stefco_

any reason for this vs. plumbum?

Right. My point is that, as with a battery, you'd need to store the product so you can reverse the reaction. Practically none of our current infrastructure for using CH4 is set up to store the output CO2 + H20; it just dumps it into the atmosphere.

If you want to reuse existing CH4 distribution pipelines, that's fine, but now you need to pipe back the CO2 to the plant that makes the CH4 in order to complete the cycle. That's much harder than pumping CH4 and would require a whole new parallel pipeline infrastructure that can transport highly pressurized CO2 back to the source plant. Alternatively, you could regenerate the CH4 on location, but then you need to pipe the energy in; in that case, you're not using CH4 to deliver energy at all.

You could make the CH4 from CO2 in coal or something, but then you're no longer carbon neutral (since that CO2 will end up in the atmosphere rather than the battery). You could also try carbon capture, but honestly thermodynamics suggests that that will never be a viable solution; far easier to order your coffee black than to unmix the cream from it.

My point is that, unless I'm misunderstanding something, existing distribution systems for natural gas are not useful for CH4-based battery storage.

This definitely applied to me. I picked up Chrome partly because I trusted Google, and I moved away from it mostly because I no longer do.

You'd still need to capture/sequester the CO2, which to me sounds much harder (or else you're only solving the "renewable" part rather than the global warming part). Water is nice because it naturally precipitates out of the atmosphere and returns to oceans/lakes/etc.

I think gp's point is that nobody does this in an organized way that's not tied to a platform. When Facebook and Google eventually die, their archives will die too. You'd ideally have some format with the openness and durability of an ASCII file that could be stored in a single location and survive multiple storage medium migrations trivially; something like a photograph in a box, which anyone can interact with easily regardless of whether the camera maker is still in business. And while a GitHub or Dropbox sounds great, the issue is that most people don't have the technical ability to collate their data from all the places its stored, and their descendants also mostly won't have the ability to pore through it unless it's pithy and organized--and if that's the case, then that it just means that the deceased came up with their own ad hoc durable "life archive" data format.

Proved his own point

I think that if you read it symbolically it doesn't really matter whether the psychological model is inaccurate. This is why literary critics still use Freud in discussions. It's not about whether the model is accurate; the point is that these symbols provide a set of archetypes that we can then relate other things to. Freudian stuff in art criticism (when used well) is more of a shared vocabulary for discussing things than a reductive framework.

Not sure if that's your gripe. I used to dislike Freudian language/themes for a similar reason but realize now that I was taking it more literally than I now do.

p=0.05 is an absurdly low bar even with an extremely well-defined hypothesis. Once you open things to p-value hacking (i.e. trying a bunch of hypotheses until one seems significant), p=0.05 is almost guaranteed for one of your hypotheses and is hence meaningless.

The physics community is pretty good about dealing with both of these shortcomings. You often need a lower p-value of around p=0.003 (3 sigma) or lower to say something was significant, and a few more sigmas to claim an actual discovery. And in addition to this, you're expected to include a trials factor to correct for multiple hypotheses. It's not perfect, but it makes claims of statistical significance more meaningful.

She may not be a crackpot within her field of expertise, but she is writing confidently without doing background research. I stand by my characterization of this piece as being of crackpot quality, and while I trust that she is perfectly capable in her own field, I think her willingness to write such a slapdash post—a post that will be read and trusted by very many given her large platform—indicates a pretty cavalier approach to physics writing in general. As someone with a reasonably large platform, I think she has a responsibility to do more research before writing a piece like this; I can immediately tell that she didn’t talk to any one of the hundreds of people who could have addressed many of her points. Instead, she cites hearsay that has been addressed in detail. I don’t consider that to be good faith writing; it’s lazy and she doesn’t deserve praise for it, regardless of the quality of her other work.

I'm a LIGO Ph.D. student and this is a crackpot hit piece. I won't go through everything, but let me address this point:

> But, as critics have pointed out correctly, the LIGO alert for this event came 40 minutes after NASA’s gamma-ray alert.

I was there the day of, when this alert went out, and the reason for that delay was that one of our detectors had its data vetoed due to a loud glitch minutes earlier and our European counterparts, Virgo, had problems getting us their data in a timely manner. As a result of that, and the fact that we had human-in-the-loop checks for everything that went public, our first announcement was slow and poorly localized, with improved data products being released throughout the day. All of this is well-described in our publications [0] on the event and can be corroborated by any one of hundreds of scientists who know what they are talking about.

[0] https://www.ligo.caltech.edu/page/detection-companion-papers

Likewise. Sticking to generic DigitalOcean/AWS/Azure services for cloud hosting.

Others have pointed out many weaknesses of this product already, but one seemingly annoying issue is homophones in the context of verbally communicating your location. They use a 40k word dictionary; surely there are tons of homophones in there. And there are surely more homophones outside the dictionary (which would just give an error when queried through their API).

When you call some delivery service and ask them to find you at course.serial.flower, they could end up querying one of:

  - coarse.serial.flower
  - course.cereal.flower
  - course.serial.flour
  - coarse.cereal.flower
  - coarse.serial.flour
  - course.cereal.flour
  - coarse.cereal.flour

If these homophones are used in the dictionary, the other person will get an effectively random location on earth. If the homophones are not in the dictionary, they will get no result from the API.

In practice, you'll either need to be extra careful about specifying locations this way, or you'll need to provide extra location data like zip/city/etc. as error-correcting code. But at that point, it seems like you're not much better off than you would be with conventional addresses or LAT/LON coordinates for common use cases.

> Why are people assuming racism?

Because a huge number of black people in America describe a frequency and quality of interactions with police that are vastly different from white people's described experiences, claims which are backed up by horrifying anecdotes (like this story, or the seemingly weekly story of some other small town police department that was "shocked" to find it had officers in the local KKK chapter) and larger scale statistics on law enforcement.

That is a strong prior. Coupled with the utter stupidity of testing bird shit on the outside of a car to see if it's cocaine, it seems entirely fair to conclude this was motivated by some amount of racism.

That explanation only makes sense if the police are already targeting black or hispanic neighborhoods at higher rates, which they do—but it's often blatantly for reasons other than the residents' wellbeing.

The Department of Justice's official report [0] on the Ferguson police department after their killing of Michael Brown concluded unambiguously (see page 9) that the Ferguson PD aggressively fined poor black neighborhoods to make up budget shortfalls, and they coordinated with the city council on this. This is no lefty, prison-abolitionist organization; it's the US DOJ concluding unambiguously that this police department overpoliced its citizens to extract money from them. The racial and economic bias makes sense in this context: black communities know they are overpoliced and have been talking about it for ages, but nobody listens to them, so the Ferguson police knew they could get away with the exact same extortion tactics used by the mob to rob the poor and downtrodden.

There are many other conflicts of interest and lacks of oversight that incentivize police departments to literally rob their citizens; civil asset forfeiture [1] allows police to confiscate, hold, and liquidate assets that they believe are related to crimes, often without trial or other due process; this mechanism is also frequently abused.

Police departments also often have quotas (though they outwardly deny it) on how many arrests/stops they need to make or tickets they need to issue; again, this incentivizes cops to go to neighborhoods that are marginalized and abuse their inhabitants (as happens in NYC, where stop-and-frisk quotas mostly impact black and hispanic citizens [2], as revealed by actual NYPD officers and corroborated by the huge number of non-rich, non-white men who are predominantly targetted by this policy).

And all of this forms a feedback cycle: cops abuse the marginalized, building criminal records for the disenfranchised citizens in specific neighborhoods; those neighborhoods become "high crime" because our metrics of crime are based on police activity, which is deliberately higher in those places; and the police then justify continuing their interference in those communities by claiming that they are "high crime" and therefore in need of extra policing.

[0] https://www.justice.gov/sites/default/files/opa/press-releas...

[1] https://www.aclu.org/issues/criminal-law-reform/reforming-po...

[2] https://www.npr.org/2013/03/21/174941454/at-stop-and-frisk-t...

I know what you mean, and I agree that the term's construction can easily lend itself to unproductive and divisive rhetoric, though I think it's a bit more complicated.

I do think "toxic masculinity" is very different from a phrase like "violent blacks" because it specifies a behavioral trait/set of cultural mores (masculinity) over which people have a significant degree of control versus an intrinsic trait (like maleness or blackness) over which people have no control whatsoever. There's also not much of a history of men being violently repressed or disenfranchised purely for being men, so the potential antagonistic aspects of a phrase like "toxic masculinity" are not instantly amplified by association/context in the same way as your other examples. Also, "toxic" is not tied to a specific harmful stereotype; this leaves one open to dissect which aspects of stereotypical masculinity are toxic, rather than concluding a priori that men are "violent" or "neurotic" like in the other two examples. I think "toxic feminity" could similarly, in good faith, be used constructively in ways that "neurotic femininity" cannot.

That said, the fact that a good number of people use "toxic masculinity" in a vague and unproductively confrontational way has given it a bad connotation to many; this sort of usage is certainly enabled by the brevity of the term. I think a lot of people (I suspect the majority) don't make the distinctions I made in my previous comment, and it undermines their efforts to use the term or concept of "toxic masculinity" in a truly helpful way. I've seen many examples of this: right-aligned people trying to discredit feminists as misandrist; feminists unempathetically erasing these distinctions; angry people in Twitter comment sections complaining about toxic masculinity; people who actually know better lazily using it as jargon on public forums where it can be misinterpreted (I'm sometimes guilty of this); and so on.

The term is also problematic in its omitted connotations; one of these omitted connotations is the interconnectedness of traditional feminine gender roles with toxic masculine traits. Many people ignore how these roles interact with and reinforce each other.

For example, where I live (NYC), it's unclear what type of masculinity is desirable to straight women. Most straight women I know are having a hard time finding guys they want to date because they've been trained to be attracted to some traits of toxic masculinity. But since they live in NYC in 2019, they also have learned recently to be wary of those same traits. Most of them are aware of this weird cognitive dissonance, but they can't figure out how to resolve it. Many of the straight men I know genuinely want to avoid toxic stereotypical behavior, and they're concerned about things like coming on strong/harassing women, but because many of the women they want to date still expect men to do more of the pursuing, it's ambiguous how precisely they should behave to make their dates feel desired but not cornered.

There are many less benign examples; I've seen plenty of women who mock boys and men for being too emotional, too prissy, too pretty, or too physically weak. It's not just men who define manhood in an overly-restrictive way. And these same women have to deal with the negative aspects of feminine gender conditioning. It's a cyclical thing, and I think the only healthy way I've seen to deal with it is to think of everybody as both perpetrator and victim. You can use "toxic masculinity" as a shorthand for the component of this approach related to masculine stereotypes, or you can use it in a divisive way that is unhelpful.

The fundamental issue is that "toxic masculinity" is non-technical jargon; it has a highly compressed meaning, but without a technical framework to limit its interpretation, it's also prone to heavy semantic drift through vagueness, malice, laziness, or stupidity. I think that someone choosing an alternative to "toxic masculinity" will have to deal with jargon's fundamental tradeoff between brevity and specificity. I won't be surprised if such an improved term (or at least and attempt at such) comes about at some point, as often happens with non-technical jargon, but picking such a term effectively is a deep problem rooted in the weakness and strength of natural language's inherent vagueness.

Not GP, but: I think it's a useful phrase, though it's often overloaded/misapplied/used vaguely.

Some (generally recognized as) non-toxic stereotypically masculine traits include having confidence, a can-do attitude, being protective, trying to be helpful, being intellectual, being silly, leading people towards common goals. Even anger in moderation is not inherently toxic (we all feel it). These things can all be terrible in excess, of course, so I think sometimes people fairly call an excess of these things toxic (even if the trait itself is great in moderation). It's also bad if you are only allowed to embody a specific set of traits and behaviors; clearly men should not be limited to these traits (though the societal pressures to conform to traits like these are strong and are often what are referred to as "toxic masculinity" in reasonable discourse). Likewise, women should clearly be seen as equally capable of possessing these stereotypically masculine traits.

I think a lot of the use of "toxic masculinity" is in reference to the fact that men are often strongly discouraged, explicitly and implicitly, from doing things that are seen as non-masculine. This causes behavior that's harmful to oneself and others, like taking out all your emotions as anger, or feeling like violence is the most powerful and respected way to make an impact. In more benign forms, it can come from an excess of an otherwise good trait, like going from confident to cocky. Many people will use the term vaguely/badly, but I think the usage I just described is useful.

I think the most important point is that "toxic masculinity" (in its useful and precise form) is very different from saying "men are toxic". It's closer to "men face toxic pressure due to people's gender expectations that cause them to suffer and sometimes hurt others as well". It's about the societal pressure to conform to a narrow and inadequate spectrum of behaviors. It's like men are expected to span a 3D emotional space with only 1 emotional "basis vector" since the other 2 are effeminate. The 1 basis vector is not itself "toxic"; it's the denial of the other 2 vectors.

They added the last important missing feature from ITAMatrix, which was a direct link to a prefilled checkout page. That part is nice.

A lot of people love working on more esoteric technical things and get more satisfaction from the intellectual component than its direct utility (I certainly feel this way about some things, though trans-architectural portability is not one of them). I would imagine this type of person is better-represented in this sort of field.

In addition, this kind of work does indirectly help keep Intel competitors viable, which helps keep Intel in check for everyone. Stuff like that is pretty exciting in its own way.

Links say you need to log in; by “anyone” do you mean subscribers?

Also worth pointing out that there are 4 times as many 777s as there are A340s. Doesn’t mean A340 isn’t great, just that the odds of no accidents go up with fewer flight hours.

The popularity of A320 and 737-NG makes their safety records particularly impressive.

> The crux of the argument is you will still need the algorithmic complexity in the end with or without LIDAR, so it doesn't add any advantage.

The approach described still uses the same ML on a Lidar data representation, but with an extra step on image recognition to put it into that data format. Image recognition only adds computational complexity post-hardware-sensor because you first need to generate a Lidar-like 3D map. So that is conceivably an advantage for Lidar in the long run given that its method for making the 3D representation of the space (upon which the ML runs) is dirt-simple physics-wise—much simpler than the image recognition algorithms and camera electronics used by image recognition. It's just not cheap yet because the market for processors and cameras is huge and the products have become incredibly sophisticated and cheap. Again, hard to imagine an idea as simple as Lidar not getting much cheaper with scale.

> I'll just paste the quote FTA

My point wasn't an ad hominem one, so I don't think the fact that an expert like Karpathy said "crutch" changes the fact that it's slanted phrasing. But given that Karpathy works for Tesla and has a vested interest in assuaging investors and consumers, it makes sense why he used slanted phrasing. "Lidar is a crutch" makes Tesla sound more visionary than "we don't want to pay for Lidar because it's still too expensive and we think we'll be able to rival Lidar with image recognition." It's a nice way to subtly jab at competitors who are investing in Lidar and frame it as if Lidar was the tech that had catching up to do (when, in fact, the opposite is true).

Since the key point of the article is that ML algorithms work better with Lidar data representations, it's pretty hard to see it any other way. They both go to the same intermediate data representation, and Lidar still wins in a head-to-head. Again, you can argue validly that cameras + image recognition will be good enough, but calling Lidar a "crutch" seems like pro-Tesla spin by a high-up Tesla employee whose job is to make Tesla look good.

Of course, they were presumably using a mostly-stock car with custom niche sensor products, so that comparison would be a bit more favorable in production.

> So in the end I think Elon will be proved right. Those currently investing in LIDAR-based systems will eventually ditch it for purely practical economic reasons. Those that don't will be completely destroyed in the open market.

High quality cameras are insanely complex pieces of electronics and optics. Ditto for processors capable of doing quality image recognition. Large scale manufacturing has made them cheap nonetheless. LIDAR is relatively niche, but if it proves useful to deploy it at scale, I'd expect costs to drop very significantly. The underlying technology uses very simple physics (relative to the algorithmic complexity of image recognition); seems like a solid basis to build a sensor off of.

> LIDAR is a crutch for bad software

You could invert this and say that high precision image recognition is a crutch for ill-suited hardware. The final combination is a product of hardware and software. If LIDAR is currently too expensive or energy-intensive to compete cost-wise at acceptable safety levels, that's one argument, but saying LIDAR is a crutch is just moving the goalposts from "good system" to "cheap hardware".

[edit] Also, just want to point out that RADAR resolutions are way too low to operate a vehicle safely (never mind road signage or other things).

This is cool for off-planet macronutrient production, but plants already do this at scale on earth with excellent efficiency and relatively little supervision. Marketing this as eco-friendly sounds, to me, like utterly shameless greenwashing. I can't picture something short of total ecological collapse or apocalyptic world-wide drought that could ever make it more efficient to feed large numbers of people this way.

Notice that they compare Solein favorably to animal products, which are an order of magnitude less efficient than plant products; if you wanted to make substitute meats, you'd be comparing it head-to-head with soy and similar protein sources. Even their "meat substitute" and "cultured meat" are more processed than Solein; you'd need extra processing (i.e. energy input) to make Solein a comparable meat-substitute product.

I'll also wager that their greenhouse gas emissions models assume some sort of renewable energy source; until we're at 100% renewable energy for industrial processes, Solein production would necessitate using non-renewables for the extra energy input it requires. If you want to switch to renewable energy, and Solein ends up using enough industrial energy to require new renewable power plants (which, at global scales, it absolutely would), you'll have to also factor in the energy, material, and greenhouse gas costs of those new plants. And given that biochemical pathways in plants have had hundreds of millions of years to perfect their efficiency, it's really, really hard for me to imagine any industrial process rivaling plant-based sequestration efficiencies.

Also, I don't know about the nutritional or culinary properties of Solein, but they are starting at a tremendous disadvantage compared to natural plant-based foods that our bodies are already acclimated to. You'll need extra energy for the industrial processes creating flavors, vitamins, and other nutrients, as well as the processes required to make this glop palatable (see point above about the comparison to processed meat-substitutes).

If you're worried about CO2 production from farms, you could switch to mostly plant-based diets with non-animal meat-substitutes. Plants literally build their own solar arrays (called "leaves"), so they don't compete with other industrial power uses for still-scant renewable energy. You could use renewables for the industrial processes associated with agriculture. Some stuff, like methane production in rice paddies, is a real greenhouse gas issue, but the general approach of using plants as an integrated solar-powered carbon sequestration and food generation solution is well established and constantly improving through GMOs and farming practices.

If you had feedback for where you were pointing I think most people would get pretty good pretty quickly (excepting people with movement disorders). It wouldn't be very precise, but I think it would be great for lazily triggering multitouch things like zoom/scroll/page-up/page-down/switch-app etc.

To elaborate on this a bit: you can talk about electric and magnetic fields as properties of space itself, and they have a fully-established physical reality to them. Moving charges appear as terms in Maxwell's equations making contributions to the field, but the field is the ground truth.

Magnetic fields from moving charges are pretty cool because you can show that they are equivalent to electric charges with relativistic effects due to charges' motions factored in. But thinking about relativistic motion equations for a huge number of particles is a tough starting point conceptually if you're just trying to find how a particle at a particular point in space will behave. So pragmatically, magnetic fields are a great conceptual tool. They also hold up straightforwardly once you bring in special relativity.

One thing that's very nice about electric/magnetic fields is that they are local. You can calculate the effect of magnetic forces on solar ejecta by just calculating the local field. It's much simpler conceptually to talk about how a particle will respond to a local magnetic field than it is to talk about how it will respond to a tremendous number of moving charges. It also ends up being very mathematically elegant; Maxwell's equations are very beautiful and conceptually simple, and their common applications are likewise beautiful and simple considering the seeming complexity of EM phenomena.

Furthermore, in some cases, like those involving light (EM waves propagating through space and time), you don't even have the luxury of knowing where the charges are that produced the field/how they moved: if a light wave is striking a piece of material from some distant light source, the only thing you know about/can measure is the field itself; it's therefore really nice to be able to do all of your calculations using the field.

But EM fields are more than just frameworks for understanding or calculating forces. They are very physical things. EM fields can literally carry linear/angular momentum and energy away from them. Let's say you have a satellite shaped like a long bar with two lasers at opposite ends pointing in opposite directions orthogonal to the main axis of the craft:

                       ^
  beam 1 points up ->  |
                       ==============
                                    | <- beam 2 points down
                                    v

Firing these lasers at very high power is like firing very weak thrusters, and your satellite will start to rotate; its angular momentum will change! Even though the light waves from those lasers are not interacting with intervening matter (at least in the classical picture), it's still carrying carrying away momentum and energy; from the center-of-mass frame, the EM field now has clockwise angular momentum (from the view of the above diagram), while the satellite has an equal and opposite counterclockwise angular momentum.

This comes up again in general relativity, where EM fields energy and momentum contribute to the curvature of spacetime in precisely the same way as mass does. In other words, if you had a huge amount of light passing through a region of space, it would be just as gravitationally attractive as an equivalent amount of mass (after converting units using E = mc^2).

When you think about it, conservation of momentum + the speed of light + EM fields' ability to move objects (i.e. change their mass and momenta) necessitates that they have this sort of physical reality. In the bar-satellite example, the laser beams are pushing back on the ship; since light doesn't travel infinitely fast, there is no equal-and-opposite reaction from some regular matter that the EM fields would mediate instantaneously, i.e. no regular matter that's pushing against the satellite to get it spinning (as would be the case with a regular thruster used for stationkeeping). The momentum has to go somewhere; as it turns out, it goes into the field itself in an extremely well-defined way.

So, again, EM fields are the real deal; moving charges contribute to fields, but the fields themselves have their own well-measured, well-described physical reality in classical physics.

Yeah, people who are completely direct about what they want in both their profile and communications tend to have a much better time on apps. I try to put things in my profile that will be charming to people I might like and alienating to people I wouldn’t, and it works very well. You have to have some standards and filter out obvious bad matches quickly. Since you’re not dating in your social circles, you don’t have the advantage of a pre-vetted dating pool, so you have to just recognize that most of the people you see won’t be a good match at all.

Oh yeah, by “swipe indiscriminately” i really meant “swipe on literally everybody and then downselect from matches”. It’s a very obnoxious thing to do. Pretty easy, but it discourages people from messaging any guy first when you know there’s a good chance they’ll just unmatch right away.

I don’t think those guys are even close to the majority of straight male users, but since they always swipe right, you have a 100% chance of matching one of them if you swipe right too. So they pollute your match list disproportionately and end up making it worse for everyone. Pretty annoying.

Right, yes, That’s what I meant: it’s advantageous for fascist agitators to encourage traditional repressive masculinity for the reasons you outline. Being susceptible to it makes you useful to fascists.

Some of us really like the parentheses!