wrs

I interpreted it as meaning that optimism is an assumption that problems will go away so there’s no need to painfully face them, whereas pessimism is an assumption that things will get worse unless you actively search out problems and fix them.

Insurance companies generally have regulated profit margins. Healthcare providers don’t. However, note that this is all percentages — if the total cost of healthcare goes up, providers and insurance companies both win in absolute dollar terms.

The Medicare “star system” attempts to compensate for this, but somewhat perversely by giving higher reimbursement rates to providers with higher quality. Logically, they should be getting lower rates because they’re more efficient at providing the service. But because the only real incentive is their profit margin, it has to be that way. And of course the star system definitions end up being a constant target of lobbying and stat-gaming.

Healthcare as a government service is one approach to the actual goal, which is healthcare that maximizes the quality of life and productivity of the citizens, rather than maximizing the profits of the providers.

When goods are discretionary and there are no externalities, it’s fine to have a purely autonomous market solution. Healthcare fits neither of those criteria. There’s a reason we don’t (or at least shouldn’t) have private subscription markets for baseline levels of law enforcement, fire prevention, public transit, food and drug quality enforcement, etc. etc. That same reasoning applies to healthcare.

Capitated reimbursement is a hybrid approach that helps a bit, but you can see from Medicare Advantage how the private sector relentlessly games the system because that’s how the incentive is inherently set up by the for-profit nature of the providers.

None of those invalidated the entire premise of the underlying technology, though. (That being: you can’t trust people, but you can trust code.)

Indeed, very shortly after live smart contracts were introduced, someone exploited a bug to steal a pile of pseudo-money, and the humans in charge decided to fork the blockchain to undo it. So the whole idea of “autonomous” smart contracts was revealed as the nonsense it always was. (Actually IIRC only some of the humans agreed to do that, so the chain was duplicated, and what does that even mean from a currency perspective?)

It’s also not just a “treat or not” decision. I have a family history and high test results so was already treating on the assumption I’m high-risk, but the scan revealed that we needed to be more aggressive. (And hopefully in a few years a scan will show we’re reversing the progression.) Personal data is always better than population data for treatment decisions.

Nobody likes to say this out loud, but because the US healthcare system is largely based on “fee for service” billing, it is true by definition that the sicker the patients are, the more profit the providers make. So incentives for actually making people healthier have to be imposed in indirect ways.

There are some pockets of “per capita” payment, where the provider gets a fixed amount per person and is thus encouraged to reduce treatment costs, which indirectly should encourage preventative care, but also encourages just refusing to treat people.

We could remove these weird constraints by making health care a government service rather than a highly contorted pseudo-market, but that would not be the American way. It has to at least look like a market, even if it’s an ineffective, inefficient, opaque, corrupt one.

Because they hired the wrong CEO for that?

The foundation has trustees, whose job is to ensure the goals of the foundation are being met, and they hire the CEO. In this case, one of the trustees is the CEO, which is a bit of a conflict of interest.

“Our legally binding purpose is to further the advancement of privacy, freedom, and democracy around the world. … We measure our return on investment not through financial metrics, but through impact.” (proton.me)

So many pitfalls here, but it’s great when it actually works.

In my experience, you can describe a culture that exists, which is useful for onboarding, but you can’t impose a culture just by writing down the one you wish you had. When the Netflix deck first came out I saw a lot of people posting “we should adopt this culture!” because they didn’t understand this.

What creates a culture is mostly just an accumulation of what behavior and speech gets rewarded and punished over time by peers and the power hierarchy. It’s good to consciously “garden” the culture when the group is small and there can be consensus on this, because it rapidly becomes uncontrollable as the group grows.

And then, especially in larger groups, the ones writing the culture down may not really be in touch with the culture, or if they are, may not want to write down the truth because it doesn’t sound good in public or just runs into the observer paradox when you say it out loud.

Moderna's business model was always to be an on-demand vaccine platform, meaning a cost-effective service for ordering a few doses of a fully custom personalized vaccine. Their infrastructure looks more like a just-in-time 3D printer service than like a traditional vaccine development lab. It was a happy accident that the platform was available when we needed to come up with a COVID vaccine rapidly, and that's what made them a household name (and wildly skewed their revenue mix from plan). But this cancer treatment is the sort of thing they had in mind all along.

I take it you stopped reading before the fifth paragraph.

It’s an equally valid (i.e., not valid) observation that most of the people with children obviously don’t give a damn, or we wouldn’t be here.

Actually, the last time I asked Claude Code about itself, it located and read its own minified source and told me something that wasn’t even in the docs.

The prosecutor is an employee of the executive branch, so use of that term is unfortunately not optional.

Turns out that’s what a Vectrex is! Or can be. As a pointless weekend project, I opened up my Vectrex, disconnected its logic board from its analog board, and gave it a new controller so an external computer can drive it. It doesn’t look as good, because the clever circuitry in the original to keep the beam velocity consistent lives on the digital board, but I think I can fix that in the controller firmware. Yet another post for my nonexistent blog.

>MAI-Code-1.1-Flash is available to download and run locally. For best performance, we recommend devices with more than 120GB of RAM.

I love the idea of promoting local models, but have you priced 128GB of RAM lately?

I’m a CMU SCS alum (actually I predate SCS being a full school). Please, someone make me feel better about this. They’re renaming the school after this person.

This conundrum (and many others) is why insurance is a highly regulated business. The commission working in the public interest decides how the compromise is drawn, not the insurers.

Mermaid is open source, no? Can we just add these placement constraints to Mermaid instead of making a whole new language?

They already bought the product, and it came with a license, then they wanted to unilaterally change the license. When they couldn’t, they declared the product a supply chain risk. The idea is you declare it a risk before you buy it.

Not to mention, they simultaneously said it was a supply chain risk and that it would continue to be used by agencies that were already using it in a critical capacity. Which makes absolutely no sense.

Or they could just not buy the product in the first place. The supply chain risk part is pure retribution.

But if they don’t jump at the chance to retroactively rewrite the already-signed license, you’re OK with designating them a supply-chain risk?

That’s going to make it rather risky for anyone to sign licenses with the DoD. In fact, it renders the licenses pointless since you just have to do anything the licensee says.

So in your mind there’s absolutely no subtextual message here in choosing wolves. It would have meant exactly the same thing if he’d chosen raccoons, or pigeons, or deer, or rabbits?

I'm talking about data not able to leave the building, not even data getting lost on the way.

One example was a site that had fiber and coax, from different companies. They might have shared a pipe at some point along their length, hard to say. But both connections went down at the same time for digital reasons, not physical. The providers had simultaneous unrelated backend router problems and the site lost contact to both gateways.

This was with a nice SD-WAN system that routed everything dynamically across both pipes to address latency and errors, but if the packets get dropped at the first hop in both networks, there's not much you can do!

The saving grace in that case was the third redundant connection, a 4G cell modem, with a lot less bandwidth but able to keep the critical transactions going. These days I'd certainly want Starlink on the roof as well.

> What has to go is the paper as the unit of achievement, the thing that gets counted and compared. If the ecosystem is the unit of research, then building something that other people can build on has to count for more than the next increment.

The focus on building stuff as the measure of accomplishment is one reason I so enjoyed being at CMU a few decades ago, so I'm very happy to see this sentiment is still expressed by the new generation of faculty.

What is the giant blinkenlight panel on the right?

I’ve had enough double-redundant links fail - and I mean pretty thoroughly double-redundant, different provider, different media, different physical path - that I’m quite surprised something like the air traffic control system only has two links.