To add more information, the intervention was guidance about caffeine intake. From the Methods:
> If allocated to caffeinated coffee consumption, patients were encouraged to drink at least 1 cup of caffeinated coffee (or at least 1 espresso shot) and other caffeine-containing products every day as per their usual lifestyle. It was recommended that patients in the coffee consumption group not intentionally increase or decrease consumption of coffee or other caffeine-containing products.
> If allocated to the abstinence group, patients were encouraged to completely abstain from coffee, including decaffeinated coffee, and other caffeine containing products.
Waymo was on a roll in San Francisco. It still is, but it used to be, too. (With apologies to Mitch.) This is utter sensationalism. Fortunately, the state's regulations have liberated the good people of SF from being able to shoot ourselves in our own foot on this particular issue.
Key Takeaway: Get a CT or CTA scan, and if you can afford it go for the CTA with Cleerly.
There is a reason that we don't recommend getting imaging for everyone, and that reason is uncertainty about the benefit vs the risks (cost, incidentalomas, radiation, etc, all generally minor). Most guidance recommends calcium scoring for people with intermediate risk who prefer to avoid taking statins. This is not a normative statement that is meant to last the test of time: it may well be the case that these tests are valuable for a broader population, but the data haven't really caught up to this viewpoint yet.
The central point of his article is that he went to a doctor who followed the guidelines, tested him and found he wasn't at risk for heart disease.
But then he went to another, very expensive concierge doctor, who did special extra tests, and discovered that he was likely to develop heart disease and have a heart attack.
Therefore he is arguing that THE STANDARD GUIDELINES ARE WRONG AND EVEN IF YOU DO EVERYTHING RIGHT AND YOUR DOCTOR CONFIRMS IT YOU MAY BE LIKELY TO DIE OF HEART DISEASE ANYWAY, SO ONLY THE SPECIAL EXTRA TESTS CAN REVEAL THE TRUTH.
I want a second opinion from a doctor. Is this true? Is this for real? Because it smells funny.
I strongly suspect the truth is both are "right", but they're both optimized answers to slightly different problems.
Mainstream medicine is hyper optimized for the most common 80% of cases. At a glance it makes sense: optimize for the common case. Theres some flaws in this logic though - the most common 80% also conveniently overlaps heavily with the easiest 80%. If most of the problems in that 80% solve themselves, then what actual value is provided by a medical system hyper focused on solving non-problems? The real value from the medical system isnt telling people "it's probably just a flu, let's just give it a few days and see" it's providing a diagnosis for a difficult to identify condition.
So if your question is "how do we maximize value and profit in aggregate for providing medical care to large groups of people", mainstream medicine is maybe a good answer.
But if your question is "how do we provide the best care to individual patients" then mainstream medicine has significant problems.
Part of providing good care is not burdening the patient with tests or treatments that are very unlikely to yield benefit. Put another way, the mission of healthcare is not "health at any cost."
The mission of healthcare in the eyes of those who provide it, isn't "health at any cost".
For the people on the other side, "health at any cost" is pretty much the goal, usually limited by the "cost" side of things, especially in the parts of the world where they haven't yet figured out the whole "healthcare for the public" thing.
Cost here doesn't just include financial cost, but also time. As an extreme example, you could surely catch diseases earlier by visiting a doctor for an hour or two every day - getting tests for all sorts of things you might have conceivably developed. But that would make your life worse, and so most people wouldn't do that even if it was free.
Research science in this area has been in agreement for a long time now that ApoB is a more informative indicator than just LDL-C, because there are a variety of different atherogenic particles, not all LDL particles are created the same, etc.
His ApoB numbers are quite readily and apparently out of range. Hell, even his LDL is out of range for the two largest lab providers in the US - Labcorp and Quest both have <100 for their reference range. But the science shows that plaque progression is still generally occurring at levels above 70 LDL-C even with low Lp(a) and other atherogenic particles - the reference ranges are likely to get moved lower and lower as practice catches up with research.
His numbers are well within the range of concern based on pretty universal consensus across the research in this area over the past couple of decades. Preventative cardiologists and lipidologists would almost certainly agree with this concierge doctor.
Thanks for the astute and informed comment. So re-reading that portion of the article, it seems to me the answer to my question is not that any general or consensus guidelines are wrong, but that a company called Forward Health is doing lipid panels and providing an incorrect interpretation of the results.
OP's LDL-C was 116 and this is on the very top end of what Forward Health's report says is OK, their report is wrong, this number is bad.
All the stuff about needing to measure ApoB, needing a high end concierge doctor, and the very long article about measuring 10-20 different numbers and doing more exercise than the guidelines and being at risk of heart attack if you don't do amounts of exercise that the author consider unreasonable etc., some of this may have value, but this all seems to be a lot of very lengthy personal opinion by the techbro author of the post. The key insight is simply that your LDL-C becomes a cause for concern over 100, perhaps even over 70, and he was not as healthy as some tech company told him he was. No surprise there, I will talk to actual doctors instead of using services from "tech forward" startups any day of the week.
I would agree that this article overstates a lot of things.
ApoB is still a reasonable thing to check though, at least once - Lp(a) is the primary cause of atherogenic particle counts being high when LDL-C isn't the culprit, and it's usually a genetic factor. Having a high Lp(a) will bounce your ApoB up and give you a better understanding of the total atherogenic particle load. You could have fine LDL-C or Lp(a) on their own but the total amount could be enough to be worrisome.
Lp(a) being problematic is definitely less common than it being more or less fine, but it's certainly not incredibly rare, either.
The claim on an individual level is not objectionable to me. The question is that if we extrapolate it out to the population and actually take this action for everyone, do we make people better off? This is what clinical trials (or at least large observational studies) try to achieve. Right now, it is not clear.
His evidence is also kinda weak. And appeal to authority largely about someone who he's paying to tell him he has health problems. The incentives aren't aligned.
I also disagree that the 50the percentile is the breakpoint between healthy and unhealthy. There's a lot more to deciding those ranges beside "well half of the population has better numbers"
If you think a 100+ LDL-C is normal you're basing things off of significantly outdated information.
Expect the normal range to drop in the coming years as well - the AHA and NLA have both been talking about how this needs to go lower, and the science is robust. See my other comments for study links, the NLA's latest guidance, etc.
If your doctor is only getting concerned at 160+, find a new doctor.
If I die at 90 of a heart attack havjng maintained the ability to live independently up until then, I’d take that as a massive win compared to my relatives suffering through a decade of me with worsening dementia.
Cardiovascular diseases are huge risk factors for dementia, so if your goal is to avoid dementia you should try to have a healthy cardiovascular system.
If health science was as simple as health outcomes are proportional to one or two measurement percentiles, sure. But that's hardly true. Health is a lot more complex than that and the disease risk cannot be quantified by a small number of parameters
Maybe he got missed--let's concede that. What about the other 10 or 100 or 1000 or subjected themselves to tests and didn't find anything? Where are their stories?
If you have enough people, the tests, themselves are eventually going to harm somebody.
For example, certain scans require contrasts like gadolinium that bioaccumulates. That's not a big deal if we only pump it into people 2 or 3 times in their lives when something in their body is about to explode. It's a lot bigger deal if we're doing that to them every year.
The bottom line is these tests aren't some sort of one-size-fits-all panacea, and nor can they perfectly predict the future. In fact Oprah herself backtracked on it, via an article by Dr. Oz in her magazine in 2011: https://www.oprah.com/health/are-x-rays-and-ct-scans-safe-ra...
A good rule of thumb is don't take medical advice from Oprah or Dr. Oz. But in the case of the latter article, he wasn't wrong.
> But then he went to another, very expensive concierge doctor, who did special extra tests, and discovered that he was likely to develop heart disease and have a heart attack.
It’s scarily common in medicine for doctors to start specializing in diagnosing certain conditions with non-traditional testing, which leads them to abnormally high diagnosis rates.
It happens in every hot topic diagnosis:
When sleep apnea was trending, a doctor in my area opened her own sleep lab that would diagnose nearly everyone who attended with apnea. Patients who were apnea negative at standard labs would go there and be diagnosed as having apnea every time. Some patients liked this because they became convinced they had apnea and frustrated that their traditional labs kept coming back negative, so they could go here and get a positive diagnosis. Every time.
In the world of Internet Lyme disease there’s a belief that a lot of people have hidden Lyme infections that don’t appear on the gold standard lab tests. Several labs have introduced “alternate” tests which come back positive for most people. You can look up doctors on the internet who will use these labs (cash pay, of course) and you’re almost guaranteed to get a positive result. If you don’t get a positive result the first time, the advice is to do it again because it might come back positive the second time. Anyone who goes to these doctors or uses this lab company is basically guaranteed a positive result.
MCAS is a hot topic on TikTok where influencers will tell you it explains everything wrong with you. You can find a self-described MCAS physician (not an actual specialist) in online directories who will use non-standard tests on you that always come back positive. Actual MCAS specialists won’t even take your referral from these doctors because they’re overwhelmed with false cases coming from the few doctors capitalizing on a TikTok trend.
The same thing is starting to happen with CVD risks. It’s trendy to specialize in concierge medicine where the doctor will run dozens of obscure biomarkers and then “discover” that one of them is high (potentially according to their own definition of too high). Now this doctor has saved your life in a way that normal doctors failed you, so you recommend the doctor to all of your friends and family. Instant flywheel for new clients.
I don’t know where this author’s doctor fits into this, but it’s good to be skeptical of doctors who claim to be able to find conditions that other doctors are unable to see. If the only result is someone eating healthier and exercising more then the consequences aren’t so bad, but some of these cases can turn obsessive where the patient starts self-medicating in ways that might be net negative because they think they need to treat this hard to diagnose condition that only they and their chosen doctor understand.
It's important to note that there's geographic variability in guidelines. Also, the article doesn't give enough information about the author's other risk factors. For a similar patient (based on the initial lab results), treated by a doctor adhering to the European guidelines, at least the following items would have been considered:
- Lipid lowering drugs
- ApoB testing
- Coronary CT (if the pre-test likelihood of obstructive coronary artery disease was estimated to be > 5%)
The year is 1846, and a doctor has a radical new idea: doctors should wash their hands between performing autopsies and delivering babies!
You're not sure of whether this is a good idea or not, so you ask various physicians, and the consensus is unanimous: the very suggestion is offensive, do you think doctors are unclean?
Not sure I follow or maybe you skipped typing a word.
You listed the risks and concluded “all generally minor.” The benefit is absolutely nonzero. So, what’s the hold up?
And how have the data not caught up? People outside the US are getting the CT scans, while US doctors prefer to lick their finger to guess the weather.
My wife’s last interaction with a doctor: patient presents with back and chest pain accompanied by occasional shortness of breath at the age of 39, doctor reluctantly asks for a EKG - which takes 5-10 minutes and is done in the next room, right away and covered by insurance with a small copay - and has the gall to be surprised when EKG showed subtle abnormalities. If she hadn’t advocated for herself, as the OP argues, doctor would just skip the EKG.
This experience left me thinking maybe doctors are discouraged from asking for imaging and guidelines are there to protect their criminally negligent behavior. I have no proof or even proxy data for the claim about doctors being discouraged from asking for imaging. But it is objectively criminally negligent to not ask for imaging in a case like this.
"Smaht" people continuously parrot things they read elsewhere, usually in a contrarian way, to assert themselves in a futile and shallow way.
There is absolutely nothing wrong with getting one CT at a specific point in your life to right a disease which, as TFA states, has a 25% incidence rate.
The smaht ones will now point me to that study of 1-5% of cancers being linked to CT scans. Yeah, sure, but those are not from people who got one-two in their lives.
One thing that wasn't mentioned is the max sustained screen brightness for SDR, which is higher on the M4 Pro (1000 nits) compared to the M4 Air or M1 Pro (500 nits).
There’s an awesome app called Vivid which just opens the HDR max brightness. I use it all the time with my M3 Pro when working outside and I believe it also works on earlier models.
There are so many base features that are inexplicably relegated to 3rd party apps. Like a better finder experience. Or keeping screen on. Or NTFS writing.
NTFS writing isn't that inexplicable. NTFS is a proprietary filesystem that isn't at all simple to implement and the ntfs-3g driver got there by reverse engineering. Apple doesn't want to enable something by default that could potentially corrupt the filesystem because Microsoft could be doing something unexpected and undocumented.
Meanwhile if you need widespread compatibility nearly everything supports exFAT and if you need a real filesystem then the Mac and Windows drivers for open source filesystems are less likely to corrupt your data.
I'll take ntfs-3g over the best implementation of exFAT in a heartbeat. Refusing to write to NTFS for reliability purposes, and thereby pushing people onto exFAT, is shooting yourself in the foot.
At which point you're asking why Apple doesn't have default support for something like ext4, which is a decent point.
That would both get you easier compatibility between Mac and Linux and solve the NTFS write issue without any more trouble than it's giving people now because then you'd just install the ext4 driver on the Windows machine instead of the NTFS driver on the Mac.
> There are so many base features that are inexplicably relegated to 3rd party apps.
> Like a better finder experience.
> Or keeping screen on.
Do you mind linking or naming which tools you use for those 2 purposes?
Asking out of pure curiosity, as for keeping the screen on, I just use `caffeinate -imdsu` in the terminal. Previously used Amphetamine, but I ended up having some minor issues with it, and I didn't need any of its advanced features (which could definitely be useful to some people, I admit, just not me). I just wanted to have a simple toggle for "keep the device and/or display from sleeping" mode, so I just switched to `caffeinate -imdsu` (which is built-in).
As for Finder, I didn't really feel the need for anything different, but I would gladly try out and potentially switch to something better, if you are willing to recommend your alternative.
I use the Finder and Raycast heavily. Raycast is not, and does not sell itself as, a Finder equivalent.
OP: I've tried all the Finder replacements. Path Finder, for example. At the end of the day, I went back to Finder. I always have a single window on screen with the tabs that I use all day. This helps enormously. I show it on YouTube here (direct timestamp link): https://youtu.be/BzJ8j0Q_Ed4?si=VVMD54EJ-XsxkYzm&t=338
Finder is the number one reason it boggles my mind that people claim macOS as head and shoulders above other OSes "for professionals". Finder is a badly designed child's toy that does nothing at all intuitively and, in fact, actively does things in the most backwards ways possible. It's like taking the worst of Explorer (from Windows XP), and smashing it into the worst of Dolphin or Nautilus; and, to top it off, then hiding any and all remaining useful functionality behind obscure hot keys.
It has been more or less the same as long as I've used it (20 years or so). Familiarity is a plus. It is a pretty simple and straightforward tool. I'm not sure what you might find perplexing about Finder.
Who said it was perplexing? If anything, it's the opposite. It's so simple and rudimentary as to be antithetical to filesystem navigation.
Back/forward operate on history, not on hierarchy; at least have an "Up" button. There's no easy way to navigate the non-prescribed folders without adding every folder to the favorites list; hell, there's not even a "Home" link by default. Simple location navigation is hidden behind Cmd+G versus being evident. Easily jumping up the tree from your current location is hidden. Etc, etc, etc. It acts like the iPhone file manager, except the filesystem isn't a sandbox on macOS and you regularly need to navigate around it.
I'm sure if it's the only FS manager you ever use then it's just fine and you've learned all the quirks. But for people that regularly use other (better) managers on other OSes, it's severely lacking in ergonomics and functionality.
cmd+g is not hidden it is a menu bar item under "Go". You can navigate hierarchy with the path on the footer. I believe home link is in fact default, its been there on the sidebar as long as I can remember (only it is called your user folder not "Home").
Eh, I feel the opposite. Finder is much more usable to me, but of course I use the shortcuts like cmd-up to go up or down instinctively now. It is a bit ironic for such a mouse oriented OS everywhere else.
Still alt-clicking on the window title to see the whole folder hierarchy is easy to remember and doesn't clutter up the UI (err cmd-clicking? It's muscle memory so I forget). The fact that it works on most native apps with file titles as well I awesome.
Welcome to the Mac ecosystem. Where basic functionality is gated behind apps that Apple fans will tell you "are lifesavers and totally needed in Windows/Linux/etc)" for $4.99-14.99/piece. And, when they get popular enough, Apple will implement that basic functionality in its OS and silently extinguish those apps.
And that's when they let you modify/use your OS the way you want.
A far as I understand Windows only has a toggle for HDR on vs off, that's not what we're talking about here, this is about forcing the full brightness of HDR always, even outside videos. It's something that manufacturers don't allow for as it reduces display life, it would actually be an anti-feature for a consumer OS to expose as a setting. It'd be like exposing some sort of setting to allow your CPU to go well beyond normal heat limits.
I don't mind that. 3rd party Mac utilities are nice: well designed, explained and do what they're supposed to because someone makes a living of it. I'm happy to pay these prices.
I would personally be afraid of using that in case it causes damage long-term to the screen either due to temperature or power draw or something. Idk if there are significant hardware differences but in this case I would guess there’s a real hardware reason for it?
I imagine what those custom brightness apps do is not magically increase the brightness, but change the various pixels' brightness in accordance to some method/algorithm such that you see what appears to be brighter whites when placed next to certain other colors.
It's not what is implied by the parent post - where the mac is limiting the brightness only to have the app unlock it.
No, I believe the issue is Apple limits the top half or so of the brightness/backlight level for HDR content only. The apps allow it to be used for normal non-HDR content.
...I'd have to say that seems like a radical reading of the text.
No; you can adjust screen brightness just fine with the built-in settings, including with the F1 and F2 keys (plus the Fn key if you've got them set that way).
This Vivid app is specifically for extra HDR levels of brightness. I've never had a problem with my M1 or M4 MBPs, either inside or outside, with the built-in brightness levels. (But, to be fair, I don't use it outside a lot.)
This is not a mainstream view of the science, and it's worth noting that this perspective is also not supported by the OP or by the JACC article that it's citing.
It's true that most doctors and pharmaceutical companies maintain that statins are effective. But there are plenty of statistically educated people that don't think they have much of an effect on all-cause mortality.
There are conflicting incentives here, and as usual we don't care about someone else's p value, we care about argmaxing our own utility functions.
> This is not a mainstream view of the science, and it's worth noting that this perspective is also not supported by the OP or by the JACC article that it's citing.
Your comment is an appeal to authority. While I have my problems with characterizing statins as dangerous drugs, the article is not particularly spicy. In particular, this part:
> Because the link between excessive LDL cholesterol and cardiovascular disease has been so widely accepted, the Food and Drug Administration generally has not required drug companies to prove that cholesterol medicines (such as statins) actually reduce heart attacks before approval. So drug companies have not had to track whether episodes like heart attacks are reduced.
...is true, and controversial only amongst people who don't know the evidence. Which, unfortunately, is many doctors and "experts".
In general, saying any variation on "experts disagree" is not a rebuttal to a question of medical evidence. You would perhaps be surprised to know how many practicing physicians have no idea what level of evidence backs the drugs that they prescribe.
The view among "authorities" is certainly something I find relevant in assessing a highly opinionated but thinly sourced medium article from someone who, respectfully, I've never heard of and know nothing about. Certainly it would be defeasible by a closer look at the research itself. But, barring that, it's a very useful heuristic.
I'm not suggesting you should take the medium article at face value either. Just that if you don't know enough to evaluate the evidence, you don't know enough to dismiss any particular opinion.
People are far too willing, today, to defer their thinking blindly to a consensus of opinions, but worse, to accuse anyone who also doesn't defer of being malicious.
I for one appreciated the clarification that it was not mainstream, since sneaking a random controversial take into a comment thread as if it was fact without noting that it's contentious is disingenuous.
No, they don't. If you don't know enough to argue on the merits, don't argue. A count of opinions is not an argument.
> sneaking a random controversial take into a comment thread as if it was fact without noting that it's contentious is disingenuous.
And again, you're justifying your judgment and dismissal based on hearsay. Saying "I refuse to believe it because experts disagree" is fine if you're unable or unwilling to look into an issue yourself, but in that case you have to realize you're basically ignorant.
I realize that we all go through life taking most things on faith, but that also means that you should not cling to the opinions of others as a substitute for thought.
yes they do. for one thing you do not make the rules around here; no one cares what you think counts as suitable grounds for arguing. For another, yes, authority has some merit. Doesn't make it fact, but certainly the prior we ought to assign for "medical authorities are correct" is quite high. Not certainty, but pretty confident, all else being equal.
edit: I see you added "I realize that we all go through life taking most things on faith, but that also means that you should not cling to the opinions of others as a substitute for thought."
Don't worry, nobody's doing that here. It's a question of weighting, not clinging. Maybe you mistook "this is not mainstream" to mean "this is definitely false because it's not mainstream"? It does not mean that. It is just helpful context for evaluating credibility.
> You're asserting that a extremely well-known logical fallacy is not a fallacy.
There are two distinctly different fallacies of appeal to authority (which overlap, since all of the second are also the first), this form is the form which is a deductive fallacy (appeal to status), but not the form that is a fallacy in inductive argument (which is appeal to false authority). It is important to distinguish them because while deductive fallacies are much more clear cut, they are also far less relevant to most real world debate, which rarely is about proving something is true by logical necessity assuming some set of axioms, but that is the only place that deductive fallacies are inappropriate, since all a deductive fallacy is is a form of argument in which the conclusion does not follow from the premise by logical necessity.
There is no logical fallacy in play here. Nobody is saying “the argument is wrong because of who said it”. When assessing the probable significance of an agglomeration of empirical data, it’s valuable to know what experts in the field think about the data and their consensus about the inferences we can draw from it—even if the consensus might be mistaken: because the consensus is usually right.
> There is no logical fallacy in play here. Nobody is saying “the argument is wrong because of who said it”.
The OP literally dismissed the parent based on nothing more than the opinions of others.
> When assessing the probable significance of an agglomeration of empirical data, it’s valuable to know what experts in the field think about the data and their consensus about the inferences we can draw from it—even if the consensus might be mistaken
I already conceded that, if you have no ability or capacity to think or investigate the issue yourself, it's perfectly fine to defer to the opinions of others. But in doing so, you remain ignorant on the matter.
> because the consensus is usually right.
No. I understand that's a comforting belief -- and even politically charged, today -- but it's just an assertion.
Well yes, exactly: it's just consensuses all the way down. Which is just another way of saying "I feel like it's right and you're wrong, even though I have no actual evidence either way."
But you have to separate the fallacy from it being supportive evidence of other data.
There is a difference from saying X must be true because Y person said it and they're an expert on Z. But when there is consensus between appropriate experts - such as researchers that specialize in this field - and it is in support of other specific evidence it is supporting evidence that the other specific evidence is compelling.
If I have 10 CPAs explaining a specific bit of the tax code to me and they are pointing out the specifics of the tax code, it is not fallacious to note that these people are experts and are pointing to specific evidence that supports their point.
as others have noted, you seem to be unaware of what exactly the fallacy refers to. You might want to look it up. It is not "citing an authority at all" but rather "citing an authority's opinion as though it were logical fact". Which nobody is doing here.
> I for one appreciated the clarification that it was not mainstream, since sneaking a random controversial take into a comment thread as if it was fact without noting that it's contentious is disingenuous.
(emphasis mine)
In other words, you didn't just passively ignore the parent (which would be fine), you posted about it, and not only that, you called it a lie. [1].
When you call something a lie like that, you're making an argument, so you'd better be prepared to bring the evidence.
[1] I realize that you're actually saying that it's "disingenous" that they posted this without some kind of disclaimer that it's a "controversial argument", but to the core of the issue: if you need that disclaimer, you aren't qualified to judge the content. For all you know, it isn't controversial at all.
... What is going on? Yes, that's what it means, that's why I used it. Notice it does not say 'lie' or anything directly synonymous with lying. It's related, sure, but not the same.
You can tell that it's not synonymous with lying because had you said to me "are you saying they're lying?" I would have said 'no'. This is always the case with semantic disagreements: if you want to know if someone intends a certain connotation, you can ask if they would agree with a rephrasing.
The disingenuousness is presenting a non mainstream theory as if it is fact. Anyone reading that initial comment probably has no idea whether "cholesterol and statins are suspect science". Had they said "some people think they're suspect science" there would have been nothing wrong. To claim they're suspect as a fact is disingenuous: it could be true, or it could be that the person posting it is one of those anti-establishment nuts who disagrees with consensus science about everything out of conspiratorial distrust and is constantly smuggling that stance into conversations all over the internet. Since it is very easy to present the state of affairs in a forthright manner, the only reason why someone would present them deceptively is (presumably) something like that. Hence knowing that the view is not mainstream is very useful for evaluating the motives of the original poster.
It's not evidence that they were lying, because that implies intent. No, probably they believe what they wrote. But it's evidence their ability to reason is suspect and possibly corrupted by some ideological motivation and so should be taken less seriously.
Not that I care, really, about any of this. Mostly this kind of antagonism is very frustrating and it's just kinda cathartic to try to shut it down.
My suggestion is that instead of engaging with commenters with your "oo! Logical fallacy! You broke the rules of arguing!" stance you instead try to find some way to productively engage with their actual thoughts. Perhaps figuring out why they said what they said instead of assuming anything you do not understand is a sign of a weak mind that needs to be corrected. You'll find people respond much more warmly to you if you do
You are confusing what "Appeal to Authority" fallacy is. Namely you are ignoring the fallaciousness of it.
The fallacy is where you use an authority in place of evidence. It is not fallacious to refer to consensus or experts.
Else, you end up basically in the "Do your own research"/vaccine denier/climate deniers/flat earth territory. Appeals to experts is not a logical fallacy. It's actually smart, because you get to leverage agreed facts (the earth is round) even though you've never actually been to space to see it for yourself.
One thing that is confusing about this write-up is that "DeepConf-low" is only mentioned once and in a screenshot, but it seems to outperform DeepConf-high for several tasks. I guess I'll need to read the underlying paper, but that seems troublesome.
> "Specifically, DeepConf-low uses top η= 10% (corresponding to the 90th percentile)
and DeepConf-high uses top η = 90% (corresponding to the 10th percentile) uniformly across
all settings. This threshold ensures that during online generation, traces are terminated when their
confidence falls below the level that retains the top η% highest-confidence traces from the warmup
phase."
I'm not sure if I'm parsing it right, but are they using "low" and "high" as descriptors of the number used as the %, meaning that the "low" 10 cuts anything outside the best 10%, while the "high" 90 leaves the best 90% ie high is less selective than low?
The confusing thing mentioned by the person you replied to is the data and naming from the actual paper, so no it's nothing to do with how the article was written. (Unless you're suggesting that the research paper was also written by an LLM, but I don't think you are?)
I don't think disclosure is necessary, but I think it can build trust in cases like this. "Please note that we used an LLM to rewrite our initial English draft." The reason to do this is that then people don't waste cycles wondering about the answer to this question.
I agree. Their LLMed English is much better than my Chinese.
Also, some of the very worst English I've ever read, has been technical prose, written by born-and-bred native English speakers with very high educational credentials.
Clear communication is important. The best idea on Earth, is worthless, if it can't be articulated well.
reply