AkshatM

I'm from India too. It's disingenuous to say things like "Bangladeshi Muslims are not desired" and not mention the massive backlash the Citizenship Amendment Act received. It's almost as if people recognize arbitrary discrimination is the opposite of just.

Attitudes like yours - justifying entitlement without holding its corruption accountable - is why people have to leave their home countries: their own insufferable countrymen.

The effort here is laudable, and this is cool! But your real competition isn't regexes, but parser combinator libraries. For cases where readability and matching matter, parser combinators are the go to. It's a bit difficult to see how this improves on that front.

I asked Kenton Varda (the creator) this one after discovering all its cool capabilities, and it seems like promoting it just hasn't been a focus.

It is heavily used at Cloudflare, though, and is the basis for a lot of cool stuff we're doing: https://blog.cloudflare.com/capnweb-javascript-rpc-library/

First off, my heart goes out to you. Thinking that you chose to be a bystander must have cut really core to your sense of identity. It's been difficult to reconcile being a good person and choosing to do nothing for you, and I don't wish that kind of doubt and angst on anyone. I'm sorry that it's affecting you now in ways you are worried can be counterproductive.

Second, I want to offer an unsolicited perspective that may help you. I read your linked post about Lowtax's death and how it went down for you. You're blaming the 2021 version of yourself for not making the right choice, but what I read wasn't selfishness or cowardice stopping you from speaking up - it was an assessment of your ability to make a difference. I'm not saying you should never speak up when you feel something is off, or demonstrate concern, but that person in 2021 understood the many, many barriers to saving Lowtax: the unpredictability of his response, the ambiguity of the message, the concern that it wasn't your place, the large possibility you were overreacting. What would you have done? What would you have said? How would you have been able to reach them that closer friends, family, and others would not? You are telling yourself you somehow had the knowledge, therefore the responsibility and power, but in reality you had no power and only the barest hint of knowledge. You had no idea if what you were doing was right or not, how it would be taken, how it would have been perceived, or what the consequences were. You are only _guessing_ that by taking some miraculous action you could have averted the outcome, but in truth no action might have been good enough - you could have spent a year and three months getting them the resources they need, and still they could have done this.

When someone takes their life, it isn't the fault of the lone stranger who wanted to say hello. You did not choose to be a bystander - rather, by taking their life, they made you a bystander. You have to separate your sense of moral esteem from the consequences of someone else's choices. Yes, anyone could have done something, said something - and for all you know people did!

Inaction by itself is not a moral hazard. Inaction when armed with the facts, the ability to make a difference, the understanding that the difference made will be significant or material, and where the action is scoped enough to not jeopardize the situation - this is the real hazard. In 2021, you did not have the facts or the ability to make a meaningful difference - only the slim possibility that a message in the dark might have ruined nothing and saved life. But that is fantasy talking, not genuine pragmatism.

I hope this gives you some permission and at least a framework to measure your "overhelping" against. Therapy has been helpful for me in the past when dealing with complicated emotions.

You're thinking of probabilistic systems at run time. People are talking about probabilistic systems at compile time.

Whatever gets generated, if it passes tests and is observably in compliance with the spec, is accepted and made permanent. It's the clay we're talking about Jackson Pollocking, not the sculpture.

Before the outage, Cloudflare had Cloudbleed: https://blog.cloudflare.com/incident-report-on-memory-leak-c...

The move to Rust was partly motivated because it prevented that entire class of errors. No more out-of-bound reads, or data races. The compiler audits these missed spots.

Now, you could say a managed memory language would suffice as well. Perhaps it could. But we needed performance, and no memory-managed language met those performance needs then or today.

I get you're making the case that Rust isn't perfect for all use cases, but Cloudflare's scenario is the exact corner case your argument falls apart in: we needed fast and safe in an environment where not being fast or safe had real business consequences, and nothing else except Rust gave both.

I find the contrast between two narratives around technology use so fascinating:

1. We advocate automation because people like Brenda are error-prone and machines are perfect.

2. We disavow AI because people like Brenda are perfect and the machine is error-prone.

These aren't contradictions because we only advocate for automation in limited contexts: when the task is understandable, the execution is reliable, the process is observable, and the endeavour tedious. The complexity of the task isn't a factor - it's complex to generate correct machine code, but we trust compilers to do it all the time.

In a nutshell, we seem to be fine with automation if we can have a mental model of what it does and how it does it in a way that saves humans effort.

So, then - why don't people embrace AI with thinking mode as an acceptable form of automation? Can't the C-suite in this case follow its thought process and step in when it messes up?

I think people still find AI repugnant in that case. There's still a sense of "I don't know why you did this and it scares me", despite the debuggability, and it comes from the autonomy without guardrails. People want to be able to stop bad things before they happen, but with AI you often only seem to do so after the fact.

Narrow AI, AI with guardrails, AI with multiple safety redundancies - these don't elicit the same reaction. They seem to be valid, acceptable forms of automation. Perhaps that's what the ecosystem will eventually tend to, hopefully.

Piece of UX feedback for the product team behind Updog: company logos are not searchable. It should be easy to Ctrl-F and find a relevant cloud on that detector, instead of scrolling alphabetically.

The Hindustan Times is widely known in India, and is considered reliable as a source of information.

I wasn't sure which thumbnails you were referring to on a quick perusal of their YT channel, but my mind instantly assumed you meant the sensationalism associated with Indian media - that's regrettably a feature, not a bug, in that ecosystem.

Having met Casey Handmer personally, I'm confident the last thing he could be accused of is being anti-science. The man knows quite a bit about his domain of expertise!

I'll also point out that it's actually harder for entry-level foreign workers (read: international students) to get jobs in general. F1 visa students are limited to one year of OPT, unless they have a STEM degree, in which case they can work for two more years for an e-Verified employer only in a field related to their major. There are significantly fewer e-Verified employers in the country. So you have a concentrated labour pool vying for a much smaller fraction of jobs than the general population - remember, we can't even be employed at McDonald's on STEM OPT.

Another factor that complicates the analysis is that foreign students do not necessarily represent the same cross-section as US students. The high cost of US education tends to bifurcate the sample size into a bimodal distribution: you have rich foreign students who can afford tuition rates, or you have scholarship students who earned their scholarship quite fairly thanks to merit. Neither population can be fairly compared against the average US citizen in terms of hiring likelihood - either you're pitting the top 1% of someone in the home country to the median US student, or you're comparing someone who's got the resources to find opportunities in ways US students can't.

Finally, it's been my experience, seeing others though the F1 -> H1B pipeline, that most people pursue that pipeline through a master's degree rather than a bachelor's degree. This is because both immigration law and tuition rates incentivise shorter programs for advanced education. If you're comparing the hiring rates of master's students to bachelor's students, naturally you're going to get a revealed preference for master's students.

tl;dr the simple statistic cited needs critical questioning.

That's not an apples to apples comparison. You don't know if the foreign workers were senior-level employees, for example, or whether there were specialized requirements they might have been hired against that entry-level grads won't qualify for.

I agree there should be more entry-level roles. But this is not an indictment of the system. You should ask what percentage of those foreign workers were hired into entry-level roles.

(Not Peter, not a lawyer)

It is hard not to be moderately worried, since the administration has already made concrete attacks on citizenship pathways, immigrants and legal residents. Most notably, it's targeted birthright citizenship via executive order, tried to push through the power to unilaterally revoke green cards, and arbitrarily cancelling the ability of universities to host foreign students.

More importantly, successful legal challenges to these efforts are being vacated or reversed by the Supreme Court. That means the administration's attempts aren't being fully checked by balances.

You should take the rhetoric fairly seriously. There is no good-faith explanation for this pattern of attacking people who have gone through the process faithfully, and no credible ongoing efforts to prevent those happening.

Need an example? Good. Coding.

You must be paying your software engineers around $100,000 yearly.

Now that vibecoding is out there, when was the last time you committed to pay $100,000 to Lovable or Replit or Claude?

I think the author is attacking a bit of a strawman. Yes, people won't pay human prices for AI services.

But the opportunity is in democratization - becoming the dominant platform - and bundling - taking over more and more of the lifecycle.

Your customers individually spend less, but you get more customers, and each customer spends a little extra for better results.

To respond to the analogy: not everyone had $100,000 to build their SaaS before. Now everyone who has a $100 budget can buy Lovable, Replit and Claude subbscriptions. You only need 1,000 customers to match what you made before.

Joe will be fine. Cloudflare is pretty good at differentiating humans from bot traffic - see how we do it here: https://developers.cloudflare.com/turnstile/

The idea behind the headers is to allow bots to bypass automatic bot filtering, not blockade all regular traffic. In other words:

- we block bots (the website owner can configure how aggressively we block) - unless they say they're from an AI crawler we've vetted, as attested by the signature headers - in which case we let them pay - and then they get to access the content

(Disclosure: I wrote the web bot auth implementation Cloudflare uses for pay per crawl)

The blog post covers this. The announcement also drops relying on spoofable user agents for crawler identification and requires crawlers to voluntarily identify themselves via RFC 9421 cryptographic message signatures to get access: https://blog.cloudflare.com/introducing-pay-per-crawl/#payme...

There are likely incentives for AI companies to try to simulate human users as much as possible, but the value proposition here is that CF is so good at identifying and stopping those that signing a request becomes the path of least resistance.

Disclosure: I am on the team that wrote the RFC 9421 message signature implementation at Cloudflare and its use in the pay per crawl project. A separate blog post went out here: https://blog.cloudflare.com/verified-bots-with-cryptography/

This blog post plays too loose and fast with its terms to be useful as a model of anything:

- It's tautological. A tribe is defined as a connected graph of members with a shared mean opinion, and then God is defined as that shared mean opinion. You can't do that - you can't assume what you're setting out to prove. Nor can you just call something a term (like 'God') and then assume it actually has other properties we associate with the term - you have to demonstrate that. In other words, all this blog post does is propose a definition and fail to show how that definition remotely captures that interesting properties we care about. I can call birdsong "trees" all I want, I doubt I'll be getting calls from the department of forestry anytime soon.

- It assumes a bad foundation. N in the author's formulation is clearly infinite (I challenge anyone to describe a procedure to enumerate all possible opinions), but the author assumes its finite. What they really want to mean is that you can assign a numeric score to a person that describes inclination towards agreement with a particular proposition, which is still workable.

This is, frankly, bad armchair philosophy.

I don't think this is the right conclusion. It's more: people with no expertise and hidden agendas should be forced to behave in line with the best available evidence unless there are no other recourses for them. People with expertise, using the standards of high-quality data gathering and analysis, should be exploring alternative hypotheses. Galileo, Kepler, Newton et. al. all fell in the latter camp. COVID denial and flat earthism is an example of the former.

This really just boils down to saying: there is a rigorous evidence-based standard for updating our best available knowledge, and people who don't know how to adhere to that standard should be disqualified from recommending deviation (they are of course welcome to believe and do unto themselves as they wish, they just shouldn't be allowed from promoting it unless they have solid evidence for their claims).

I mean, the Simpsons has predicted several monumental events multiple times in its history. Random can be just that good.

I believe the examples in the post are all definite integrals, not indefinite ones.

The Greeks knew the Earth was round because they could look at ships appearing over the horizon and observe that the tops of their sails would appear first. Eratosthenes was further able to calculate the radius of the Earth to a decent first approximation simply by using the shadows cast by two sticks at reference locations. If they could do it, you can do it too[0].

The idea that you "cannot verify" is a very pedantic comment, and using flat Earth as a basis only makes it more comical. Of course you can't know anything "with absolute truth", but nobody cares about that. The relative distribution of evidence strongly disconfirms some hypotheses, there is already a culture of strong distrust and independent verification in the hard sciences (see the LK-99 saga), and you can rely on the long-term output of that process in the same way you can rely on your GPS to just work without needing to independently launch your own satellites. Needing to be your own scientist now is like needing to be your own farmer: completely unnecessary for most people.

[0] Yes, flat earthers claim these results are spurious because of optical illusions caused by hot air or the like. But the relative distribution of the evidence for _that_ hypothesis is pretty darn slim, which only furthers my point.

I would push back on the interpretation that syntactic sugar, by virtue of making programs shorter, is more expressive. APL folds entire compositions into single character sequences, but one would probably not argue that it is more expressive than Brainfuck.

Rather, it seems expressiveness as used in common parlance is closer to ease of comprehension. We like programs that are easy to "think of", that are "natural" to write in that they are succinct and "natural" to read in that they mimic what you are conceptually trying to accomplish. This has very little to do with programming language design and more the experience of the language user - aesthetics, really.

I would like to see social factors put to a similar rigor of analysis then, since it seems to me that you are proposing socialization is (1) a major factor and (2) that the socialization of women towards non-leadership roles is incorrect from a normative standpoint.

This is accurate, though, to be clear re: (1), I am saying that socialization of both men and women plays a role in these disparate distributions. Men are also socialized, to a great extent.

My counterpoint is that motherhood, pregnancy, chastity, fertility, these are tied up in the biologic of a woman, and these fulfill important roles that can’t be explained by social factors. We are not at the stage where we can replace women with mass artificial wombs and then no child has a “mother” they know and grow with. Rather we know from studies that a nurturing mother is incredibly important for a child at the epigenetic level all the way to the social level.

I think you are blurring an important distinction here: motherhood during pregnancy and parenthood of a child. Yes, you clearly need women for the first, but that doesn't have anything to do with the latter - a nurturing parental figure is needed, but not necessarily women. Surrogate children, for example, thrive despite their biological mothers being absent in their developmental lives, as do adoptive children from early ages.

If you think about it, children have a variety of developmental needs, but none that are conditioned on requiring their parents to be of a specific gender. Children raised by successful gay male couples are simply not developmentally harmed in any way by any study, metric or measure, though neither parental figure is a woman.

The narrative that you need a nurturing, attentive and competent parental figure is certainly valid - but the idea that only a woman can fulfill that role isn't, because there's practical and ample evidence otherwise.

So much of a social role is devised around motherhood and so much of male mating and paternity is focused on not wanting the woman to have children who are not his

But what about this is biological, exactly?

The trouble with the biological explanation is it doesn't explain exceptions very well. There are women who don't want children, men who don't want children, asexual people who don't care about sex but love the idea of raising a family, hikikomori who don't want any social contact, and more. That these people exist, are rational thoughtful individuals with full lives, and don't conform with the expectations biology supposedly places on them suggests biology can't be that strong a force in the first place. (You could claim these people are defective, of course, but that's circular and motivated reasoning - if biology is infallible except when it isn't, then we should just accept it's fallible instead of demonizing the exceptions).

The other trouble is that the biological explanation is very selective. As a species, we do a great deal of unnatural things. We eat cooked food, we wear clothes, we have manners, we employ language, use toilets, and more. Yet, somehow, when it comes to the topic of finding a mate, we somehow argue that our instincts, rather than our society, has shaped us into who we are, despite having shaped almost everything else.

Take one example where socialization has overridden this supposed base instinct: beauty standards. Small feet are not correlated with reproductive success, yet at some point in history foot binding was introduced. There are tribes that engage in neck elongation and other forks of bodily mutation that people learn to find attractive. Yet it would be foolish to argue that this is somehow a biological imperative - after all, we no longer include small feet as an index of beauty. Why does sexual reproduction get the special treatment?

To bring this idea back to the original talking point, the idea that motherhood is something you need to aspire to can't be something innate because there are many women who do not want (or even like!) children at all. Promiscuity and parenting preferences aren't "male" phenomena or "female" phenomena - these are people phenomena, and the currents of what we encourage, enshrine, highlight, reference, and consider weird shape how people internalize what parts of themselves are okay and aren't okay.

I also challenge your notion that scientific studies conducted in a research study fashion are the preferred criterion for evidence. A study without context and understanding is dangerous.

Of course. But bad methodology at arriving at said context and understanding is much more dangerous, and significantly more common than bad analytical science. One rule that is usually overlooked, for example, is that explaining the outliers is much more important than explaining the average of the distribution.

So when you say women are pushed into caregiver roles… this is obvious if you understand the concept of motherhood.

It's not just that women are pushed into caregiver roles - it's also that men are pushed away from these roles. It's an invisible pipeline that begins with how we think about feelings and how to process them. In general, men don't receive the emotional guidance women do. We're encouraged to think about sex as a prize, status as a measure of self-worth, anger as a primary means of self-expression. Close male friendships are rare in comparison to close female relationships. Suicide is much more prevalent, as is violence. The joy of emotional labour and pure authenticity is never presented to us until we experience and mine it for ourselves - or rather we are told it is only possible to have that when we are in relationships with submissive women, home with children who are supposed to love us for all our faults we never work on, at which point we explode because we cannot handle the idea that mature adult love is so much more than about just blind devotion. All these things shape our perspective on what's right for us.

So, no, I would say it is not obvious. I would say it glosses over the lived experiences of many men and many women to arrive at that specific conclusion.

As for a rise in women leaders of countries, I will say overwhelmingly the political and business class is male,

But that doesn't mean anything? Of course they are overwhelmingly male - you've accepted that doors were formally barred to women for a long time, and there are invisible doors that continue to operate even now.

We saw a rising “feminization” of society in terms of corporate interactions and corporate decorum now being a default in many respects for how society operates; there are also ideological reasons to elevate “women leaders”; all of these are reasons I suggest you see a rise in women leaders of countries

I don't understand what you mean by "corporate decorum/interactions". It sounds like you are arguing that it's only out of politeness that women are now allowed to be leaders. I assure you, when Boudica led a revolt against the Romans, savaging city after city, it was not because the men she led were being polite.

Take a first principles approach. It's possible to articulate what qualities or skills are required for competence in leadership: competence, assertiveness, popularity, diplomacy, and decision-making skills. No item on this list disqualifies women or even disadvantages them. There is no shortage of tough women out there.

If you look at societies where ostensible crude military control and authority are important to display, or even in war situations, we see that men are by far the ones involved in combat. ... A big question to ask, why so few women in the military roles then, if they have been open to women now? If you look at women’s physical performance it becomes obvious. We aren’t at the stage of all robot armies yet; and the militaries of today are still by far men.

I've written a lot of thoughts above, but I'll reiterate once more because I'm tired and can't do full rebuttals: biological explanations are weak explanations, because they don't explain the outliers. The fact there are military women in the first place is the interesting finding, not their rarity.

There are some errors in your reasoning, though.

Here's one. When people collect datapoints and observe a difference in one case not present in the other, they sometimes fall into the trap of thinking that their dataset is exhaustive - that it captures all the variables that matter.

In other words, it's easy to start with the observation "women tend to be more represented in fields that exhibit qualities of care and less represented in leadership roles" and conclude that the absence of the variable "has leadership" makes up the core difference.

But this is a conceit that you, the observer, made. Your choice of variables to look at influenced your observations of what is different and therefore important.

Here's a counterpoint that contradicts your observed conclusion. Female heads of state and politicians have been increasing in number throughout the last century, and have done remarkably well in those roles. Has the nature and autonomy of executive power changed like medicine (according to you) has? It has not. It is a sign that you have not taken all variables into account.

Here's another error in your reasoning. Why those variables specifically? How do you know those are the variables your subjects are thinking about? When women sign up to be doctors and nurses and not CEOs, are they expressly telling you the shared decision making are the important variables? Or is it because you inferred that those variables are important because you can measure them?

In the interest of correct methodology, here are some other variables that people have suggested to explain the same data: women are conditioned or socialized into early expectations of caregiving roles; women are rewarded for pursuing caregiving roles in a way men aren't; there are more barriers for women than for men, and so on.

Now are these variables the correct answer? We don't know. It is a complex topic, because humans are complex. It is possible biology and perceptions of role play a part, as does socialization and others.

But what we can do is go a little meta and ask the likelihood that a biological variable is a good explanation in the first place. Consider what a good explanation has to do: it has to provide a causal mechanism, and be able to explain how the cause led to the effect you see. The trouble with every single biological explanation that's been proposed in this area is that they don't provide this causal mechanism. What exactly is it that makes women gravitate towards submissive roles? Is it estrogen levels? Hormones? Menstrual cycles? But then people on hormone therapy, people born with extra chromosomes, people with hermaphroditic parts, and so on should all display manifestly different behaviour and choices, which they don't. To explain these differences, you need to invoke more and more factors, until you end up with epicycles all over again.

More generally, the history of biological explanations simply don't fare all that well in comparison to sociological explanations. A reasonable prior is to assume that it may play a small part, but that larger effects are driven by how society treats and works with people.

With respect, there are some errors in your thesis. Specifically, you're attacking some strawmen.

Here's the steelmanned version of the ideas you're saying are causing issues:

- Instead of defining men / women, think of each human as a person. There are no strong men, sensitive boys, intelligent women, vulnerable girls - there are only strong people, weak people, hard people, soft people. Our job as society is to compassionately accept people as they are and encourage them to be their best authentic selves, instead of arguing there is some role they must force themselves to fulfill because they were born with a set of genitalia, because this promotes individual happiness and is the kind thing to do.

- People's virtues are learned, not innate. Independent people can learn to accept help. People who are unsure of themselves can gain confidence. People are allowed to be vulnerable, but also to have the space to find roots, solidarity, strength and growth. Roles that traditionally would have gone to someone with a penis - head of household, soldier, provider - are achievable and manageable by anyone, because the virtues needed to hold those positions are learnable and not rooted in biology. The same holds in reverse - roles that would traditionally have been held by women are open to everyone.

- People should be given freedom to flourish as they best see fit. They have rights to their bodily autonomy, securing their financial futures, achieving scholastic pursuits, respect and equality for their contributions, and more. Historically, this has not been the case, and this does not just apply to women's rights here - it applies to anyone who has had these opportunities denied.

- Some human but deleterious attributes we should all grow beyond. Brutishness is an excellent example - as is shallowness, closed-mindedness, entitlement, ignorance, greed, dishonesty, and more. Some attributes are contextually awful - being stoic, for example, is a boon when you need to handle stressful situations, but it can also manifest as a lack of empathy for other's emotional lives.

Note that none of these are attacking masculinity - if anything, point two contradicts exactly that, literally anyone can learn the virtues associated with masculinity and is why trans men are supported even conceptually. Rather, point four holds - there are a subset of awful virtues that are unacceptable or contextually awful from anyone that have been attached to being a man, such as entitlement to sex, resorting to anger in lieu of healthy expressiveness, and assuming they must shoulder all burdens instead of being allowed to seek help. In other words, mainstream movements are attempting to abolish toxic masculinity, not masculinity itself.

You discuss the emotional needs of boys and girls as being different (which may or may not be valid, depending on the best available evidence), but then you argue that treating boys like girls is not the right thing to do, which is not what anyone is proposing at all. Again, the idea is that people are people, not their bodies. Girls and boys are people, capable of feeling, wanting and expressing the full range of emotions. People want to treat children in ways that allow them to be emotionally expressive and mature adults.

Further, when you talk about the patriarchy, you are again not taking point one and two into account. Nobody wants authority, strength, boldness, vitality, etc. to go away - these are all great virtues. Rather, they want the idea that these qualities are somehow gendered to go away. If you want strong leaders, find them in both genders. Additionally, historically, the patriarchy you are discussing has tended to enshrine point four virtues rather than eliminated it - it's the same system that inflicted foot binding tortures, made Indian women throw themselves into funeral pyres when their husband passed, allowed female infanticide to flourish, and so on. These are all terrible things and bundled into a parcel of ideas about what men "should" be: not just protectors, but architects of their offspring / wives / sister's fates. We have managed to overcome many of these core underpinning beliefs, but there still remain a lot that should not go unchallenged.

So, respectfully, I would ask you to at least attack the right thing. It is possible your value systems are opposed to all the points above, and that's fine. But if a value system has to reach towards attacking strawmen of other viewpoints, then it is not a good sign that that value system has been arrived at fairly.

Actually, you can find online examples that demonstrate how easily you can teach a neural network how to do y = x^2

https://stackoverflow.com/questions/70407674/how-to-teach-a-...

Neural networks, from one perspective, are good at finding hyperplanes between data points from two different sets. You can teach a neural network how to recognize X belongs in your dataset of choice, so it's relatively easy to find other points in the dataset of y = x^2 or so with that building block.

I am no expert, but mechanistic interpretability seems to make the most sense in understanding individual NN layers: each layer recognizes a different combination of features corresponding to a certain pattern and feedback from the layers below. At a scale of 40B+ parameters and 120K context windows, we start to lose track of what the machine is doing, but it's not inconceivable that with such large numbers it can regurgitate very large portions of text with 100% accuracy.

Scale matters a lot.

I get what you mean, but I think you've formulated this incorrectly. Let P(Y) be your prior about government corruption. Let X be the event that SBF is arrested. You want to compute P(Y|X) using the Bayesian update formula and then set P(Y) = P(Y|X). That is what is meant by re-evaluating your priors.

You're modelling X and Y as propositions and you're correct about the inference of ~X and ~Y, but Bayesian updating is about degree of belief in those propositions, which your inference is not a claim about.

My trouble with the argument that crypto will liberate underprivileged classes in general is that it seems to miss key holistic issues I'm enumerating below. Perhaps I'm wrong, and I'd appreciate knowing the way I'm wrong.

1. Crypto wealth only translates into real wealth if there is either (a) vendors who accept native crypto as tender or (b) there are exchanges that convert from crypto to authorized legal tender in a country.

However, it seems (a) has not happened in a meaningful scale without (b), and (b) has been frustrated because of underlying price volatility, high fees, low transaction processing compared to traditional finance, etc. There's the problem of "too many cryptocurrencies" which frustrates user choice.

(I'm aware independent proposals to "fix" all of this have been proposed, but there don't seem to be any clear winners e.g. stablecoins should have "fix"ed price volatility, but the leading stablecoin - TerraCoin - collapsed.)

The net impact of this is that people are cut off from real-world markets they live in. Someone who has no access to traditional finance will not have an easier time purchasing goods and essentials with crypto wealth, and this their ability to participate is only marginally improved.

2. Crypto has higher barrier to entry than traditional finance because it requires higher than usual technical literacy, which people in these classes don't have. Wallets need protection, smart contracts need auditing, and so on. I would struggle getting people to use (say) hardware tokens.

3. There's nothing ultimately stopping governments from banning crypto exchanges or crypto vendors or more if they really want to prevent the disenfranchised from getting some. They might not eliminate the network, but they can certainly go after people for suspicion of participation in it (such as vendors). The crypto model requires governments to be foiled by technical limitations, but this is rarely the technique they need.

Based on all of this, it seems like basically you need to solve the problem of vendors really caring enough about crypto for regular folks to get an advantage when the advantage is not clear. Is this not the case? What have I missed?

Related concept: Kung Fu Chess, which is truly real-time and simultaneous: https://www.kfchess.com/

Nancy Leveson's _Engineering a Safer World_ (available for free here: https://direct.mit.edu/books/book/2908/Engineering-a-Safer-W...) is required reading, in my opinion. It talks about applying systems theory to reducing industrial accidents, using various case studies such as the Bhopal gas tragedy, which maps automatically to reducing outage severity and incident severity for tech workers.

The key insight it applies is that root-cause analysis to fix and improve situations is limited because choice of root-cause is somewhat arbitrary - it really depends on where you stop probing further. Engineers should not rely on root-cause analysis to identify failures in _the system_, only on failures of components. Nancy proposes alternative techniques instead to identify and improve broken systems.

The detail, quality and depth of her treatment makes it immediately practical and useful. My one complaint is that it is a bit long, and not all of it is easily condensed into short treatments - but I guess verbosity is the tradeoff for quality?