The calories in/out model isn't wrong exactly, but it's so reductionist that it becomes misleading in practice. It omits hormonal responses (insulin, leptin, ghrelin), the thermic effect differences between macronutrients (your body burns 20-30% of protein calories just processing them vs 0-5% for fat), gut microbiome composition, sleep quality, stress hormones, meal timing, and individual metabolic variation. Two people eating identical calorie counts can have very different outcomes. Telling someone "just eat less and accept the hunger" without any of that context can set them up for a miserable yo-yo cycle - or worse, a disordered relationship with food.

Sure - how did you arrive at the 800 kcal figure? Most likely a wearable or an app, and those estimates are based on rough linear regressions from weight, age, sex, and heart rate - not actual calorimetry. The error margins on those numbers are significant, but the devices present them with false precision that makes people treat them as ground truth.

Even setting accuracy aside, the framing is the issue. Your basal metabolic rate - just keeping your organs running, blood circulating, cells maintained - accounts for 60-70% of your total daily energy expenditure. Add the thermic effect of food (~10%) and you're at roughly 70-80% of your energy budget spent before you even lace up your shoes [1]. Exercise typically makes up the remaining 20-30%. So that hour of running, while genuinely beneficial for a hundred other reasons, is a relatively small slice of your total daily burn. And not all calories are equal on the intake side either - your body spends 20-30% of the energy in protein just to digest it, compared to 0-5% for fat, so "800 kcal burned = 800 kcal of anything eaten" doesn't hold up.

That's what I mean by "myth" - not that exercise burns zero calories, but that the popular mental model of "I ran for an hour so I earned X calories of food" is built on inaccurate measurements, treats all calories as interchangeable, and overweights exercise relative to what your body spends just existing. Curious though - do you track your intake with the same rigour, and if so, do you find the numbers actually add up in practice?

[1]: https://my.clevelandclinic.org/health/body/basal-metabolic-r...

Right, and that's kind of my point - the "2000 kcal" figure is itself part of the problem. It's a rough global average that doesn't account for your sex, age, weight, body composition, activity level, or even climate. It's a number on a food label, not a physiological reality for any specific person.

And even if you could nail down your actual total daily energy expenditure, calorie counting treats all calories as equal, which they aren't. Your body spends 20-30% of the energy in protein just to digest and metabolize it, compared to 5-10% for carbohydrates and 0-5% for fat. So 100 kcal of chicken breast and 100 kcal of butter are not metabolically equivalent - your body nets significantly less usable energy from the protein. This is the thermic effect of food, and it alone accounts for roughly 10% of your total daily energy expenditure.

Speaking of which - basal metabolic rate (just keeping your organs running, blood circulating, cells maintained) accounts for about 60-70% of total daily energy expenditure. Add the thermic effect of food on top and you're at roughly 70-80% of your energy budget spent before you even stand up from bed [1]. Physical activity - including your 100km rides - typically makes up the remaining 20-30%, though obviously that range is wide and shifts dramatically for endurance athletes.

So yes, of course people who cycle 100km need more fuel. Nobody is disputing that. My point is that most people vastly overestimate how many calories exercise burns relative to what their body spends just existing, and they use kcal as a universal unit of nutritional value when the body's actual energy extraction varies significantly by macronutrient composition. People optimizing purely on calorie numbers are working with a model that's far rougher than they think.

And this whole picture gets worse with wearables pushing calorie counts front and center. You see it all the time - "my wearable shows that i burned 300 kcal with that activity so it zero outs that sweet thing I ate earlier that was also 300 kcal" That's wrong on multiple levels - the device estimate is inaccurate to begin with, the thermic processing of that pastry isn't equivalent to the "300 kcal" on its label, and your body doesn't do neat arithmetic like that anyway. But with every fitness app and smartwatch plastering a big kcal number on your workout summary, it's becoming the default way people think about food and exercise, and it's reinforcing exactly the wrong mental model.

[1]: https://my.clevelandclinic.org/health/body/basal-metabolic-r...

And also that the calorimetry from wearables is highly flawed and it seems to that we don't have super accurate data and what sort of activities burn the most energy.

I am also a big opponent of folks that start equating the "my wearable shows that i burned 300 kcal with that activity so it zero outs that sweet thing I ate earlier that was also 300 kcal" which is wrong on so many reasons but with a lot of workout apps and devices pushing the (inaccurate) kcal count front and center becomes more and more a of a thing.

The big copy on the front page says:

> Your Apple Watch *tracks* VO2 Max—one...

While you have to read through FAQ where you see:

> The watch *estimates* your cardio fitness during outdoor activities and stores it in Apple Health, which our app reads automatically.

All emphasis are mine.

I think it's a little disingenuous to sell this as "Your VO2 Max, finally visible" when it's actually just an estimate from a watch, based on biomarkers. When the real VO2 is measured in a lab with a more involved equipment.

A 2025 validation study involving 30 participants found that Apple Watch underestimated VO2 max by a mean of 6.07 mL/kg/min (95% CI 3.77–8.38) when compared to indirect calorimetry, the gold standard method. The mean absolute percentage error (MAPE) was 13.31%, and the limits of agreement showed considerable variability ranging from -6.11 to 18.26 mL/kg/min [1]. Another 2024 study found similar results, with the Apple Watch Series 7 showing a MAPE of 15.79% and poor reliability (ICC = 0.47) [2].

[1]: https://journals.plos.org/plosone/article?id=10.1371%2Fjourn...

[2]: https://biomedeng.jmir.org/2024/1/e59459

Reading this article I'm a little confused by the author's conflation of brain energy and the energy expenditure of the body as a whole. In the beginning they mention:

> "Your brain consumes roughly 20 to 25% of your body's total energy at rest"

while later they say:

> "Even chess grandmasters, who sit for hours in states of intense concentration, burn only about 1.67 calories per minute while playing, compared to 1.53 calories per minute at rest"

That second figure seems to refer to whole-body expenditure, not just the brain. And intense cognitive work doesn't happen in a metabolic vacuum - there's increased cerebral blood flow, elevated sympathetic nervous system activity, changes in heart rate variability, hormonal shifts (cortisol, adrenaline). These all have systemic metabolic costs that go beyond the glucose the neurons themselves consume. So the "it's just a banana and a half" framing might be undercounting by quietly switching between brain-only and whole-body measurements.

Also somewhat related - the link to businessinsider about chess grandmasters is broken, but another very interesting rabbit hole here is how energy expenditure is actually measured. A lot of what consumer devices and even many studies report is based on proxy biomarkers like heart rate, HRV, weight, age, and sex, run through linear regression models. True calorimetry (indirect via gas exchange, or direct in a metabolic chamber) is expensive and impractical outside lab settings. That means the precise calorie figures cited with such confidence - the "100 to 200 extra calories" from a day of thinking, or the per-minute burn rates of chess grandmasters - likely carry wider error bars than the article suggests. We don't really have a great way to measure real-world energy expenditure accurately at the individual level, which makes me a bit cautious about the neat narrative of "thinking is calorically cheap, full stop."

That said, the core point about adenosine accumulation and perceived exertion affecting training quality is fascinating and well-supported — that part of the article is genuinely useful regardless of the calorie accounting.

See: https://cyberdefence24.pl/cyberbezpieczenstwo/blokady-w-poci...

That’s what all the other tools like ts-node and tsx do already.

I’m not sure what more are you expecting to do?

Typescript is build time type checked, there is no runtime component to TypeScript. If you want type checking you run tsc.

I think this is a great step in the right direction by node. It will save transpiration on the server and improve stack traves and whatnot.

> Once you start using the type system more extensively I suspect this will blow-up in your face.

I don’t see why. There isn’t any more runtime information in “complex” TypeSceipt types than in simple ones. It’s all build time - see above.

> What a missed opportunity to do it properly

Please explain in more detail what “doing it properly” means to you. Including the typechecker? If so that wouldn’t make sense - they would be competing with TypeScript itself, they shouldn’t, all the “third party plugins” rely on tsc for type checking.

But I guess that falls apart when we treat Safari as dependent on Apple, which it is.

I was surprised author made no mention of it (mqtt.org) but come to think of it it might be because author is specifically looking for queues it seems and MQTT works better as a PubSub, and its durability story which seems the main focus of the author is way different with very cool features - QoS - for delivery reliability but still not a classic queue

I think you should mention that „I looked” was done with help of your AI project. That does not spark much confidence given current state of LLMs and their „research”.

From your twitter article:

> So across the ~50 or so prominent whistleblower cases against big co's that I researched with futuresearch.ai, retaliation is common, harassment is rare but does happen, but murder is not.

I mean it all sounds fun but when you consider for example low-cost airlines in Europe (WizzAir, Ryanair) they’re really efficient without any of this, since their margins depend on it. Surprisingly their boarding isn’t as structured (only priority and then the rest) and they still only take around 30 mins total from first to last passengers being let through a gate in my experience. Their trick is probably some combination of motivated passengers, asking people to put stuff under the seat in from, remote stands that need a bus and flight attendants constantly announcing „please don’t block the aisle”.

it’s not as bad as you think. Either the hosted version or on a php hosting which are cheap.