I used to have CFS, but apart from the occasional temporary post viral fatigue that many get, it’s gone. And what is CFS but long term post viral fatigue?
One thing I learned is to ignore figuring out the exact supplements, because you’re playing an impossible balancing game with poor feedback mechanisms. There’s too many inputs.
What helped me was a combination (no one thing can solve it) of therapy (being able to listen to and not suppress emotions), key supplements (magnesium/iron - check out lactoferrin and anaemia of chronic infection), exceptional oral hygiene to reduce inflammation (4 minutes per brush), exceptional gut health (many viruses cause problems with the gut), exercise (eventually), and more…
I never used niacinamide or any of the supplements you used, which shows you that there’s no single approach. I agree that it appears to correlate with an unaddressed infection.
I'd like to argue that it is possible to fight the balance game to an extent.
I've also fought a similar battle to you in a similar manner by "fix everything according to best practices."
When it comes to supplements and nutrition, it may be an impossible game, but what's important in finding the right approach is understanding at least the basic mechanism of action and knowing what you're targeting.
What I'm hearing you say is that you had an app with a performance issue, and you couldn't pinpoint whether it was CPU, Memory, Disk, Network, etc, so you solved it by doing a wholesale system upgrade by giving it all the basics that modern science says are the typical health best practices. Magnesium/Iron, exercise, gut biome, oral hygiene, etc to help. All the things you've listed are anti-inflammatory (of course there's other benefits) but generally, anti-inflammatory things are pretty good at making the body run better.
NAC on the other hand, is a precursor to glutathione, a significant anti-inflammatory molecule in the body, and usually the limiting reagent for glutathione synthesis. So it's also arguable that NAC had an anti-inflammatory effect on the body similar to the effects you received from your regimen.
Finding the right supplements are possible for sure. Usually what is needed though is a thorough analysis/observation of someone's diet and then working back the potential malnourishments that are most likely to occur and in alignment with the symptoms. But usually this takes months and years of learning and understanding to even know where to begin when it comes to suggesting a supplement.
Have enough supplements to cover your bases and ideally cover it off with whole foods when possible.
To be clear, I found it wasn't a good use of time to spend years experimenting with many supplements that end up working temporarily and then having an antagonist effect on something else that appears months down the line.
The best use of time was taking a holistic approach. Supplements didn't save me - but without some basic supplements I wouldn't have been saved. And I agree, some basis in nutrition is important.
I always wondered if some hidden pattern would be exposed when visualising numbers in unconventional ways in numbers with no known pattern such as Pi or prime numbers. A sort of multi-dimensional rendering that suddenly reveals a hidden pattern.
This is sort of how Fermat's last theorem was solved by Andrew Wiles (forgive me if I misrepresent this proof, my math is a few years rusty) - by creating a different kind of representation of elliptic curves, it was possible to compare them to modular forms in a way that created a contradiction that proved the theorem correct.
Well some numbers expose patterns when written as a continued fraction. In particular e becomes pretty regular.
You can modify the continued fraction slightly to make pi regular as well, but the normal continued fraction sequence doesn't give much of an insight. Other than the fact that 3 + 1/(7 + 1/16)) is a damn good approximation (7 digits, pretty good for something that can be written using only 4 digits total: [3;7,16]).
Larger integers in continued fractions mean you get 'more information' out of the limb. That means not only is Phi "1s all the way down" it is the continued fraction that converges the slowest. If you've ever used the iterated matrix product (which is a specific edge-case of the algorithm to convert continued fractions to decimals), you'll know how slow it is!
Well, they buried the lede with this one. Using LLMs were better for some tasks and actually made it worse for others.
The first task was a generalist task ("inside the frontier" as they refer to it), which I'm not surprised has improved performance, as it purposely made to fall into an LLM's areas of strength: research into well-defined areas where you might not have strong domain knowledge. This also is the mainstay of early consultants' work, in which they are generalists in their early careers – usually as business analysts or similar – until they become more valuable and specialise later on.
LLMs are strong in this area of general research because they have generalised a lot of information. But this generalisation is also its weakness. A good way to think about it is it's like a journalist of research. If you've ever read a newspaper, you often think you're getting a lot of insight. However, as soon as you read an article on an area of your specialisation, you realise they've made many flaws with the analysis; they don't understand your subject anywhere near the level you would.
The second task (outside the frontier) required analysis of a spreadsheet, interviews and a more deeply analytical take with evidence to back it up. These are all tasks that LLMs aren't strong at currently. Unsurprisingly, the non-LLM group scored 84.5%, and between 60% and 70.6% for LLM users.
The takeaway should be that LLMs are great for generalised research but less good for specialist analytical tasks.
I was thinking about this last night. It’s a new version of Gell-Mann amnesia. I call it LLm-man amnesia.
When I ask a programming question, chat GPT hallucinates something about 20% of the time and I can only tell because I’m skilled enough to see it. For all the other domains I ask it questions if I should assume at least as much hallucination and incorrect information.
I see this as for drill-down thinking from a broad -> specific concept AI seems to be helpful when supplementing specialist work. However like you both mentioned: when needing more focused and integrated answers AI tends hinders performance.
However as the paper noted, when working within AIs areas of strength it improved not only efficiency but the quality of the work as well (accounting for the hallucinations). As you mentioned:
> When I ask a programming question, chat GPT hallucinates something about 20% of the time and I can only tell because I’m skilled enough to see it
This matches their Centaur approach, delineating between AI and one’s own skills for a task which—with generalized work—seems to fair better than not using AI at all.
It's because a lot of engineers are learning to become better plumbers, not better engineers.
Trying new technologies means you're mostly becoming better at using someone else's APIs - this is the path to eventual burn out as the churn continues.
There is a better path - See through the hype. Ask those around you what's the downsides to this approach? And you'll often get blank stares... Why? Because they don't know either - And if they don't know the downsides, they don't really know. They are following the hype curve.
Focus on the fundamental engineering principles and asking better questions - take the bottom-up approach and the reward is you'll find teams that aren't taken by the hype curve so easily.
PS. There are a lot of good technologies that come out, but staying behind the hype curve a little helps you make better judgements over time.
Her son talks about the paradox of how she raised him abusively, and yet it was only through her books that he was able to heal himself many years later.
The timeline here is important -- her son was born in 1950, and her first book wasn't written until 29 years later.
I think it's much fairer to view her work as a heroic attempt to atone for her own sins. She was abused by her own parents, did the same to her children in turn, and then dedicated her entire professional life to figuring out how to break that cycle. It was too late for her own kids, but she was able to help others instead.
So it's unfair to judge her by saying she didn't practice what she preached, because they came at different times. What she wrote about was sadly informed by her own personal experience.
The Body Keeps the Score (by Bessel van der Kolk) is another great book, but it's about trauma generally, and especially things like PTSD. It doesn't focus much on specifically parental abuse the way the blog post, and Alice Miller, do.
By that logic we should avoid most psychologists(think Anna Freud as a complete opposite here). These are all great books, but there’s something to learn from them all. If you read Miller’s most popular book and then her son’s, it will go to show you how difficult it is to break the generational cycle. Or simply put, how much people actually practice what they preach.
It gets you more of whatever you love doing - even if no one reads it - because you get better at whatever you write about.
If you knew no one would ever read your writing, would you still write it? If yes (the likelihood is no one will read it apart from your future teammates) you'll have found your subject.
It can give you jobs, learning & connections, but it also takes time. Time that can be used for other things that could get you the jobs, learning & connections you want without writing. There's no one way to approach it, you need to find what works for you.
What are the genetic variants? I can’t access the paper, but I assume if anyone can we’d be able to run our DNA results (from 23andme, etc.) through this to see how high we score.
Unlikely as those services barely scrape the surface of genetic coding, if I recall they only sample from specific regions of the DNA to generate their data.
But I am also likely wrong, so I am hoping someone can challenge/confirm this thought.
The $200 is their sales price but it happens frequently enough that it’s not hard to get. The regular price is $600 and they do try to upsell with running disease specific tests against your genome. I thought the current wholesale cost of the genome sequencing was $75. After shipping costs there wouldn’t be much profit out of $200 So it is cutting things pretty lean but AFAIK still should be doable.
The cost of whole genome sequencing claimed by manufacturers is around $200 for the latest Illumina tech and sub 100$ for the latest MGI tech. All this assumes full load on max throughput machines - e.g. enough samples to sequence (otherwise the price invariably goes up).
These are not reference points for the pricing over the last couple years since these products were only announced last/this year respectively. 600$ WGS seems reasonable price for the last couple years in Europe, $200 is not. See this spreadsheet by @AlbertVilella who's a very useful source for his kind of stuff:
It was actually €200 but at the time I did it the exchange rate was pretty close to 1:1, so a recent exchange rate drop could explain some of the discount. There is an added delay of 6 weeks to the special so I guess if they’re running under max utilization it could make sense to fit these in at above the cost of the reagent. There is also competition from Nebula Genomics so there is a market share and mind-space component to denying a competitor potential customers. I think I read something somewhere about cheaper bootleg reagent being used to get around patents. I think Nebula Genomics was hit by a lawsuit around that. Plus there is the potential outsourcing to less scrupulous countries that may go on to sell access to data. I think there was a rumor that Nebula Genomics was using a Hong Kong lab and later a Russian lab. I’m not worried about my DNA getting out, I already consider my DNA privacy defeated which was confirmed by my ability to find second cousins with it.
One thing I learned is to ignore figuring out the exact supplements, because you’re playing an impossible balancing game with poor feedback mechanisms. There’s too many inputs.
What helped me was a combination (no one thing can solve it) of therapy (being able to listen to and not suppress emotions), key supplements (magnesium/iron - check out lactoferrin and anaemia of chronic infection), exceptional oral hygiene to reduce inflammation (4 minutes per brush), exceptional gut health (many viruses cause problems with the gut), exercise (eventually), and more…
I never used niacinamide or any of the supplements you used, which shows you that there’s no single approach. I agree that it appears to correlate with an unaddressed infection.