mac-mc

Dude, disclose the AI writing; it has AI smells all over it, such as contrastive sentences.

No, I mean for some, a high saturated fat diet can do amazing wonders. And for others, it causes horrible issues. The studies are not well segmented genetically and by body state since that is signficantly more expensive and genetics only got cheap in the past 10 years or so, so they wash out these large sub-population dynamics.

Different countries are different, some are far more trigger happy about it like Canada. What you suggest as an alternative other than 'git gud' diet & exercise also changes it.

I've tried designing information-dense things for colorblind coworkers, and they seemed a bit disinterested in testing it out with me. Even with tools that simulate it, you can still be off, I've found.

There can be some sensitivity about trying to figure it out with them. I've added little affordances here and there, and ironically, I rely mentally more on color coding things because I am bad at finding things in a visual field than most.

I've also found that colorblind family members and friends just never tell you and they tend to suffer in silence. Even my own half-brother (which I have a 15 year gap with) didn't tell me he was colorblind until recently.

17 years is far from rapid or move fast and break things. ApoB has been known about for quite a long time, since the 90s its effects have been obvious, and showed up in research in the 70s-80s!!! It's still not part of standard testing!!!

Guidelines also leads to standards of care being random and heavily driven by politics & financial reasons disguised as medical best practice. South Korea and India are "parallel testing" places, which saves time, while the USA & others are serial testing places mostly because of their funding models.

Talk to any American doctor and they will give you a bunch of emotionally wrapped cope about why it's bad because the cognitive dissonance sucks and there are liability reasons to avoid admitting your wrong. I would argue that in many cases, parallel testing is cheaper because $300 of tests is cheaper than 4 chained $500 doctor visits. But whatever.

If you dig into research and follow the low-risk experiments that people do online to reduce their Lp(a), you can find techniques and evidence to do so. It doesn't have to be an impossible-to-fix issue.

I like this list of experiments by Greg Muschen: https://x.com/gregmushen/status/1924676651268653474

If you fix it without statins through better lifestyle and diet, that is the preferrable route.

As to why medicine is like this, it's because it's conservative, usually about 17 years behind university research[0], and doctors are shackled to guidelines in most health systems or risk losing their licenses. It isn't a coincidence that the article author had his out-of-pocket concierge doctor tell him the more up-to-date stuff.

[0] https://pmc.ncbi.nlm.nih.gov/articles/PMC3241518/

IMO, I think that is more of a saturated fat issue, and only a subset of the population is like that. Others solve their health issues through eating a lot of red meat.

It's a fairly useful tool if you know how to use it. People will also play with it as a toy. It's much like the masses getting access to cheap video cameras and smartphones with good cameras. It's going to enable different content, it's not going to make more hollywood movies. This is an early example of what people will make: https://www.youtube.com/watch?v=jBwluRXtS2U . It's just one person making all of this on the side.

Yes, much like wet streets are a high predictor of rain. Or smoke, firefighters and wood are a high predictor of fire. The firefighters are not causing the fire, neither do the wet streets cause rain. This is what people are trying to tell you. If you remove the firefighters only, then you might make it worse. If you do something to cause the firefighters to go away, probably because there isn't a fire anymore, then you did the right thing. The important thing is not to goodhart's law yourself into doing the wrong thing.

Don't you sign something to the effect that you have not hidden testing like this that you are aware of?

Cholesterol is more of a proxy "smoke" or "firefighter" measure than a measurement of the actual fire. It's very much a wet streets cause rain kind of thing.

Artificially eliminating the firefighters doesn't necessarily mean you've solved most of the problem.

Heart disease is a far more complicated problem than "cholesterol" or "cholesterol + inflammation", but humans and patients mentally gravitate to silver bullet thinking, which makes it really hard to work with. One interesting measure I've encountered is the lipid clearance rate, but it costs something like +$20k to measure and is not something a doctor can order from a lab; it's typically only performed in research settings.

It is ultimately a form of insecurity.

ok what is DMV, RTP is research triangle park from what I can see. DMV is unsearchable.

I was the manager in this case, and I also hated JIRA with a passion. It's often the managers doing the alternative spreadsheets too and only using JIRA as necessary. I found you didn't need to "hound" developers with Phabricator and part of the hounding was me being hounded further up about it. Tools matter! Developers love automated organization!

Yeah, burnout is hard, and real. Everything you've described, I've literally been in an equivalent. To get out of these ruts requires a sober assessment of where you are and some hard decisions, and maybe the end result is acceptance for someone in your case. As I said, it's belief that drives this, and you do have ultimate control & choice over your beliefs.

For many tho, what I said is a revelation and it can save them from yet another ineffective pomodoro exercise. Many are not middle aged people stuck in a hard place and this advice helps them.

JIRA is so dysfunctional in many places that people do a good chunk of intra-team planning in spreadsheets and docs instead, and use JIRA sparingly to make everyone faster.

I saw this effect live at my previous big tech after they moved to JIRA. JIRA got used way less than Phabricator because of all the friction it introduced and a lot more informal google docs + slack bot usage increased instead.

I remember to this day asking a report to plan more stuff in JIRA and seeing a beautiful task tree in Phabricator they did in the past. I asked why, and he shrugged and said it was just easier. That's when it really clicked for me. Linear can't come soon enough and burn JIRA to the ground.

IMO you need all types. A well-functioning team has people with different strengths that can get pathological when they go too far, and they cover for each other's strengths. The "just do it" high work ethic guys like that are great, but they can sometimes "just do it" in the wrong direction and don't stand back and ask if we should even do these things. "Mr. strategic" can sometimes stand back a bit too much and overthink things, etc.

Heh, yes it was inspired by that game once I saw it.

Procrastination is not a time management problem; it's an emotional management problem.

What it comes down to is not having belief in what you do, so you do other things. You might feel trapped, so you pass the time with stuff like YouTube because that is the most compelling thing available to you. A man will walk on broken glass with a smile if he truly believes in what it will accomplish.

When I was younger, I was into video games because they gave me a sense of accomplishment and progress compared to high school, which I found relatively meaningless. I called it progress quest.

When there is a rare game or youtube topic that really obsessively catches your attention, like Factorio, pay attention to it! It helps show you what drives you, and you can try to leverage that into things you find healthier.

Also, it might be worth it to look into ADHD testing if this has been a persistent pattern your entire life.

I tend to reject full on large homework assignments like that unless I get a guarantee of feedback.

Ironically, they turn on luck even more.

There isn't, but you can also sometimes deduce things you didn't do as well if they were apparent. Like, if you couldn't answer a bunch of questions or finish the task on time, and all the other interviews were good, there is a good chance it was that thing, and you should brush up on it.

Sometimes interviews are designed to be a hard grind that everyone fails and it's based on 'did you fail the least', but those are rare and once you've done a bunch of interviews, you can tell the difference.

If your half way through a process, recruiters are often ok to tell you what you didn't do as well on and offer that as feedback and tips. At the end they tend to be quiet so you can't figure out the final reason for a rejection unless you somehow have a friend on the inside that can find out informally for you. If your lucky to get a rejection call, you can even get vague hints as to why it wasn't a go from the recruiter if you have the social skills and the right recruiter. I often play guessing games based on my deductions.

But these are more obvious parts.

There are entire online social groups on discord of teens encouraging suicidal behavior with each other because of all the typical teen reasons. This stuff has existed for a while, but now it's AI flavored.

IMO I think AI companies do have the ability out of all of them to actually strike the balance right because you can actually make separate models to evaluate 'suicide encouragement' and other obvious red flags and start pushing in refusals or prompt injection. In communication mediums like discord and such, it's a much harder moderation problem.

I found I couldn't listen to music with words in them while coding. Maybe try wordless music, or music that is in a language you don't understand.

It's pretty obvious, it's costly to make one that is up to the level of quality of commercial ones. It's not a mistake that the 2 mobile oses are owned and created by some of the largest and most profitable companies in the world.

IMO, looking at most budget spend, it's the art & content that is the bottleneck.

Like a full sprite sheet, one sprite at a time or a sprite animation loop?

Each one would require a different kind of model and model technique to make, so I wouldn't be surprised that ChatGPT has issues with it. A sprite animation loop would be better done by a potentially specialized video-oriented model, for example, and the current image and video models are barely trained on that kind of video data.

Others would quibble that those are integration tests, "UI" tests, or other higher-level tests, etc.

You see this dynamic even with swift which has a corpus of OSS source code out there, but not nearly as much as js or python and so has always been behind those languages.