Apes

No argument against VMs - just that they have a different risk profile and a different set of trade-offs than containers. They're not a silver bullet, but if they're working for you, then go for it.

You can't really do anything useful with a VM either unless you start punching holes in those boundaries.

If you're connecting to a host on a port < 1024, then you know a SysAdmin must have set it up, and it must be trustworthy. It was a simpler time.

If you have write access to nginx.conf, you can set "user root root;"

The theory behind the US having a large military is that it acts as a sort of fleet in being - that the US prefers other methods of engaging with countries, and having a stronger military precludes other countries from engaging militarily. In turn, having stable global relations and protected global trade provides the US with a huge economic boon to fund its large military.

That's the theory anyway - our Idiot King and his idiots have completely missed the point of the US military existing and are using it as a primary method of engagement, which is causing the economic boon used to fund the military to evaporate.

As an aside, it's not a huge issue, but China's military costs use different accounting than the US, and seem lower by comparison. Apples to apples, China probably spends about half what the US does on military.

An MRI is great at identifying which ways your body doesn’t look like a textbook reference body, but it doesn’t necessarily tell you what those things are or whether they will ever cause you problems.

Its the doctors doing this, not the MRI.

There's this weird definition switch that always happens with the "overdiagnosis" defense where the information gets blamed for the overdiagnosis. An MRI doesn't provide any diagnosis in any sense of the definition. A doctor does.

Claiming an overdiagnosis defense is essentually implying the medical industry is worse for most than doing nothing.

I've got this game in my steam library, and I wondered why no one ever talks about it. I never realized it was delisted and made impossible to buy so long ago.

The year is 1846, and a doctor has a radical new idea: doctors should wash their hands between performing autopsies and delivering babies!

You're not sure of whether this is a good idea or not, so you ask various physicians, and the consensus is unanimous: the very suggestion is offensive, do you think doctors are unclean?

A clear conclusion has been achieved.

I strongly suspect the truth is both are "right", but they're both optimized answers to slightly different problems.

Mainstream medicine is hyper optimized for the most common 80% of cases. At a glance it makes sense: optimize for the common case. Theres some flaws in this logic though - the most common 80% also conveniently overlaps heavily with the easiest 80%. If most of the problems in that 80% solve themselves, then what actual value is provided by a medical system hyper focused on solving non-problems? The real value from the medical system isnt telling people "it's probably just a flu, let's just give it a few days and see" it's providing a diagnosis for a difficult to identify condition.

So if your question is "how do we maximize value and profit in aggregate for providing medical care to large groups of people", mainstream medicine is maybe a good answer.

But if your question is "how do we provide the best care to individual patients" then mainstream medicine has significant problems.

I don't think there's a silver bullet coming within our lifetimes.

There's no single points of failure: as you get older, everything just starts wearing out and failing.

If you cure heart disease and cancer, then others will just take their place: strokes, respiratory disease, Alzheimer's disease, falls.

And even if you do extend your lifespan, the reality is quality of life at 90+ is a lot worse than in your 20s or 30s.

Whoops, I haven't worked with bonds directly in a bit and forgot the coupon rate was split across payments.

I'm not sure there's a significant difference in practice, but technically the $50 would be part of the bond's debt principal, not interest.

A bond with a face value of $1000 means the government has $1000 of debt regardless of what is paid for the bond.

The coupon payments represent the "interest" on that debt - the $20 coupon means the government is paying $20 of interest per year.

Paying below face value doesn’t make the difference "interest." It simply means investors are buying the bond at a discount, so the government receives less cash upfront in exchange for repaying the full $1,000 at maturity. Bonds differ from traditional loans in that their market price can fluctuate, but the debt obligation remains fixed at the face value.

In practice, the government's accounting labels the discount as an "interest expense", so it still gets captured as interest in the budget.

I think you have a strong argument here, but there's a problem of deeper and widespread rot at play.

The reason Uber can get away with pretending it's just a "connector" is because the entire tech ecosystem has been allowed to normalize that kind of control without accountability.

Look at Apple and Google: they take a 30% cut on every sale and ban any competing payment systems. That's the same pattern - absolute gatekeeping disguised as "market facilitation."

Our regulators have become so complacent that this behavior is now seen as the default way digital markets work. The problem isn’t just Uber's misclassification; it's that the entire platform economy is built on pretending these companies are neutral middlemen when they're really gatekeepers.

Byuu was fantastic - I remember when they came onto the scene and made a big push for better emulator compatibility across the board. Byuu made positive changes in the world, and I still think about them from time to time.

It seems odd that you're fixated on the tool used rather than the content itself. Do you refuse to wear clothes because a machine made them?

I created the list and had an LLM review it to ensure I wasn't just pulling stuff out of thin air. Looks like it switched my hyphens for em-dashes.

People often get hyper-focused on a single cause, but there is a whole range of forces driving a sharp decline in jobs. All of this is visible, yet we are marching toward an economic crisis of our own making. I believe the best term for this is "gray rhino".

* Tariffs and Trade Uncertainty – Elevated tariffs and rapidly shifting trade policies are raising costs for manufacturers and discouraging hiring and investment (Investopedia; Atlanta Fed survey; CBO analyses).

* Automation and AI Displacement – Automation and AI, especially in low-skill occupations, are reducing new job creation and wages for some workers (academic studies in arXiv and PMC).

* Restrictive Immigration Policies – Tightened immigration and visa processes are straining labor supply, particularly in sectors that rely on immigrant workers (Axios, 2025 labor coverage).

* Small Business Strain from Economic Pressure – Tariff-related uncertainty is leading small businesses to slow hiring or lay off employees (Joint Economic Committee report, 2025).

* Offshoring and Outsourcing Trends – Technological advances are enabling offshoring and automation, substituting domestic labor with remote or machine-based alternatives (academic research in World Development, 2024).

People often get hyper-focused on a single cause, but there is a whole range of forces driving a sharp decline in jobs. All of this is visible, yet we are marching toward an economic crisis of our own making. I believe the best term for this is "gray rhino".

* Tariffs and Trade Uncertainty – Elevated tariffs and rapidly shifting trade policies are raising costs for manufacturers and discouraging hiring and investment (Investopedia; Atlanta Fed survey; CBO analyses).

* Automation and AI Displacement – Automation and AI, especially in low-skill occupations, are reducing new job creation and wages for some workers (academic studies in arXiv and PMC).

* Restrictive Immigration Policies – Tightened immigration and visa processes are straining labor supply, particularly in sectors that rely on immigrant workers (Axios, 2025 labor coverage).

* Small Business Strain from Economic Pressure – Tariff-related uncertainty is leading small businesses to slow hiring or lay off employees (Joint Economic Committee report, 2025).

* Offshoring and Outsourcing Trends – Technological advances are enabling offshoring and automation, substituting domestic labor with remote or machine-based alternatives (academic research in World Development, 2024).

RTX 5090 is about as good as it gets for home use. Its inference speeds are extremely fast.

The limiting factor is going to be the VRAM on the 5090, but nvidia intentionally makes trying to break the 32GB barrier extremely painful - they want companies to buy their $20,000 GPUs to run inference for larger models.

My understanding is the DGX Spark is optimized for training / fine tuning and the Jetson Thor is optimized for running inference.

Architecturally, the DGX Spark has a far better cache setup to feed the GPU, and offers NVLINK support.

The “employee” isn’t living off an H-1B salary — they’re already wealthy enough to bankroll the whole arrangement. The company is just a shell to win the auction and sponsor them. If an auction system were adopted without safeguards, it could turn the H-1B program from a labor-market tool into a plaything for the ultra-wealthy.

That's technically true, but I have seen PhDs hired under H1B - I've never heard of, let alone seen someone hired under the O-1A program.

It’s not hard to imagine a loophole here:

• Someone with a lot of money wants to get U.S. residency.

• They set up or work with a shell company that “hires” them under the H-1B program.

• That company uses part of their payment to win the visa auction.

• Once in the U.S., the “employee” does whatever they want — the job is just window dressing.

This is essentially what’s happened in other visa categories when money alone becomes the main filter.

It’s not even clear that an auction would be the most efficient way to allocate H-1Bs if we define efficiency as maximizing long-term economic and societal value, not just short-term revenue. An auction favors companies with deep pockets right now — meaning a startup looking to bring in a world-leading PhD in a critical field could lose out to a much larger firm simply filling headcount. That’s hardly an optimal outcome for innovation or competitiveness.

People want at least the appearance of a holistic process that considers a candidate’s broader value to society — even if that process isn’t perfect. An auction drops the pretense entirely and just says: pay to win.

They dropped 60 eggs each from three different heights (8, 9, and 10 millimeters)

Based on the photos, they measured this as the distance from the surface of the edge to the surface they were dropping it onto.

But for the vertical egg drop, the center of mass is several millimeters higher than for the horizontal drop, a pretty significant difference when you're only dropping 10mm at the most.

Maybe I'm missing something, but based on how they set up the experiment, maybe they're not measuring how resistant the egg is in certain positions, but instead just measuring that higher potential energy is more likely to break an egg?

It's very expensive, but Geisha varietal coffee might be of interest to you. It's extremely smooth, and tastes more like a light tea than a coffee. I've only had it once, and it was by far the best cup of coffee I've ever had.

It feels like "Democratic and open" in the US has become "shadowy oligarchy and kleptocracy run for the benefit of whoever can control the media propaganda machine controlling the masses"

I really liked Shoup's book. It contains a lot of practical advice what causes issues with parking, lays out bad designs that promote parking issues, talks through how to resolve these issues, and even covers how to sell it in a way that encourages people to buy into it.

San Francisco formally adopted demand-based pricing for curb parking in 2017.

The result was a rare win for San Francisco governance: We know from pilot areas that implementing demand-based pricing reduced congestion and parking citations while speeding up transit and increasing overall sales tax revenue. The program is now widely regarded as a model for parking management.

The SFpark program was by all accounts a huge success, and the areas where it was implemented feel a lot better off - you can actually find parking when you need it, and traffic isn't as bad there. I would love to see it put in place in the Sunset and Richmond - those areas are almost entirely free parking, and have the worst parking in the city.

[flagged]