicable. Unfortunately I think it's terrible to learn from unless you already know the stuff, and (unless it has been refreshed to the point of a major rewrite) the copy I have is extremely outdated that I never really recommend it to anyone, and I haven't opened it in a decade.

Unfortunately I don't know of such thing, but if anyone here knows a course from the Neuroscience side doing experimental work, you could see what the prerequisites for that are, and go from there.

But if you are not like me and can still learn a lot of new things without forgetting too much, go for it all and live the dream!

How does this work? In the case of Covid, I understand (for some value of "understand") how they can make the cells create virus proteins, which then train the immune system without an actual infection to be ready for when the real virus strikes. But if someone already has cancer cells, wouldn't all the proteins to train against already be in the body in the first place?

I think probably most astrophysicists did and realized GR does not matter that much at galactic scales. See my top level comment to this article for an explanation (disclaimer: not a professional physicist, I could very well be wrong).

I am (as many here it seems) on the side of wishing that all the phenomena currently ascribed to dark matter will be all explained away by some more elegant and less "tunable" theory. However, imo the chances of the specific idea from the article being the case are very low, and it's more likely that there is a problem with the new way of calculating the gravitomagnetic contribution, either in theory or with the code/computation.

I've had the same idea a few years ago. It's nice because the effect is that the moving matter in a certain orbit causes an increase in the centripetal acceleration (i.e. higher velocity) for orbits further out, and a reduction for orbits further in, which is in the right direction to flatten the rotation curves.

Being a procrastinating amateur and not knowing/being able to invent sophisticated techniques such as the one in the article, I didn't fully go though with the calculations, because after some unsuccessful attempts I convinced myself that the effects involved are way too small to matter.

Consider two equal electric charges a certain distance from each other. If they move perpendicularly to the line between them at constant velocity, there will be an attractive force due to the magnetic field that opposes the electric repulsion. However, the ratio of the magnetic to electric force is the familiar (v/c)^2, at least for non-relativistic speeds, and in any case the magnetic force can never exceed the electric force, because perpendicular distances are preserved in the Lorentz transformation. Hence the most the magnetic force can do is slow down the repulsion between the charges, exactly in the amount consistent with time dilation in the moving frame (where the charges are stationary so there is no magnetic field).

Transposing that to gravitomagnetism, I think that within that approximation the situation is the same. At the very best the gravtomagnetic force can equal the gravitoelectric force when masses move close to the speed of light, making it mathematically impossible to account for the DM mass being greater than the baryonic mass. On top of that, at typical velocities in galaxies the magnetic effect is ~1M times smaller.

Together with the facts that 1. there are other unexplained phenomena that won't be explained by this, and 2. it's trivial in retrospect so probably some truly smart person thought of it already and discarded it a minute later without even thinking it's worth publication, I just abandoned it, always wanting to get back to it eventually just for fun when I have the time (sadly this seems to always mean never).

Fast forward a few years, and I see this article on HN, and really regretted having been lazy with my dismissal, maybe I missed something along with everyone else, and should have really done the calculations. So I found a few hours this weekend, learned julia for the Nth time, and finally did what I should have done years ago... and as expected the gravitomagnetic effects have zero material impact on the rotation curves.

Now, admittedly my calculations are not fully correct, but I think my approximations are reasonable enough that they wouldn't be many orders of magnitude off, validating the intuition about relative strength of electric vs magnetic force. So I'd be very surprised if the results in the article turn out to be right.

I don't think it's fair to attack scientists for not having simulated galaxies with full GR. They realized with much less work than me that it does not matter. At the sizes, masses, and velocities of galaxies, the effects are minuscole, and using Newtonian gravity is sufficient, especially if the phenomenon to explain is so large. It's like in the solar system: GR can explain Mercury's precession, but it wouldn't help if we had orbital anomalies equivalent to effect of 5 solar masses scattered throughout the orbital plane.

Anyway, another thing I've put off forever is making an HN account after a long time reading. So since I saw this story here and it made me finally get that idea out of my system, what better occasion to also create an account and get that out of my system too :)