"I still get the occasional joke from colleagues about my ‘crackpot consultant business’, but I’ve stopped thinking of our clients that way. They are driven by the same desire to understand nature and make a contribution to science as we are. They just weren’t lucky enough to get the required education early in life, and now they have a hard time figuring out where to even begin."
In some sense, labeling somebody a "crackpot" assigns a sort of malicious wrongness to them. It is interesting to see somebody who's dealt with a significant subset of these people and discovered more actual, honest misunderstanding than we (or at least I) would have expected.
NNs do not model the brain, and have basically nothing to do with it. I'd imagine rereading that chapter would be a good idea.
But above all, "pseudoscience" is something different and is certainly not the right word to describe this issue of terminology.
The definition of pseudoscience is "a collection of beliefs or practices mistakenly regarded as being based on scientific method."
Arguably the idea that neural networks reproduce how brains work is pseudoscience.
In the very least, it's a deeply misguided marketing blurb. It's now considered a metaphor, just like the ones driving some corners of the artificial intelligence field.
Moreover, you're dismissing an entire extremely active research field, which tries to understand up to which point artificial and biological NNs are or aren't similar. It's the field, for example, of Turing laureate Yoshua Bengio.
Neural networks CAN reproduce how the brain work (if you're doing computational neuroscience, in which case what you call NNs is something different, and it's an issue of terminology). Even simple binary NNs were originally born to understand the cognitive functions of the brain (McCulloch and Pitts, 1943). The field later diverged (with the advent of backprop), but still today they have a lot in common even inadvertently, for example the representations they learn in navigational tasks (Banino et al. Nature 2018) or in the visual cortex vs. CNNs for computer vision.
The terminology is now fixed and anyone who has more than a passing acquaintance with these things knows that neural networks have almost nothing to do with biological neurons.
either way, the technique has absolutely nothing to do with the biological cells we call neurones -- as much as decision tress have to do with forests.
It is metaphorical mumbojumbo taken up by a credulous press and repeated in research grant proposals by the present generation of young jobbing PhDs soon to be out of a job.
The whole structure is, as it has ever been, is on the verge of a winter brought about by this shysterism. Self-driving cars, due in 2016, are likewise "just around the corner".
ANNs don't yet model the structure of the brain but it seems plausible that they could do in the future as the result of some "convergent evolution".
ANNs have a fair model of individual neurons. Artificial and biological neurons do roughly the same thing when evaluated, but they are connected and trained very differently.
For me it's too much of a coincidence that the two most generally intelligent systems (ANNs and BNNs) are both "linear networks of activation functions".
We have not managed to build general intelligence from any other formalism, and neither has nature.
Viewing ANNs as a poor model of BNNs may be looking at the question backwards. You could say that BNNs are trying desperately hard to model the pure mathematics of ANNs within the confines of biochemistry. The fact that a biological neuron is not exactly a ReLU may say more about the limitations of biology rather than the limitations of ReLUs.
I am unconvinced that "linear network" appropriately defines BNNs. Can you clarify this?
They most certainly do not. The idea that biological neurons are a kind of programmable on/off switch is completely wrong. There is plenty of computation happening inside a single neuron.
The fact that we once thought only the interconnect of neurons is important, and not the individual neurons themselves, is actually pretty strange. We know full well that every single cell is capable of computation, as it must react to its internal and external environment to be able to function. Neurons are more specialized for computation than other cells, and of course the biological neural network is a still a huge part of animal intelligence, but the simple model where we ignored what was going in inside each neuron should have been seen as extremely unlikely to be the full picture from the start.
Light-ether is real. It's [very oversimplifying] the EM pressure from maintaining every EMF-free zone (the insides of particles superconductors), plus pressure of all the cosmic background radiation.
It's like cargo-cult physics. They (try to) do all the things, but the starting point is wrong, so it can't work.
So showing them "where to even begin" is in fact the solution... if they'll listen.
This is where I find the line easier to draw. Starting from a wrong premise and reaching a wrong conclusion (even if reached rationally and logically) does not, itself, make a crackpot. The crackpot is the one who won't listen to the counterarguments. And when they do listen, they just shift the goalposts instead of actually adjusting their basis to better fit reality.
A looser, less perfect, but more easily applicable and economical heuristic that I have is:
- Is my (layperson) interlocutor mostly contrarian and skeptical of others? Then there is very little chance that trying to point them in the right direction will be productive. Being a reasonable contrarian requires a very high level of expertise.
- Is my interlocutor mostly curious and skeptical of their own understanding? Then my friend, this person is a rare golden perl, and I will spare no effort or patience, because in almost every case these people are the ones who haven’t lost the capacity to learn.
The difference is that in P. the metaphors and mappings create consistent experiences for everyone and can be manipulated symbolically using math.
That's a huge change, which is why science is everywhere and hardly anyone attempts alchemy today. But under the hood science is still a machine made of metaphors which are partial mappings of experience.
I don't think anyone believes all the different metaphors physics uses mesh together neatly, so at least some of them must be misleading. Improving the coverage is almost the definition of physics.
But it's also possible the consistent symbolic metaphor approach is limited in ways we don't yet know, and at some point it will reach its limits and have to be replaced with something else.
Newton is the father of modern physics to us because we have perspective that he lacked. At the time he was working—in the moment—it wasn’t yet clear how the universe worked.
His trial and error gave that perspective as a result. Somebody had to waste time on alchemy to then succeed in physics, the time spent on alchemy was not a waste.
Anyway, Newtons contribution with Principia was to collect facts known in his time and put them into one unified mathematical framework, he didn't actually discover any of "his" laws (as predicted by Stigler's law).
Call me a crackpot, but didn't it?
Newton is forever remembered as [one of] the granddaddy[ies] of physical sciences.
Alchemy, from any initiated source seems to be a series of rituals designed around the refining of the self into its full potential, rather than a material goal of becoming wealthy by being able to produce gold. That is, it's metaphorical.
I haven't read anything by Newton on his alchemy, but if I were to take my rudimentary understanding of mystical thought and read it against Newton's example—I would think he were quite successful having produced such novel mathematical and scientific breakthroughs—whether he was aware of that or not.
If there was gold to be found in a human experience and work, I'd say he produced endless amounts of it...
edit: I'm not making it up. Whether or not you want to believe in whatever, this is what alchemy is to magicians like Newton: https://en.wikipedia.org/wiki/Hermeticism#Alchemy
He didn't endeavour to just turn physical lead into physical gold. I know we shit on humanities around here, but facts still exist in that sphere.
Further, this isn't my opinion but that of Keynes - https://royalsocietypublishing.org/doi/10.1098/rsnr.2012.005...
> Keynes underlined the last sentence of this passage, which could easily substitute for his own assessment of Newton's use of experimentation in ‘Newton, the Man’, that ‘His experiments were always, I suspect, a means, not of discovery, but always of verifying what he knew already’.13 Keynes repeatedly referenced Newton's ‘intuition’, calling it ‘pre-eminently extraordinary’, almost to the point of relying on mathematical proof and experiment only as a matter of social convention, rather than as a means of revealing some insight that had not already occurred to him. The mathematical proofs in Principia were alleged to be ‘dressed up afterwards—they were not the instrument of discovery.’
https://www.openculture.com/2016/11/videos-recreate-isaac-ne...
I'm teasing at proposing, after Keynes, that maybe his alchemy worked, and he just didn't understand how.
But there's no salvaging the storybook version of Newton here. Newton was many things, but the innocent, curious monkey sitting under an apple tree he was not. (well, he was—at least insofar as we all are, anyway)
It pidgeon-holes and ridicules people, we know it's insulting, and we use it anyway - that seems close enough to malicious.
Needless to say the experimental results differed slightly from their theory. But what struck me is, they were actually doing science .. which is wonderful. I remembered thinking if only our local schools would do a little bit of science, or math for that matter..
1. There are labs / practicals as part of a scientific curriculum, though ultimately in a controlled fashion.
2. There are projects that allow a student to pursue an open-ended like of inquiry.
I'm not in the academia myself but I can totally see how the nth time someone offers their "insights" into something they almost certainly don't understand could make you think of them as crackpots.
Also it isn’t just physicists on the receiving end. As a software engineer I get people wanting me to help them with their “crackpot” business ideas (although they sometimes don’t want to tell me their secret, and I heartily encourage them not to!). I have seen an acquaintance explore a crackpot electrical mechanism for perpetual energy. Mathematicians surely get their fair share of crackpots too!
It turned out to be right, of course.
Sometimes entire fields get stuck in bad equilibria and need to be disrupted.
Theoretical CS is of course niche enough that it probably doesn't attract mainstream people from outside of the field, so I suppose the field has its own crackpots. I suspect mathematics would be similar except somewhat more mainstream.
Of course some day someone will probably come along and actually prove something about those open questions but claims of proofs are much more common than actual proofs.
As for crackpot things within the academia itself, computer science also has areas such as AGI that are so hard to get actual scientific results from that it might be easy to deem it as just handwaving rather than actual research. Trying to build general AI with automated reasoning and knowledge bases back in the early days may not have been exactly a crackpot idea -- if anything, it was overly optimistic about hard logic -- but it certainly wasn't realistically going to succeed either, at least in hindsight.
You would think this would be empirically verifiable but I nor others could make heads or tails of his argument.
When I saw the guy also had a theory of unified physics, I gave up.
Still don't know, he might have been onto something.
We'd be better off with a theory of everything designed by a neural network, at least people wouldn't belabor under the mistaken assumption that "theory of everything" actually corresponds to anything real.
Anyway, a string theorist on the faculty doesn't cost any more than a mathematician. For the others you are continually having to pay for LHCs, nanokelvin freezers, and optical tables.
That's exactly the problem - instead of trying to produce new theories, people keep researching string theory, in massive numbers (relatively to the size of the field of theoretical foundations of physics, of course).
Also, there are a few other interesting avenues to explore - coming up with a relativistic version of de Broglie-Bohm pilot wave mechanics (or as close as possible to get, given Bell's inequalities), for example.
If the string people were putting all their effort into coming up with a testable prediction, we might be better off. But they are mostly trying to show their untestable thing is better than the next one over, instead.
I don't have a solution... I do have a problem in that this drive-by dismissal of mine is getting attention when what I was excited about in another comment goes ignored. =\
Isn't that explained by the GP's point? Some things are much easier to engage with than others, regardless of whether it's for or against. Your statement above is two sentences, and while it undoubtedly accurately describes some occurrences, it also likely doesn't describe others. Without statements to clarify how absolutely you intend it to apply, people are free to make their own assumptions and if that assumption casts your statement as absolutist, it's easy to refute.
Communication can be hard, asynchronous communication even more so.
The article touches on this in how the author quickly deduces who is one of them and who is not. Perhaps therein lies the entire issue at hand.
It's not a truth any more than 'humans murder others when they're angry'. Sometimes humans belittle others and sometimes they gain status from it. Sometimes they lose status, and I suspect they almost always lose a little self-respect and feel more intensely the shame that was so strong, they just tried to project it onto someone else.
> Being highly educated is apparently no defense against this common weakness.
I think it depends on the fields of study. If you study humanity, you may understand this behavior. If you study quarks, I agree it is little defense.
I mean -- as a programmer, my significant other thinks that my presence soothes the Machine Spirits and makes the computer magically cooperate. Imagine if this was a superstition! The random requests for IT support would be really annoying.
Quiet a few of the misinformed are HN commenters.
Generally, the pattern appears to be "Foo is complex. Conspiracy theorist asserts there is a conspiracy making Foo complex. We cannot trust the mainstream experts because they are lying. Gishgallop with endless bad faith questions which take an enormous amount of time to answer properly but ultimately do not convince the question asker".
The general takeaway, for me at least, is that even the most intelligent person can be sucked into crackpot conspiracy. The problem is "I'm an expert in this field, this means I'm smarter than most" and then thinking that intelligence in one area grants you insights into others.
The worst part is conspiracies like to hide under the motto of "I'm just being a good skeptic". However, a good skeptic needs to actually dig deeper than just surface level assertions. A good skeptic needs to look into the counter claims to their conspiracy. What are the sources? Do the actually point to data or are they just naked assertions? Do the sources reference only other conspiracy websites? Do they have references? If this claim is actually true, why is it rejected by the mainstream? Who is actually benefiting from hiding the truth?
I love this. Also hilarious.
In my experience, also true in certain cases.
You observe something that is being done a way you would not do it. There is no reason to suspect this is the source of the problem, but you do it the way you think it should be done. Lo and behold, the process works and the problem is gone. In retrospect you can see why, maybe the filename ':/%20' was bad luck, or whatever. Looking at the process, though, it's hard to claim success was achieved by a rational scientific process...
They're both definitely in my favourite author list, of any genre.
The naive enthusiasm, self directed exploration, and passion many of these crackpots have are winning traits in many pursuits. Just a bit of poor luck they picked an especially high wall to scale with theoretical physics.
It's why they rage at people not taking them seriously - if anyone tries to show something mathematically they can't understand it, don't want to admit it, and either wander off or ignore it and go back to the extensive prose.
Usenet science boards used to have various "personalities" who were clearly ill. They would post variations on the same word salad over and over, invariably claiming that Einstein was wrong about something or other, while they were right, which made them super-geniuses.
I understand high-profile physicists particularly tend to attract these people, so dealing with them must be a challenge, and there must be a temptation to skip over anyone who looks like they fall into that group even if they're just asking questions.
There was - possibly still is - a crossover group typically made of electrical/electronic engineers who knew just enough traditional EM theory to be dangerous, but not enough to know what they didn't know about GR and QM. For some reason they were also obsessed with trying to prove that relativity was wrong - perhaps because GR is so weird it's almost offensively strange to people who want to live in an un-weird universe.
My contribution to this severely under-served department:
"It can't possibly be correct to use relativistic field theory to write effective fields for excitations of a wave medium when the whole idea of SR was that there wasn't a medium."
"The introduction of a “luminiferous ether” will prove to be /superfluous/ inasmuch as the view here to be developed will not require an “absolutely stationary space” provided with special properties, nor assign a velocity-vector to a point of the empty space in which electromagnetic processes take place."
In 1920, after developing GR, his views were considerably more nuanced, as expressed in his 1920 lecture at the Leyden University, "Ether and the Theory of Relativity":
"Recapitulating, we may say that according to the general theory of relativity space is endowed with physical qualities; in this sense, therefore, there exists an ether. According to the general theory of relativity space without ether is unthinkable; for in such space there not only would be no propagation of light, but also no possibility of existence for standards of space and time (measuring-rods and clocks), nor therefore any space-time intervals in the physical sense. But this ether may not be thought of as endowed with the quality characteristic of ponderable media, as consisting of parts which may be tracked through time. The idea of motion may not be applied to it."
Einstein, Albert. Sidelights on Relativity . Public Domain Books. Kindle Edition. (0.99$ on Amazon)
That lecture, by the way, is a superb example of presenting the frontier of 1920 physics - before Schrodinger, Dirac, and modern field theory - without using any mathematical notation whatsoever.
On the other hand it also gave us marvelous energy sources such as nuclear power. However this connection is often missed amongst lay observers.
We weren't getting FTL anyway.
I don't believe anybody would have expected the western allies and the Warsaw pact countries not to go to war with each other had nuclear weapons not existed.
This will unfortunately remain true only until a nuclear war does break out. After such a moment, the few that survive may say "if only nuclear weapons weren't invented, we could have had another WW2 instead of this disaster".
The other arguments about nuclear physics that you make are much more convincing as an unmitigated good.
Ahhhhh, those were the days.
I think this is why many profs have a secretary to screen this stuff out
https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect#...
The D-K effect is: "The Dunning-Kruger effect is defined as the tendency of people with low ability in a specific area to give overly positive assessments of this ability." That is exactly what the parent described, and is an empirically identified behavior. The criticisms are aimed mostly at the "meta-cognitive" and bias-based explanations for the behavior.
> GR is so weird
Someone said about Quantum Mechanics that if you understand it, you obviously don't. Because it is so weird.
“Science popularizers often use real-world metaphors to bring specialized knowledge to the public. This was not the case with Feynman. When he delivered that famous quote during a conference at Cornell University in 1964, he was trying to convince his listeners not to try to understand his explanation ‘in terms of something familiar.’ Instead, he announced that he would simply describe how nature works, inviting those present to ‘relax and enjoy it.’”.
The same can be said for physicists and mathematicians who know just enough EE. As a practicing EE who specializes in RF and antennas, I've had to debunk many crackpot antenna and circuits, who developers are looking for funding. The nice thing about RF is hunks of dielectric and conductor are pretty easy to fabricate or simulate, but they don't even try to prove their ideas.
If you are ever even around physics or math at all, you will see the crackpot letters. I was a little surprised engineers are largely unaware of the phenomenon and are shocked, for example, that anyone seriously tries to make perpetual motion machines at all.
Here's a couple links about the mathematical equivalent:
The mathematical physicist John Baez proposed a “crackpot index”[1] [snip]
Mathematician Chris Caldwell was inspired by Baez’s list and devised a mathematical version. Some (lightly edited) examples from Caldwell’s list are
1 point for each word in all capital letters;
5 points for every statement that is clearly vacuous, logically inconsistent, or widely known to be false;
10 points for each such statement that is adhered to despite careful correction;
10 points for not knowing (or not using) standard mathematical notation;
10 points for expressing fear that your ideas will be stolen;
10 points for each new term you invent or use without properly defining it;
10 points for stating that your ideas are of great financial, theoretical, or spiritual value;
10 points for beginning the description of your work by saying how long you have been working on it;
10 points for each favorable comparison of yourself to established experts;
10 points for citing an impressive-sounding, but irrelevant, result;
20 points for naming something after yourself;
30 points for not knowing how or where to submit their major discovery for publication;
30 points for confusing examples or heuristics with mathematical proof;
40 points for claiming to have a “proof” of an important result but not knowing what established mathematicians have done on the problem.
[1] https://math.ucr.edu/home/baez/crackpot.htmlEdit: The second link has a scanned advert at the end from 1983 for “The Science of Programming” by David Gries: the title of the book amuses me.
> 10 points for not knowing (or not using) standard mathematical notation;
I personally hate non-standard notation so I switch to the standard when I know about it, but at first, when moving out of well known waters you may be using non-standard notation.
> 10 points for expressing fear that your ideas will be stolen;
Ha ha, I don't express fear, but who is not afraid a little?
> 10 points for stating that your ideas are of great financial, theoretical, or spiritual value;
A lot of papers I have read starts with this actually, not to mention PhD or MSc thesis.
> 10 points for each favorable comparison of yourself to established experts;
Most papers boast about how they beat the baseline that everyone uses. I do it myself although I try to be honest about the circumstances when this happens and when it's better the alternative.
> 30 points for not knowing how or where to submit their major discovery for publication;
Sometimes it happens
> 40 points for claiming to have a “proof” of an important result but not knowing what established mathematicians have done on the problem.
It has happened lot of times, since the beginning of time that some people rediscover previous results, sometimes famous people on famous and important results. To put an example the FFT, but there are a lot of them. And besides, who likes to read others people code?
I think the difference between a crackpot and other people is not accepting the mistake. It's curious it happens so much to engineers. I'm an engineer myself and if I learnt something during my studies is how little I actually know. I usually say that one thing STEM education gives you is the certainty that most of your ideas are wrong.
There were like 10+ physicists and 10+ grad students in this meeting, they respectful when they smoked this time crystal guy, but of course he just used it as an example of the cabal keeping the truth down.
They fell into a couple easy categories. Perpetual motion machines, casual ideas that are obviously impractical, fundamentally failing to understand the subject area. Not a single one of them seemed new or useful.
Then again Nicola Tesla had some strange ideas. Turns out half of them were brilliant and the other half still seem weird. Efforts to demarcate crackpot and real might be approaching the problem from the wrong direction.
I like the author’s strategy of helping people gain the knowledge to understand what they’re trying to talk about. Teslas ideas about motion and flow as it pertains to physical health were weird but that doesn’t mean he was unable to learn about biology, had he spent time learning about it, maybe he’d have brought his ideas back to reality and contributed something interesting.
We should celebrate people for their curiosity, encourage them to gain the knowledge that has already been discovered and help them express their ideas in ways that others can relate to.
In addition, I believe the article demonstrates that in these cases, the facts people are discussing and debating are only part of the picture: there is an emotional undercurrent to what they are saying.
Dr. Carl Rogers would likely argue that facts would only change their minds if they felt properly heard / seen / understood.
For example:
> One summer I went to visit three trisectors... My first trisector was... enormously pleased to see me and couldn't stop talking. He was bursting with energy and couldn't keep still... We talked about his trisection and I tried (I was younger then, and not as wise) to show him the error of his ways. He seemed to be listening to what I was saying, but none of it was making any impression.
I'm not saying "feed the trolls," only that there is more going on than meets the eye.
While pop-sci has been radically successful at making physics concepts appear accessible, it has woefully failed at making actual physics accessible. It's no wonder that Americans increasingly view science as either the tony stark like practice of wizards - or as a ceremonious endeavor practiced by those whose beliefs are no more testable than a TV pundits.
We need to do better at Science writing.
The median person who can drive, pay taxes, be a loving partner, and parent is just ... disappointingly not smart if you've been surrounded by brainiacs for most of your life. Some very bright people have other personality quirks that make it hard for them to integrate new ideas outside their comfort zone.
We need better people.
There are some folks who are incredibly motivated and also fundamentally incapable. But for most people, the biggest "issue" is that there are multiple lucrative and lower-stress exit ramps.
Learning higher maths and physics is not intrinsically more difficult, but it is hard for most to maintain focus on a topic with only vague applications to anything and leads for most to financial ruin.
Software skills are more common merely by virtue of how applicable they are.
Given that the income delta between a researcher at CERN and a professional software engineer at Google is >10x I'd be skeptical that this selection always translates to intelligence. Anecdotally, many PhD Physicists transfer into software for monetary reasons and do not come across as intrinsically smarter than the average engineer of equivalent experience.
I would start with most terms named after their inventors/discoverers, trying to give them meaningful names.
> [...]
> We need to do better at Science writing.
Rather: we need to get better at not believing what advertising tells people.
To a certain extent, fundamental background knowledge, in particular mathematical, is needed. So to present something to a general audience will necessarily involve a simplified model of some kind. This can be a bad analogy, but even a "good models" are flawed at the edges and seem likely to be the simpler the model becomes in relation to a more complex topic.
Also saw it around COVID vaccines. I talked to several people who claimed calling it “vaccine” is fraudulent because in their minds a vaccine protects you 100% from the disease and if it only avoids hospitalization or severe cases it’s not a vaccine.
How do you mean? I think of popular science writing as always being for the layperson.
"A typical problem is that, in the absence of equations, they project literal meanings onto words such as ‘grains’ of space-time or particles ‘popping’ in and out of existence. Science writers should be more careful to point out when we are using metaphors. My clients read way too much into pictures, measuring every angle, scrutinising every colour, counting every dash. Illustrators should be more careful to point out what is relevant information and what is artistic freedom. But the most important lesson I’ve learned is that journalists are so successful at making physics seem not so complicated that many readers come away with the impression that they can easily do it themselves. How can we blame them for not knowing what it takes if we never tell them?"
Of course, this writing often isn't for the layperson, it's for an audience who can tell the difference between diagram, artistic license, and metaphor, but even so, it's good to think about, especially when communicating science to the general public.