I actually just responded with almost the exact opposite, but maybe I'm the "lipid hypothesis skeptic".

Seeing as the threat is calcium build-up in the arteries, and because cholesterol is a vital component of health, I believe that if you are in good health, and don't have a history of heart-disease, or have diabetes or other auto-immune disease which increases risk of atherosclerosis, lowering cholesterols is an in direct measure.

It's about understanding your personal risk and making decisions based on that.

I actually don't think your response is the exact opposite, but you touch on some of the skeptic stuff, so I'll respond to here:

First of all, I agree with your points that you should consider the individual. My long term interest in this is also from being a very fit, low blood pressure, metabolically healthy person who always had at least somewhat elevated LDL (sometimes very elevated) that doctors would flag.

PCSK9 people are as close to a natural experiment on the effects of life time low LDL as you will get and they get near total protection, even when they have no other risk factors. People like smokers, hypertensives and diabetes have ~90% less than other high risk people, but people without any of those factors also have significantly less heart disease. People with two broken PCSK9 genes have close to zero LDL and have noticeably completely plaque free arteries as adults. I do think this does pretty fatal damage to the theory that you must have some other health issue for LDL to be bad.

It's very likely that "LDL-C" the lab measurement isn't as good as measuring ApoB, but for most people, they are concordant. And ApoB is a different way of looking at low density lipids, by particle count instead of weight. Dietary stuff like the fats in the article that lowers LDL measurements typically also lowers ApoB in most people.

So, in part, I agree that more precise biomarkers can help adjust individual risk. But most people are concordant. And the evidence that the underlying "low density lipids", no matter how you measure them, are causally part of the disease process is very strong.

There is push back on the high LDL leads to cacium build-up in arteries (being specific instead of just "longevity").

The challenge is that some people, like myself, have outrageously high LDL, yet no calcium build-up in arteries via calcium score testing.

This is why ApoB is the newer more common test. Molecules containing ApoB can stick to the walls of arteries, and the theory is that the more ApoB molecules in the blood, the more likelihood of a molecule sticking and then becoming calcified.

If you have an auto-immune disease or diabetes, which increases the amount of time it takes for tears in the arteries to repair, you have a longer exposure time of fats sticking to the arterial walls which increases the likleihood of calcifying.

However, that doesn't mean that everyone with high LDL and high ApoB are at higher risk.

This is why I went for the calcium score. Don't show me the things that might lead to calcium build up later, just measure my calcium levels and let's see if I am currently at risk, and we can keep monitoring this.

Doctors are still trying to push me on to statins, but without a history of heart-disease and every other biomarker being off the charts high, I am taking that risk myself, knowing that calcium score is 0, so my suspected risk is actually very low.

Long answer, but hopefully that clarifies the understanding.

I'm not a doctor, I work in neurotech so am around health and have gone through this process myself.

I'm with the sceptics, but also they don't show it in use. But from the product screenshot of the person hitting the button, it seems not to be wearable, so.....when would I use this? When I'm in a room that has a smart-speaker? So I go to a friends place for dinner and I put this device on the table so that my friends Alexa can't hear me.

Good thing I wore my tin-foil hat to dinner, but sadly, my friend didn't wear one, and now they can't get a timer for their cooking and the meal is ruined. Brilliant.

I think the thin-client/flat-client is a pendulum that swings every few years.

Main-frame (thin) -> PC (fat) -> Internet/Cloud (thin) -> Mobile (fat) -> AI (thin)

I expect this to continue until the next technology transition.

In each of these shifts, and there have been others, things are not completely fat or thin, more of an in-between state but leaning to local vs cloud.

Of all the things wrong with windows, I don't feel that having a Microsoft account is the worst of them, or the one that needs the most attention.

Is the reasoning that if you don't have a Microsoft account, they'll do less of the ads nonsense which is baked into the OS? Maybe I don't get what the issue is.

I've tried linux, but haven't converted (though I'm tempted), but my mac has a mac account and nobody seems to complain about that, my android has a google account, why is a Microsoft account so much worse.

The things I feel it is more important for Microsoft to get rid of 1) the push for OneDrive everywhere - Mac is as bad if you don't have iCloud 2) updates requiring "set-up" and trying to trick you into adding services in the process 3) Windows Hello moving the "sign in" button down 10px once it recognizes you....WTF!! 4) ads, ads, ads (though if you don't use start button much or Edge, I think this is mostly avoided 5) letting apps add shortcuts to your desktop on an update.

What am I not understanding?

"Requiring an account" means leaning in the Apple/Disney direction - upselling, walled gardens and milking captive customers. Dropping the account requirement is a symbolic (and practical!) step in the other direction, towards using the product as a tool.

That's fair, so is the outrage focused on Microsoft because they are a "more" open system without the app store being the default method of installing apps?

I guess I'm wondering why Apple and Google don't get the same pushback?

You don't need and apple account to use macos

I was wondering about this, as this is how I thought it works, but is memory reducing the context window rather than loading in so much each time?

Some people are comfortable living on the edge, but it isn't for most.

I was developing in the metaverse space, and the problems we were facing led me to learn about the state of AI image generation (2018), and where the world was headed.

People assume the thing you are focused on is the thing that must win in the end, but that just means you are too focused on your little part of the world to take in the bigger picture of what's out there.

Prior to working in the metaverse (really a form of volumetric video, but I won't go into details), I was working in telehealth (2014). I did some research in augmented reality (2009), and lots of other areas of interest as well.

Some people would say I wasted my time on these, but there was a mass of secondary learnings which I value every day.

Are these sorts of general advice on how to do X even valuable today when you can put the details of your start-up into AI and get a more customized and moderately more thoughtful actions based on what your start-up does, who your customers are, etc?

Who's still going through these kinds of docs?

I know micro.so (I'm not affiliated with them) have documented how to build agentic B2B sales AI that you can download (if you give them your email address). https://www.micro.so/guides/sales

I have a friend who was working in this space in 2019.

Their customers were hiring something like 10k jobs worldwide annually, which means 500k+ applications to go through.

AI was used for the first filter to get a person through to later rounds.

It makes sense at that scale, and not for "hiring" but just to make decisions as to who gets to the next round.

The alternative is that you end up having to hire so many people to go through the applicants and then those people get bored of asking the same initial questions again and again.

I remember hearing an anecdote, back in the days of paper resumes, that hiring managers would take the huge stack of resumes they got, divide them in half and throw half in the bin. That half would be considered unlucky, and you don't want to hire unlucky people.

But seriously, with the number of job applicants, for certain positions, what are the alternatives to getting AI to help?

If you have 10k jobs and 500k applications, that's 50 applications per job. Spending a few minutes on each application would mean a few hours for the initial screening for each job, but if that job is filled, you'll get a person who'll stay there for a few years. How is that not worth it? Why would you need to automate it? If you're interviewing for people to do some quick task and leave, sure, but companies want long-term hires.

I probably should have worded that better. 500k applicants for 10k positions. The jobs aren't different. Think Amazon Warehouse job where there are 650k workers. High turnover, and an HR department is running those interviews, not the manager.

So even if each screener is running 15 minute interviews, they're asking the same questions 20 times a day. Every day. The mental task and repetitiveness just isn't something a person is going to be good at. An AI can do this more effectively and pass on the top candidates.

Realistically there's a point where you haven't interviewed enough people and another when you've interviewed so many you're wasting time.

Do you need the global optimum candidate, or do you need a very good candidate? If you need the global best then you're probably better off headhunting than posting a job listing.