Anyone know what's going on with the ACA marketplace?
I like to take a peek at it every so often and it's just stupendously worse than employer healthcare. There is no plan in my market (Idaho) which doesn't have extreme out of network deductibles. The cost is also identical to what I and my employer pay for insurance.
Is it just that the ACA is mostly used by sick people or something?
For one thing, as I understand it, the ACA was pretty effectively sabotaged by removing the mandate, which destroys the math that makes insurance work.
Because pre-existing conditions have to be covered, you're free to wait until you have a serious (expensive) condition, then sign up for an ACA plan. And there's no mandate that spreads that risk around to healthy people, so the population is severely skewed.
I have no definitive data on this, but it seems self-evident that the system can't work well.
this is mostly right, but it's not true that you can just sign up at any time. there's an open enrollment period for the aca marketplace and if you miss it, you won't have the opportunity to buy health insurance until next year.
The individual mandate merely guaranteed a captive audience for health insurance companies. That doesn't do anything for cost. What drives cost is the ACA limiting profits based on the cost of care, thus more expensive care equals more profits. Can you see where this is going?
Five states plus DC still have an individual mandate (ugh), and yes health insurance premiums are spiking there too. A large part of that is due to the tax credits (generously labeled subsidies) expiring – even with an individual mandate health insurance is prohibitively expensive.
Even with the individual mandate, it seems self evident that incentives are for private insurance companies to happily take in the extra premiums from healthy people, and then drop or have prohibitively high premiums after people get sick. The medical loss ratio tries to mitigate this, but just kicks the can down the road where the incentive becomes to dump the money into administrative overhead and self-dealing with providers/pharmacies run by affiliated companies.
As a solo dev, I’ve been on ACA with my family for the past four years. In my market, a major insurance company is not returning for 2026 and the prices have gone bananas. Our premiums are increasing over 40%, all of our copays are more than doubling, and the deductible is astronomical. In many cases the cash pay price is probably cheaper than the insurance negotiated rate.
I’m the plus side, all bronze plans can contribute to HSAs next year.
We compared individual plans off market, but they’re mostly the same without HSA access or they’re targeting people with specific ailments (e.g. the diabetes plan).
On top of that, consolidation of hospital systems has made wait times for GPs months unless you’re lucky enough to find a cancelation. We have some older Canadian ex-pat friends and it seems like our health system has all of the downsides of socialized systems, plus we get to pay Cadillac prices.
I’m really hoping 2026 is so bad that something breaks enough to resolve some fundamental issues with private health costs, but I’m not naive enough to think it’s likely. ACA was such a monumental gift to someone, but it definitely not individuals who need family coverage.
It's because health insurance companies need to pad their profits. Because we have a backward system in the US where profits are extracted by everyone without any competition from a not-for-profit. That's why the idea of a public option is so popular (it was originally a Republican idea until it was decided insurance company profits are more desirable). Personally, I don't think any health insurance company should be for-profit. But the for-profit companies sure as hell should have a non-profit competition with the same bargaining power they do.
Most (and the biggest) BCBS licensees are for-profit. Kaiser is not but is still eye wateringly expensive. For me, a silver plan will be about $700-800 next year. And then you take a look at what a bronze plan covers…
This is correct. BCBS is for profit and most others are. Kaiser Foundation Health Plan is one of the few not for profit insurances and only available in 8 states.
Yeah, some individual states do have not for profit, and it does help. These are the vast minority of states. That's why it should be available nationwide as a public option.
The acid test is if when presented with new evidence, someone is willing to change their opinion or if they simply retreat and double down on their skepticism.
It can be really hard to abandon a false belief. Especially if you take the time and effort to shore it up with bad data. It starts with recognizing that everyone, including yourself, is wrong about something.
> The acid test is if when presented with new evidence, someone is willing to change their opinion or if they simply retreat and double down on their skepticism.
My opinion is that on the specific topic at hand of COVID-19 vaccine "scepticism", we are far past the point of rational scepticism being sustainable.
It's solicited advertising. Something I don't think almost anyone has a problem with.
Unsolicited advertising is what everyone hates.
If I go onto my grocery store website and see "we have a sale on xyz" I'm not bothered because I went to that website to see what they have. I'm also not bothered by sales displays in the store. All forms of acceptable advertising.
But what I absolutely hate is navigating a webpage unrelated to my store and seeing "Did you know you can buy widgets at your local store!" or watching youtube and seeing an unskippable 30 second ad for my store. Or getting a newspaper that is actually just 90% advertisement with 2 paragraphs of actual news.
I see - thanks for the explanation. I try to filter out those sorts of ads too, because I don't want my decisions to be biased by the money someone else spends, but they certainly are less annoying than the usual sort.
I wonder whether you would consider ads for fashion houses in a fashion magazine to be "solicited" or "unsolicited"?
I was born in '85 and got chickenpox at 7. My parents aren't quiet antivaxxers but they also have a sort of "you don't need all vaccines" attitude (so no way I was getting the new chickenpox vaccine).
I've recently had shingles and, let me tell you, it sucks ass. I thought my liver was going to explode.
I'll be getting the shingles vaccine as soon as possible.
For a lot of people these aren't rational beliefs, they're beliefs based on appeals to emotion. They will only rationally re-evaluate those beliefs if you change the kind of media they consume.
If the results showed that mRNA vaccines had negative health outcomes, then the obvious next question to ask is "are they worse or better than COVID's health outcomes?". If they are better then yeah, I'll still say take the shot. If the negative outcomes only occur in certain demographics, then I'd say they should limit their exposure to the shot.
The most common skeptic position that I've seen (which admittedly isn't all of them) is that the shots should be banned altogether until they can be proven 100% safe for everyone. Very similar to the general vaccine skeptic position. It ends up being a moving goalpost as well.
A truth seeking individual realizes that very few things in the world are black and white. They avoid trying to frame things as a black and white. Nobel and villainous framing. If you are truth seeking, you won't try to turn a non-binary evidence into binary thinking.
The most common skeptic position that I've seen (which admittedly isn't all of them) is that the shots should be banned altogether until they can be proven 100% safe for everyone.
That's not what I've seen. I live in very-red Tennessee. What I see is more like what you said yourself:
If they are better then yeah, I'll still say take the shot. If the negative outcomes only occur in certain demographics, then I'd say they should limit their exposure to the shot.
The conclusion to this (within my bubble) being: since covid risk to young children is negligible, why the heck are you requiring them to get a shot?
It's common enough that there are state legislatures trying to ban mRNA vaccines all together. [1]
> since covid risk to young children is negligible, why the heck are you requiring them to get a shot?
Kids spread covid and no vaccine is 100% effective.
I've not seen any evidence that the vaccine is dangerous to any age group. There is plenty of evidence that COVID is deadly to the very young, the very old, and a bunch of other people (including those with compromised immune systems).
It's exactly the same reason kids should get the flu shot.
That said, there's no requirement anywhere for kids to get either.
Well, I'd say that the standard movie format just isn't what people want anymore.
The problem movies have is they have a relatively short amount of time to deliver a complete story. 90 to 120 minutes just isn't a lot of time to be compelling. That's why some of the best movies are split into parts.
Consider Andor as an example. It's some of the best media ever made (IMO) and it simply would not work in the movie format. What makes Andor work is the excellent character development and the time spent building and shaping the universe under a fascist government.
Andor had no length constraints per episode. That allowed it to tell complete satisfying stories with the promise that you'll get more in the next episode.
Telling a detailed story is different than telling a compelling story.
Andor isn’t as compelling as the original movie or significantly longer than the Harry Potter series of movies. Babylon 5 is probably the poster child for a long running space opera series with a planned story arch, but they added plenty of filler because you don’t actually need that much time.
If anything movies tend to be better than TV shows because of the time constraints rather than the budget.
Eh, the current 10-hour seasons are the worst of both worlds.
Telling a story in a "tight 90" means making very deliberate choices about what to include, what not to, and how to make scenes do double duty. Having 23 episodes a season lets you slow down, spend time with the characters that's not all focused on the season plot, it lets you have B-stories in every episode. A 10-hour season doesn't get to do that, but it doesn't enforce the same discipline as 90-120 minutes.
Compare Star Trek: Deep Space Nine to Star Trek: Discovery or Star Trek: Strange New Worlds. I greatly enjoy SNW, but the characters and their relationships with each other are in no way as substantial as in DS9 (or even TNG, which was much less character-focused than DS9).
AV1 was specifically designed to be friendly for a hardware decoder and that decision makes it friendly to software decoding. This happened because AOMedia got hardware manufacturers on the board pretty early on and took their feedback seriously.
VP8/9 took a long time to get decent hardware decoding and part of the reason for that was because the stream was more complex than the AV1 stream.
Hmmm disagree on your chain there. Plenty of easy hardware algorithms are hard for software. For example, in hardware (including FPGAs), bit movement/shuffling is borderline trivial if it's constant, while in software you have to shift and mask and or over and over. In hardware you literally just switch which wire is connected to what on the next stage. Same for weird bit widths. Hardware doesn't care (too much) if you're operating on 9 bit quantities or 33 or 65. Software isn't that granular and often you'll double your storage and waste a bunch.
I think they certainly go hand in hand in that algorithms relatively easier for software vs previously are easier for hardware vs previously and vice versa, but they are good at different things.
I'm not claiming that software will be more efficient. I'm claiming that things that make it easy to go fast in hardware make it easy to go fast in software.
Bit masking/shifting is certainly more expensive in software, but it's also about the cheapest software operation. In most cases it's a single cycle transform. In the best cases, it's something that can be done with some type of SIMD instruction. And in even better cases, it's a repeated operation which can be distributed across the array of GPU vector processors.
What kills both hardware and software performance is data dependency and conditional logic. That's the sort of thing that was limited in the AV1 stream.
> Bit masking/shifting is certainly more expensive in software, but it's also about the cheapest software operation. In most cases it's a single cycle transform.
He's not talking about simple bit shifts. Imagine if you had to swap every other bit of a value. In hardware that's completely free; just change which wires you connect to. In software it takes several instructions. The 65 bit example is good too. In hardware it makes basically no difference to go from 64 bits to 65 bits. In software it is significantly more complete - it can more than double computation time.
I think where software has the advantage is sheer complexity. It's harder to design and verify complex algorithms in hardware than it is in software, so you need to keep things fairly simple. The design of even state-of-the-art CPUs is surprisingly simple; a cycle accurate model might only be a few tens of thousands of lines of code.
Right. It's bit packing and unpacking. Currently dealing with a 32 bit system that needs to pack 8 11 bit quantities each subsisting of 3 multi bit values into a 96 bit word. As you can imagine, the assembly is a mess of bit manipulation and it takes forever. Ridiculously it's to talk to a core that extracts them effortlessly. I'm seriously considering writing an accelerator to do this for me
> I'm not claiming that software will be more efficient. I'm claiming that things that make it easy to go fast in hardware make it easy to go fast in software.
I don't think you have an accurate view on what makes an algorithm slow.
The actual constraints on what makes hardware or software slow are remarkably similar. It's not ultimately the transforms on the data which slow down software, it's when you inject conditional logic or data loads. The same is true for hardware.
The only added constraint software has is a limited number of registers to operate on. That can cause software to put more pressure on memory than hardware does. But otherwise, similar algorithms accomplishing the same task will have similar performance characteristics.
Your example of the bitshift is a good illustration of that. Yes, in hardware it's free. And in software it's 3 operations which is pretty close to free. Both will spend far more time waiting on main memory to load up the data for the masking than they will spend doing the actual bit shuffling. The constraint on the software is you are burning maybe 3 extra registers. That might get worse if you have no registers to spare forcing you to constantly load and store.
This is the reason SMT has become ubiquitous on x86 platforms. Because CPUs spend so much time waiting on data to arrive that we can make them do useful work while we wait for those cache lines to fill up.
Saying "hardware can do this for free" is an accurate statement, but you are missing the 80/20 of the performance. Yes, it can do something subcycle that costs software 3 cycles to perform. Both will wait for 1000 cycles while the data is loaded up from main memory. A fast video codec that is easy to decode with hardware gets there by limiting the amount of dataloads that need to happen to calculates a given frame. It does that by avoiding wonky frame transformations. By preferring compression which uses data-points in close memory proximity.
It was a presentation on AV1 before it was released. I'll see if I can find it but I'm not holding my breath. It's mostly coming from my own recollection.
Ok, I don't think I'll find it. I think I'm mostly just regurgitating what I remember watching at one of the research symposiums. IDK which one it was unfortunately [1]
I've heard that same anecdote before, that hardware decoding was front of mind. Doesn't mean that you (we) are right, but at least if you're hallucinating it's not just you.
The entire point of diagnosis is because a person needs help.
The point of treatment isn't to "fix" or "make normal" someone. It's to give them the tools needed to participate.
For example, someone with autism might be more prone to having a meltdown. What therapy does is give them the tools to both identify that they are on the verge of such a meltdown and to de-escalate themselves.
The point of treatment is to help someone take care of themselves, hold a job, and generally not burden anyone. It's also to help a person feel better about themselves.
For OOME problems I use a heap dump and eclipse memory analysis tool.
For microbenchmarks, I use JMH. But I tend to try and avoid doing those.
reply