Comments by "John Smith" (@JohnSmith-op7ls) on "Sabine Hossenfelder" channel.

The issue with climate change is that it’s really all about power and money. Whether or not it’s happening, whether or not humans are causing it, not causing it, or just speeding it up, how quickly a given scenario will occur, whether it will occur anyhow just further in the future and many other questions. They all get lumped together as “climate change denial”. When a supposedly scientific endeavor, with so many unanswered questions, decides it already has all the answers it needs to force everyone to do things which haven’t been demonstrated to be necessary, and nobody is allowed to question it, oh, and it just so happens to require expanding government power and tossing trillions of dollars to organizations which have close ties to governments… yeah, that’s not science. It may incorporate aspects of science, but it’s not science.
22
“The Universe” is just a term for everything. We’re in that everything. We can think. The Universe can think.
9
This is semantics. Oil extracted from the ground, for example, does not fit the traditional definition of fossilization. The organic material is not preserved in any tangible meaning of the word preserved, and it is not mineralized. Oil is considered a fossil fuel because it’s made through similar external conditions but the end result is not the same as fossilization. So really, at best, it’s overloading the term fossilization. I’d say it’s a misuse by whoever decided to use it for convenience to vaguely convey the process which leads to the creation of oil and coal.
8
Indian Company. Didn’t need to hear anything after that
4
@Stratosarge If you were knowledgeable about climate change and the many incorrect predictions made by it in the past, you would know we don’t have answers to all these questions. At best we have flawed models that give a rage of probability which gets less accurate the further out you go. And I didn’t even list any specific scenarios, so what kind of fool just says, “Uh, yeah, we have answers to that, no matter what the question is, because sciences knows literally everything.”
4
@somdudewillson And yet eventually nothing close to emulating human neurons, so it’s a moot point really unless you can prove that this sort of timing on its own has a significant positive effect without the countless other aspects of a biological network being accurately emulated down to the last detail.
3
Uhh, any language with the usual stable of logical operations can compute anything that’s in the solvable problem space. And “almost anything” is totally misleading as there are a lot of things outside the solvable problem space. So this isn’t saying anything interesting. Higher level languages don’t increase the things you can solve, they just abstract things so that the programmer doesn’t have to manually write them out.
3
There are countless other ways that digital NNs are nothing close to biological ones. Literally countless because we know so little about how the brain works. It’s nonsense to even compare the two.
2
We do know, bad training data plus gradient descent, which will always cheat when it can.
2
@andrezcabara2774 Oh right, politicians and big companies are in this for fun and altruism. And none of these universities are the slightest bit in bed with the companies and government agencies that fund so much of their research, or employ anyone who would put personal political beliefs ahead of objective science. Never seen any universities pushing politics at all. They are the epitome of integrity and morality.
2
A combustion motor on each wheel can and has been done.
2
They don’t rely on any objective point. They simply show where a middle ground is on a spectrum, a sort of average. An average doesn’t imply objectivity, it’s just a point of balance on a range of data points.
2
Tyson is a narcissist and has contributed nothing to science. He’s a talking head. Who cares what he thinks. This whole, “He’s an astrophysicist so he’s super smart, so he understands everything” nonsense needs to end. I’d call it an appeal to authority fallacy except 90% of what he talks about he’s not an authority in, and that’s being generous, assuming he still remembers most of the astrophysics he learned.
2
This isn’t a new issue. We’ve been dealing with older encryption algos getting broken after years or even decades of use, for a long time now. And large state actors have been mass storing internet traffic since the 90s, back when 56 bit keys were considered safe. All that data has long since been trivial to decrypt on conventional computers. So has 64, 128, and even some 256 bit keys. But quantum computers will not speed up brute force attempts by that much. Not enough to matter on longer keys of 1024 or higher. Attacks on the algo or other attacks are what are used. Bruteforcing, even with a quantum computer just isn’t practical on longer keys or at mass scale on mid sized keys. Anything old enough to be encrypted with a short key can be compromised but you have to consider if 30+ year old secrets being compromised is that much of an issue.
2
The regulations are often ridiculous and add nothing to the safety. Like irrationally tight tolerances on walls which not being exactly met to zero loss in safety requires them to be torn down and rebuilt at great expense. Endless NIMBY nonsense as well which, again, has nothing to do with safety. You have activists for profit who sue to block everything because that’s how they grift donations to pay themselves million dollar salaries. These things don’t add “a few pages” of paperwork they can add hundreds of millions of dollars to a project.
2
@sk-sm9sh There’s nothing here saying that this experiment produced code that was more efficient at solving problems much less on par with something like quantum computation that expands solvable problems. They didn’t even let the code do anything beyond some trivial things, they explicitly said they didn’t let it evolve into anything complex, and there’s no guarantee it would, or that if it would, it would do so with any scale of computing currently practical. You can’t compare the countless molecular interactions that happened all over the earth over billions of years to what you can simulate on any computer humanity could even begin to create.
2
I thought Germany wanted to power the country with Sharia law
2
Saying this mimics biological neural networks is just nonsense. First off, we don’t even understand how biological NNs work beyond the surface level. There’s over 3000 types of cells in the brain and we don’t understand how almost all of them fully work much less how they all work together. It was only recently that we learned the brain has a lymphatic system. Only more recently we learned the glial cells, previously thought to have nothing to do with cognitive function, is actually very important to it. All systems like this do is speed up digital NNs, their processing and maybe their training. Digital NNs are so primitive compared to biological ones that the man who coined the term neural network layered regretted it as it implied so much hype and misunderstanding. Digital NNs are nothing more than a bunch of simple functions which take an input value, do very simple math to it against some fixed internal constants set during training, then output the value to the next layer of functions. Essentially, they’re just a way of encoding an algorithm which can handle a lot of variability in the input quality. You could get the same effect by hand coding out countless logic branches if you wanted to. These are called expert systems. So, yeah, how our neurons and synapses and the supporting cells operate are far, far more complex not only in terms of processing information but in the fact they can dynamically change in response to thousands of chemical compounds, not to mention physically changing their structure. Even if you used something like a FPGA processor to dynamically change the functional pathways of the processor, this is a very crude approximation of what biological NNs and you still wouldn’t know how to emulate their function because we don’t know how they function. And don’t even get stared on how nonsense it is to claim a brain has the equivalent to X FLOPS. That just shows a complete misunderstanding of the topic. You can’t make a comparison to transistors when you don’t even know what the brain is actually processing at any given time. Simply taking an average of neurons and how often they fire or number of neurons and synapses and claiming that represents a workload capability which translates to transistors is comical.
2
If you stub your toe, it’s because of climate change
1
The other physicist was surprised you wouldn’t know basic thermodynamics?
1
⁠ @peter2024-n6d The purpose is always money and power. The right question is who is making money and gaining power off this. Because you don’t just happen to get corrupt governments, greedy mega corporations, and researchers pushing the same narrative, while the media just happens to label any questioning of the data, models, or interpretation of them as “conspiracy theories” from altruism and genuine concern for the good of humanity.
1
@kugeltmg Why would you even put any stock into predictions not only made with incorrect data, but by an org which knows its data is incorrect yet chooses to pretend it’s not and make proclamations anyhow. The Paris Agreement for example was always a complete joke. It was explicitly non binding, nothing more than a global virtue signal. Various other self imposed goals have failed to be met either on purpose or because they were unrealistic. The CCP never had any intention of meeting any of their goals, all for show. Climate researchers either have to know this or are incompetent to overlook all this, either way, their models are flawed. So the question is, if they know of these flaws and aren’t fixing them, or are ignorant of them, why are we putting any stock into the predictions?
1
LLMs are also trivial to dupe. They have no way to reason, no real world experience to draw from, no ability to physically verify a study. They’ll just blindly consume whatever they’re given and garbage in, garbage out.
1
@FalkEbert Maybe, it’s not like there is just one test/survey and who knows how often they’re being done. And you don’t need a to know how sentiment changes over time to get a snapshot of what it is at a given moment, those are different things being studied.
1
Fear mongering by con artist environmentalists, who were often working with the fossil fuel industry, caused the obscene over regulation of nuke plant construction and led to millions of premature deaths from pollution.
1
Lol what? Both of those things can be a problem at the same time, they’re not somehow mutually exclusive
1
@svenkrackow5550 Goal oriented? Setting goals through a reward system is a basic tenant of how most ANNs are trained
1
Most of the large LLMs available as a service are overtly biased against white people and this is easy to demonstrate by what it refuses to produce or in the case of images, produces when asked directly or indirectly to produce depictions of white people. Google’s Gemini is probably the biggest example of this, where it was shown that it had a system prompt to force “diversity” AKA, non-white people, when producing images. There are countless examples of this and the manager of the project was exposed as a far-left type with some disturbing social media posts. There was not subjective context here. They went out of their way to discriminate.
1
@c.augustin Opinion is not speech? Lol, I don’t think you know what speech means. And many have been fined and put in jail for expressing beliefs or simply telling jokes which the current government doesn’t like. They then use the media to try and justify and normalize this censorship and oppression. This is overt fascism.
1
@alansmithee419 What a sad straw man you built there. Nobody but maybe some irrelevantly tiny fringe of people deemed “conservative” is complaining that others are allowed to say their opinions are bad. The complaint is that they’re not allowed to have opinions without oppressive actions taken against them, up to being put in prison. And this whole nonsense about “You don’t have freedom from consequence”, well yes, that’s how free speech works. You have to fear being assaulted, killed, or put in prison simply because some zealots with power decide you’re a threat to their wealth and power. And those who harp on about how you’re not free from consequences are the first to whine the moment they’re exposed for their fascist opinions and actions. They cry about how it’s “racism” or “sexism” and “doxing” for anyone to disagree with them or not give them money. Every movie or tv show that flops because it’s pushing far-left rhetoric has someone involved come out and claim it’s some “ism” that they suffered the consequences of not being given money.
1
Anyone who isn’t overly trying to rig the results would take where the person is from into consideration. Sociology is a weak “science” but even they know that culture, economics, and other local factors are relevant. Unless, as I said, they are making fake science to use as propaganda.
1
@delavan9141 these talking head science content creators assume they know as much about everything as they do their field of science. Neil DeGrasse Tyson is the most obnoxious of them.
1
@BossAwesomeSauce Yeah, no, doesn’t make me look bad to point out that knowing one thing doesn’t magically mean you know everything else. Makes you look bad for believing that nonsense. And what’s the “average person” exactly? You implying that being a physicist puts you on a pedestal and everyone else is tossed into the average category? The people you’re putting into that category have actual experience in domains these science content creators love to delve into after a modicum of research.
1
@BossAwesomeSauce If she had, she wouldn’t have been wrong on this, would she.
1
@oliverwilson11 Again, playing semantics. Point is, the term inapt and arbitrary. Methane reclamation plants bury organic materials but we call those renewable. Geothermal requires digging up subterranean heat sources, we don’t call that heat source a fossil fuel. So oil is preserved, mineralized organic material? Strange, I’ve got a fossil of a leaf here and it doesn’t look like oil or want to burn. Probably because it’s an actual fossil.
1
@BossAwesomeSauce The whole video is a diatribe about how these two guys are wrong about some fossil fuels not being fossil fuels. It’s all a nonsense argument of semantics. And just because am I apt term is commonly used doesn’t make it inapt. So many dimwits misused “literally” that they added their use as a definition. So now we have a word that means the opposite of itself.
1
@petitio_principii Racism is a belief, AI has no beliefs so it can’t be racist. It can make incorrect conclusions based on race if fed incorrect data, but that’s not a belief. Also, these models will often produce very true conclusions about certain groups of people but people will call it racist because they don’t care about truth, only pushing propaganda and an agenda.
1
Well, either you believe the wild predictions of doom and accept it’s already too late to avoid it, only slightly reduce severity and duration, or you ask whether the wild predictions of doom are even accurate to begin with, and then you get labeled a “climate change denier” by the media arm of large corporations, governments, and the researchers they fund.
1