Hacker News new | past | comments | ask | show | jobs | submit | Frieren's comments login

The police investigating thief costs more than the money they get back. Should we stop prosecuting thieves?

When criminals see that they get caught for evading taxes then many of them reduce their activity. Dissuasion is also part of the equation.


I don't know where you're at but where I live the cops don't investigate petty theft unless there's violence involved.


> won't do anything to balance the budget.

Many goverment expenses come from corruption, from expensive payments for health care (that exists because billionaires put money into politicians pockets, to overspending in the military that also come from kickbacks, etc.). So, reducing the numbers of billionaires reduces corruption and overspending. (And even what is spend can be done so more efficiently, money going to teachers and nurses instead of profits for big corporations).

Cutting on corruption has a big impact into goverment spending. Billionaires and inequality increase corruption, as there is easy money used to bribe politicians. End one and the other follows.


do you have any sources you can cite for this?


-https://www.hks.harvard.edu/publications/inequality-and-corr...

VI. Conclusions and Implications

In summary, income inequality is likely to be a significant and no less important determinant of corruption than economic development (and thus many other variables for that matter). ... Inequality increases corruption, which in turn deters investment and growth.


“Correlation is not causation” The paper shows that wealth inequality and corruption are correlated, and pontificates that maybe reducing inequality with wealth redistribution might reduce corruption, but offers no evidence that this is the case. Your chain of reasoning is not really supported by the paper.


If I stopped paying my taxes my goverment will not even notice. It cannot run on my money alone for even seconds.

Is that an argument for me to not pay my taxes?

> (and crashing the stock market in the process)

Some scaremongering there for a straw man's argument that you made up.

Put thieves, tax avoidance is theft, into prison. Getting the money back is not enough. Society needs justice, and people stealing from everybody should not only pay their fair share but also be punished when they steal from society.


Here's a graph of America's national debt. Its rate of increase clearly starts spiking...right after the Bush tax cuts.

https://www.statista.com/chart/28393/us-public-debt/

I think it's indisputable that tax policy has an impact on a government's health.

Yes, that's not exactly the same thing as prosecutable tax avoidance. But my sense is the real problem here is quasi-legal tax avoidance, and that this small $1.3 billion figure is just hinting at how easily more taxes could be recovered with less lenient tax policies. (As well as better-funded enforcement officers...) But that gets dangerously close to also saying, "let's raise the tax rates on millionaires (not just billionaires) back to where it was a few years ago..."

And let's be honest: that's why we're having this discussion. Government spending is bad and part of the problem, according to one side - with taxes just a way of enabling it. There's been a lot of words devoted to extolling the values of lower taxes in creating jobs and innovation.

So I think at some point, people just start rooting for tax avoidance.


legal tax avoidance has nothing to do with the $1.3b discussed in this article. I'm all for collecting legal taxes - but I'm against pretending like we can fund the budget with any reasonable tax policy. i'm trying to make the point that if we try to fund the bloated budget at the current gdp, it won't (just) be the super-wealthy who suffer, it will be the people who can't retire, the people whose kids still need scholarships and loans for colleges, the workhorses of the economy, the middle class.


> u.isAdult()

Being adult is not a property of the user but of the jurisdiction that the user is in. In some places or some purposes it is 18 but it could be, e.g., 21 for other purposes.

If you software is not going to just run on the USA it is not a good idea to implement isAdult in the user but in a separated entity that contains data about purpose and location.


With proper OO you could still implement that on the user object.

    boolean isAdult() {
        return this.age >= this.location.ageOfAdulthood();

        // or this.location.isAdult(this.age);  pick your poison!
    }

…anyway it’s just an example of how to introduce OO concepts. As everything in programming it depends


Just having some fun with bikeshedding here: Yeah, that could work but IMO in a big/international system the responsibility should ideally live elsewhere, since:

* You may need to determine adulthood for a different jurisdiction than where the person currently resides. Their citizenship may be elsewhere, or you may be running a report that expects "adulthood" to be by some other region's standards, etc.

* Sometimes the underlying kind of adult-need wanted is slightly different, like for consuming alcohol or voting.

* There may be a weird country or province has laws that need additional factors, like some odd place where it's a different age-cutoff for men and women.


Yes this is just extreme bike-shedding at this point. But none of this is impossible with more OO principles, like interfaces:

    class User {
        // Convenience function to check if the user is an adult in their current location
        boolean isAdult() {
            return this.location.isAdult(this);
        }


        boolean isOfDrinkingAge() {
            return this.location.isOfDrinkingAge(this);
        }
    }

    interface Location {
        boolean isAdult(User u);
        boolean isOfDrinkingAge(User u);
    }

    class WeirdLawsLocation implements Location {
        boolean isAdult(User u) {
            return switch (u.gender()) {
                case MALE -> u.age() >= 16;
                case FEMALE -> u.age() >= 18;
            }     
        }

        boolean isOfDrinkingAge(User u) {
            return u.age() >= 21
        } 
    }
In the hypothetical that you want to check somewhere the user is not currently:

    class SwedenLocation implements Location {
        boolean isAdult(User u) {
            return u.age() >= 18;
        }

        boolean isOfDrinkinAge(User u) {
            return u.age() >= 18;
        }
    }
    var sweden = new SwedenLocation();
    sweden.isOfDrinkingAge(user);


That feels like unnecessary levels of indirection to provide a method that shouldn't be on the User anyway.

    j = Jurisdiction.fromUserLocation(user);
    j.isOfDrinkingAge(user);


> that shouldn't be on the User anyway.

That’s just your opinion. It’s ok to provide convenience functions. I see no difference between the amount of indirection in our implementations, except mine is in the more natural place and you don’t have to know how to get a location or jurisdiction to answer the question: “is this user an adult?”. Knowing that it uses a location or jurisdiction is an implementation detail that you shouldn’t couple yourself to.

Cheers mate, I think I’m done moving goal posts for this conversation :)


On a side note, this discussion really made me realize how useful concise method bodies would be in Java: https://openjdk.org/jeps/8209434


Wow, thanks for proving that OOP was a mistake.


I challenge you to do it better then.

Also from the guidelines:

> Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something.


> Being adult is not a property of the user but of the jurisdiction that the user is in.

It's a function of both. And I'd argue it's really mainly a function of the person: Independently of jurisdiction, there's at least a rough global consensus what "being adult" means, and most jurisdictions set rather similar (many of them, identical) limits.

A four-year-old isn't an adult anywhere; a fourty-year-old is everywhere.


> Sexual exploitation concerns > Promotion of homosexuality > Incentives for children through virtual currency

The official also highlighted issues with “robux,” Roblox’s virtual currency. Reports indicated that bot accounts were distributing robux to incentivize children’s participation in the aforementioned problematic activities. Challenges in content oversight

> The inability to effectively monitor and regulate inappropriate content on Roblox

The homosexuality part is kind of absurd, but very in line with other Türkiye policies.

For the rest they seem very reasonable. Predatory monetization methods in games for adults are bad, applying the same methods to children is pure evil.

I will prefer strong regulations than banning. The same the we do with gambling.


I don't really see the homosexuality part as absurd. It is to be expected in their less tolerant culture. Remember, Turkey is the only majority Muslim country in the world where homosexuality is legal. They're pretty tolerant comparatively, but their culture is such that they don't want to encourage it. I wouldn't want to live in their society but I do believe in national self determination.


simply false and a simple search brings up more results


> "Even though Superfest glasses were ubiquitous in every bar, canteen and household in the Soviet satellite state, few people had heard of Paul Bittner, Fritz Keuchel and Tilo Poitz, the design collective who gave them their shape."

I guess that it is in the same way you do not hear about the engineers and designers at Apple, Space X, etc.

> I believe the technology in itself, is a bit more admirable and the people involved mentioning, than the mere shaping.

But we hear all the time about CEOs and their opinions, thou. And they are way less interesting that the tech and the engineers that work at their companies.


Yeah, how many designers and inventors of industrial processes, or materials, or techniques or such we know widely in the west?

Never mind that even for those well known to us engineers, regular people couldn't name what they did either, except maybe Edisson, Tesla, Wright Brothers, and a few more. E.g. would even more modern PhD holders know what Watt contributed? Or Lord Kelvin?

People like Gates, Bezos, Musk, and Jobs on the other hand, everybody knows.


Yes, the engineer CEOs like Gates, Musk and Jensen Huang certainly become very well known. Not so much for their engineering prowess (though it is typically great) but for being the symbol of the products they create.


> Not so much for their engineering prowess (though it is typically great) but for being the symbol of the products they create.

They are mostly known for being rich. Very very rich. Way better engineers creating way better products are not well known.

"Great engineering" is not the word choice I will go for to describe their legacy.


> Coubertin called this sport in the service of global harmony – nothing short of a new “religio athletae,” or “religion of athletics.”

I think that he succeeded. I see many people celebrating their nations pride, but also human accomplishment of athletes of any nationality.

One can be cynic about it, but one can be cynic about everything and enjoy nothing. But even that the Olympics exist in our reality, with its problems and injustices, it is still inspiring and enjoyable in many levels.


We need more things that connect the world this way, but they only work if you can set aside the cynicism a bit or at least see the good and the bad and enjoy the good when appropriate and try to fix the bad when appropriate. It's been genuinely fun watching this year and just enjoying the competition and surrounding celebration and coverage.


> I think he succeeded

… and hopefully continues to succeed. Humanity needs more celebrated and shining beacons of what we can accomplish together.


> But the project’s hefty price tag—an estimated $312 million over 6 years—drew immediate concern from several members of Congress, who asked GAO to conduct a review just 2 months after the initiative was announced.

Many AI and Crypto projects have gotten similar amounts. As a society we decide were to invest our money.

When that decision is removed from society and put in the hands of a few investors we get this result. Money for buzz-word pyramidal schemes while projects that would change the life of millions are stopped.

There is something broken in the economy right now.


I don't think the problem is with society, or with the economy. The money was already available for this project, and funding was essentially assured.

The problem is that, without putting too fine a point on it, the folks running the National Institute on Aging couldn't find their rear ends with both hands and a flashlight. There is a profound degree of stagnation in some of the National Institutes of Health, and the NIA is no exception, as exemplified by the fact that the current NIA director has been in that role for over 30 years. The leadership group puts out a list of goals for their "strategic directions" every 5 years or so, and to my knowledge, they've never actually achieved any of those goals.

As outlined in the report mentioned in the article [1], the NIA's approach to spearheading this large (for them) initiative basically involved identifying job opportunities, putting out a call for grant applications...and that was it. They didn't bother to articulate any actual vision for the project, nor develop any way to account for how the money would be spent. So at the end of the six-year program, there would have been 100 different projects, storing data in 200 different silos, $300 million spent, and zero to show for it.

In the end, NIA chose to can the initiative rather than open up scrutiny to how they spend the *rest* of their money (their yearly budget is over $4 billion [2]).

1: https://www.gao.gov/assets/gao-24-106886.pdf

2: https://www.nia.nih.gov/about/budget/fiscal-year-2024-budget


The success of a project isn't measured by how lofty the goals are multiplied by the budget it draws. The 'lack of basic project management practices' is pretty damning a statement for something which is expected to have hundreds of millions in budget.


$312 million sounds big, but it’s just $1 per US person.


We are throwing 10X that at the middle east and eastern europe on a weekly basis.


> Many AI and Crypto projects have gotten similar amounts. As a society we decide were to invest our money.

Biotech companies (including those tackling Alzheimer’s) have gotten much larger amounts from VCs.

The U.S. government also directly funds medical research to the tune of tens of billions of dollars annually.

I guess people must find a reason to bash investors.


> estimated $312 million over 6 years—drew immediate concern from several members of Congress

Napkin cost analysis:

• $8.50 per year per patient, or $0.02 per day per patient // via: https://www.wolframalpha.com/input?i=%24352%2C000%2C000+%2F+...

• $0.17 per year per American to fund it // via: https://www.wolframalpha.com/input?i=%24352%2C000%2C000+%2F+...

---

> As a society we decide where to invest our money.

It's hard to believe there's any American that doesn't know someone with a family member affected by Alzheimers.

Prevalence:

An estimated 6.9 million Americans age 65 and older are living with Alzheimer’s dementia in 2024. Seventy-three percent are age 75 or older (see Figure 2, page 23).

Of the total U.S. population:

About 1 in 9 people (10.9%) age 65 and older has Alzheimer’s dementia.

The percentage of people with Alzheimer’s dementia increases with age: 5.0% of people age 65 to 74, 13.2% of people age 75 to 84, and 33.4% of people age 85 and older have Alzheimer’s dementia.

People younger than 65 can also develop Alzheimer's dementia. Although prevalence studies of younger-onset dementia in the U.S. are limited, researchers believe about 110 of every 100,000 people age 30 to 64 years, or about 200,000 Americans in total, have younger-onset dementia.

Source: https://www.alz.org/media/Documents/alzheimers-facts-and-fig...


If you look for optimal you are going to spend more time looking for that textbook than learning.

The optimal solution is to find a good enough textbook and start as soon as possible to learn and tonstop procrastinating.


It's a common issues with self-learners, mathematics or not: there is no perfect course out there, and switching from courses to courses can be wasteful.

In my experience, focusing on a single, good-enough course (when in doubt, go for a famous/respected author/field contributor) and looking for other sources once in a while, has been the best approach.


Applies the same to job search too. Find one that is good enough and then work from there for future prospects. Often, the definition of “good” changes over time as priorities in life change.



We do this with all demanding endeavours. Circling around the tactical perimeter is easier than knuckling down and getting it done. But it's close enough that we can fool ourselves into thinking we are being productive.

Many examples:

- It's easier to research the "best textbook for me" than it is to study and do problems.

- It's easier to read about the optimal periodization cycle while sitting on the couch than it is to go sweat in gym.

- It's easier to read about dieting (it must be the best diet for meee!) than it is to just stop ordering pizza.

- It's easier to order business cards and redesign your logo than it is to find customers.

- It's easier to fiddle with your vim config than it is sit down and write code.

Unless you are already in top 10%, focusing on optimality is a distraction.


And if you pick up the wrong one, you might just end up dropping the whole ordeal. It's not so black and white, it makes sense to spend a bit of time and figuring out a good resource. At the least you'll get a sense of the domain's main trunk of knowledge, get into the jargon, etc.


If you pick up one knowing you can try a different one, putting it down isn't dangerous.


Yeah, this is it. A year ago if I tried to find the perfect textbook to learn Linear Algebra, I would still be looking.

There are certainly good and bad textbooks, and a book good for many people might be unsuitable for your style, your goals, and your background. But there are plenty of good enough textbooks, trudging through any of them will yield far more benefits than getting that ideal book.


If you're still looking, Gilbert Strang makes the best introduction book I know of: https://math.mit.edu/~gs/linearalgebra/ila6/indexila6.html


I like that he leaves determinants to a later chapter and doesn't _start_ with them, I never understood why they were useful or made sense. His view, represented on the cover, is great for learning


I don't understand the anti-determinant brigade. Many linear algebra books don't don't start with determinants.


They're fine where they are useful, I guess, but my undergrad put way too much emphasis on them when they're not intuitive, don't help (me) much with comprehension, and aren't useful in that many cases compared to the other techniques.


This seems a false dichotomy to me.

Surely the optimal solution would be to spend a few hours / days in the first week picking the textbook, then 51 weeks studying it, as opposed to literally picking the first one you see and studying it for 52 weeks.


> A year ago if I tried to find the perfect textbook to learn Linear Algebra, I would still be looking.

You know, there is a textbook for Linear Algebra that's literally titled "Linear Algebra Done Right". It's pretty much what it says on the tin.


And it is strongly discouraged as a first book, by the author himself!

https://linear.axler.net/

> This best-selling textbook for a second course in linear algebra is aimed at undergraduate math majors and graduate students.

> No prerequisites are assumed other than the usual demand for suitable mathematical maturity.


Equally there is also a text called "Linear Algebra Done Wrong"


Not equally, better. It's intended as a book for learning the concepts of Linear Algebra intuitively and with some introductory rigor, before doing it "right" in a professional way.


An important principle in learning


and it was not. No plane was flying with that software. It was booking services and similar needs. Planes could fly just fine, it was impossible to book people, thou.

My guess is that it's similar in this case. (Site is down)


Hospital networks and computers running Epic are very much indirect life-support systems, faulire of which can cause lots of injuries and deaths - as we're learning now in real-time.


I'm hearing that most hospital cybersecurity insurance requires Crowdstrike (or a product like it) on all the endpoints, so if that's true the liability might fall back on them. It will be a protracted argument for sure.


This is likely the case in a lot of places. We're still in the midst of ransomware groups targeting hospitals.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: