Roundup #22: USB standard transition, Clippy the paperclip, and medical claims
I am happy to report that I have fully completed the annoying USB transition, and I recommend it!
All of my things are now USB-C
My laptop, iPhone, Airpods, Watch, iPad, Kindle and flashlight now all charge with USB-C. I bought a new travel adapter with 3 USB-C ports and even a small retractable cable to charge stuff.
This is obviously not a huge thing, but it’s pretty nice. With two cable (the one for my watch and my laptop charger) and this block I can travel and always charge everything I own. There is still one USB-A port for the odd thing that needs it, like my toothbrush.
There is a lot of valid criticism of the EU’s regulations milieu, but they did us a good one mandating USB-C for all mobile devices. That influenced Apple to adopt it. Also, it is a damn good standard actually in my opinion (not knowing much about the technical details, but speaking as a user). It’s compact, it can take a lot of current for charging stuff, and it’s symmetrical so you cannot plug it in ‘upside down’. That was always super annoying about USB-A and Micro-USB.
Fun piece about MS Office ‘Clippy’ AI assistant from the 90s
Bill Gates' response when first being shown the concept of 'Clippit' aka Clippy, the paperclip AI assistant in Microsoft word. 😂😂💀💀💀
But the entire concept of Clippy is back with a vengeance now with LLMs. I read a mind-blowing 2021 Substack post from Steven Sinofsky yesterday sharing all about how and why Clippy was created.
People in the mid 90s did NOT know how to use a PC. And the capabilities of the tech improved quickly, widening the gap between what was possible and what people were doing more and more. Microsoft's intent was to help people get the most out of their software.
Does this sound familiar in the context of AI in 2025?
The post also blew my mind in it's casual mention of the AI Agent hype from 1995:
"...the industry was buzzing with the idea of agents that would be able to do work on your behalf such as find cheap airline flights or schedule meetings. Everywhere from Apple to the MIT Media Lab were talking about agents. There was ample evidence this was not simply a weird vision in our corner of the tech world. In fact, by some accounts we were in a race to have the first and best guru in the box."
Wait... WHAT? 🤯
Sam Altman and I were both 10 years old in 1995. We were doing WordArt and laughing at Clippy. But we weren't really 'there' for the web boom. It seems like history is repeating itself.
Maybe this is just another platform shift after all. A generational one, nonetheless.
Link to the post from
:
A great post about health science by
Dr Katz has an interesting history, as one of the medical advisers/fact checkers of Dr Oz. Dr Oz helped millions of people live healthier lives, but also spread tons of misinformation and maybe even helped write the blueprint for the sensationalist ‘health influencer’ of today.
Along with Degrowth, cynics, and stupid AI takes, I file misinformation about science under the list of things that really trigger me. Although the part about Statins is not currently relevant to me or anyone around me (thankfully), he debunks a list of health claims that are currently trending in the piece as well.
📣 The Misinformation Machine
Health content is everywhere. Intermittent fasting. Seed oils. Coffee extending lifespan. Vitamin D reversing aging. The ketogenic diet preventing heart disease.
Everyone has an opinion. Almost no one has expertise.
And that’s a real problem because when it comes to my patients, this stuff isn’t just about abstract ideas.
It’s about real people making real decisions that affect their lives, often based on bad interpretations of bad data.
Take a few recent examples:
The New York Times recently covered an observational study suggesting that coffee improves healthy aging. It’s not a randomized trial - this paper is incapable of proving causality. But the article gets a lot of press. It will shape - or at least justify - behavior. And it’s based on useless, hypothesis-generating data.
My wife showed me a post from some influencer on Instagram claiming that vitamin D is an anti-aging drug - probably because they saw that Harvard/Mass General Brigham Communications endorses the conclusion. Why? Because a sub-study of the neutral VITAL trial found an association between vitamin D and telomere length.
This ridiculous logic is two steps removed from reality:
The VITAL trial showed no benefit in preventing cancer or cardiovascular disease.
Telomere length as an actionable target to slow aging is probably discredited — it hasn’t held up as a useful or meaningful clinical tool.
Or consider the Keto CTA study, where researchers followed patients on a ketogenic diet with coronary CTAs and found essentially the fastest progression of heart disease ever recorded. But the lead authors are now misrepresenting their own data, obscuring endpoints, and claiming there’s nothing risky about very high LDL cholesterol.
What’s wild is that the people downplaying that study live in the same online information sphere as many of the anti-statin voices. These perspectives often amplify each other and reinforce a narrative that’s really quite wrong.
More of my own thoughts on health claims and what to do about them
I certainly care about my health. I exercise, I’m mindful of what I eat, I take supplements. But I’m very wary about unsubstantiated, sensationalist claims. Be it microplastics, or the health benefits of all kinds of compounds or diets I am always going to be on the look out for a couple of things:
What kind of study was this? Is it a meta-analysis, a population study, in-vitro (a test tube), in animals, or in humans?
What was the quality of this study? What was the sample size? Was there a control? How was the control configured, was it double-blind? If it was a data-science exercise (e.g. a claim that drinking coffee promotes longevity), how was it set up? Is the dosage that was tested representative of what one may encounter in real life?
What logical steps are there between result and claim. Is this a mechanistic claim? A repeatable experimental outcome? A human trial?
What are the implications of this being true or false, and what actions one might take based on that? Is it safe? What is the least-regret option to take.
Mechanistic claims
The kind of claims I am most skeptical of are things like: “People with Alzheimers have lower levels of Phosphocreatine in their brain, therefore Creatine supplementation helps to prevent Alzheimer.” To make this sound more credible, the influencer or newspaper will usually add more details on the biologicial mechanism: “When creatine enters muscle or brain cells, it can be phosphorylated by the enzyme creatine kinase to form phosphocreatine (PCr).”
Sure, it CAN be. But is it? Since you can’t measure phosphocreatine levels inside of a brain without taking the brain out of the animal, you can’t verify if this is true in humans. The finding that Creatine levels are low in people with Alzheimer’s might reverse causality, or old people might have low Creatine levels anyway for some other reason. This is not to say that we should reject the potential of using Creatine for preventing neurological diseases outright, animal studies do look promising. It is also quite safe. If I had Alzheimer’s in my family, I’d probably take it. But I don’t, so I’m not taking it.
Dosage studied not in same ballpark as real life exposure
In the immortal words of Paracelcus: The dose makes the poison.
Alle Dinge sind Gift, und nichts ist ohne Gift; allein die Dosis macht, dass ein Ding kein Gift ist.
All things are poison, and nothing is without poison; the dosage alone makes it so a thing is not a poison.
—Paracelsus, 1538[1]
If you eat half a cup of apple seeds, you’re dead (Cyanide). 40 Bananas will do it too (Potassium). 80 grams of salt (too much salt per liter water in body), 6 liters of water (too little salt per liter water in body).
People keep trying to somehow proof that Aspartame is bad for you. And they are able to show all kinds of effects. In mice. At dosages that would require a human to drink 10 bottles of diet coke per day for years in a row. The way a body metabolizes and clears things from the body isn’t linear. Drinking alcohol is harmful for you1, but below a certain point (let’s say 1 glass per week), it’s just… not relevant. Likewise, construction workers who worked with Asbestos have a 1.5-2.5x higher chance of getting lung cancer than the base population2, but spending an hour on such a construction site once won’t really matter.
Please don’t take this to mean that there is no biohacking or healthy interventions in life that can be done. But I’m confident that a lot of people who obsess over processed foods, or sugar, or even microplastics can safely lighten up a bit.
The question remains: With all this uncertainty and unprovability of what is really harmful or healthy, what decisions do you actually make? What is important?
I emailed Gregory Katz to ask him how he decides what to eat
I emailed this:
So what IS a healthy diet. I often struggle with the term 'processed food' because technically if I make a bread myself that is a processed food. Even the flour is already processed. Olive oil is processed, every organic fruit or vegetable is processed in some way.
There are specific compounds that you can try to avoid: transfat, that red dye, highly processed meat like hotdogs (but what about butchers sausages?)
But there are also so many compounds incorrectly targeted: aspartame, msg, that scare about black plastic utensils where they multiplied the effect by 7x accidentally. There are obviously macro decisions you can make: go easy on the sugar, prefer whole grain {rice, bread, etc} over white.
Do you have a set of running guidelines, that go beyond sweeping heuristics like 'avoid processed branded food'? How do you approach it yourself?
And he replied:
i think a healthy diet is one that keeps people in calorie balance and an unhealthy one is one that makes someone overweight or underweight. a healthy diet should also be something we enjoy and is sustainable - so there's a lot of personalization
i try to prioritize protein, eat vegetables, and don't eat too much junk. i eat white potatoes and white rice. i don't have too much bread. i am willing to eat dessert, especially if it's in a social situation or with my kids
i don't worry about processed vs unprocessed, i don't worry about msg/aspartame/sugar in moderation.
i think most branded products and things that don't go bad on the shelf for weeks are probably bad for us and i try to avoid when i can
In many countries in Asia, people are obsessed with not putting bags on the floor
Many, many times, after putting my bag on the floor in a cafe or restaurant, a waiter came over as fast as possible to pull up an extra chair or any other item for my bag. I was reminded of this because of this cute little bag-bags at a cafe.
I always just filed it away as having to do with the bags potentially getting dirty, but this time I decided to google it. It turns out there is a superstition about putting a bag on the floor. It’s bad FengShui.
包包放在地上, 钱财也走掉
“Bag on the ground, money walks away”
Some extra googling brings up a whole bunch of potential explanations. Putting a bag on the floor is ‘disrespectful towards your wealth’. I guess that hints at something like gratitude for being well-off enough to even have a bag, and being callous with it might jinx that. Another more practical explanation suggested the bag would be less likely to be forgotten this way. I don’t buy that one, I’ve forgotten my bag many times whether it was on the floor or not.
As often with these kinds of things, the ultimate actual origin is probably lost to the times. But it’s nice to know how it comes across when I put my bag on the floor. Like “Wow this guy is so disrespectful of his own wealth.” It’s nice of the waiters at restaurants to help prevent my money from walking out the door. Although I guess it would also be pretty annoying for them if it walked away before settling the bill…
A good moment to mention the most likely explanation for the observation that people who drink wine in moderation live longer. It is not because of the resveratrol or whatever else in the wine. People who drink wine regularly:
are rich enough to afford it (meaning they also on average eat healthier, can also afford healthcare, and probably exercise),
not having any serious diseases that preclude them from drinking,
having people to drink with (not lonely).
Even this is to the best of our knowledge and pretty uncertain. To what degree are lung cancers attributed to the right cause? If someone is a smoker (30x increase in lung cancer risk) and asbestos worker, do you blame the smoking or the asbestos? Or both? This probably depends on whether the ‘cause of cancer’ form at the hospital supports adding two causes or only one.