From one machine analogy to the another... Mitochondria have been observed moving between cells. These are living organisms! Reducing their role to that of "microprocessor" seems suffocating.
The idea of comparing evolved structures to the current height of technology is usually an oversimplification. It's amusing to think back to when the switchboard was compared to the brain, for example. It wasn't that long ago that we considered the brain and a computer to also be quite similar, but we know it's only in certain aspects.
I remember reading a kids science section in the Sunday newspaper that explained the internet like our circulatory system, but instead of moving blood around it moved information around. (This was mid 90s)
A decade+ later I randomly picked up a Sunday paper and they were explaining the circulatory system "as if it was a computer network, but instead of moving data around, it moves blood around".
I thought it was funny that going from 95 to 2005, the expectation for innate understanding of biology vs digital technology had flipped.
It's also interesting because there are so many analogies you could use, off the top of my head I would say the cooling loop in car would be a much less "strained" analogy even if it's still very reductive.
it's called a model. all models are wrong but some are useful. the plugboard model of the brain was useful in the sense it suggested pieces working together and functional connections varying for different activities.
nobody is suggesting the brain runs an instruction set or a compiler - but it does compute in a general sense.
> Do you have some solution to this age old conundrum?
Well, as it happens, the person you are replying to did have at least an outline of a solution in this case, which could be stated thus: analogies may be regarded as models to the extent that they are useful for making predictions or testing hypotheses. it probably needs some refinement (for example, analogies carefully-crafted to "predict" the outcome in very narrow cases, but which diverge significantly with modest changes, probably don't count), but it is about as good as you can get in pinning down the semantics of ordinary language.
For example, a hydraulic analogy to electric circuits may be useful in simple cases, but once reactive elements are introduced, it is probably time to abandon it. Similarly, the 'bowling-ball on a mattress' analogy for general relativity cannot even predict the precession of Mercury, while some hydraulic and acoustic models have, apparently, been useful for better understanding black holes (even though they also spawned ridiculous headlines like "Scientists Close in on Creating Black Hole in Lab."[1])
In the case of TFA, I personally feel that the analogy being pushed is at best useless and probably misleading for any purpose.
Especially like bringing up the hydraulic example.
I was reacting to a common trope on HN to react harshly to any type of metaphor about the brain.
When, where is the outrage when describing 'voltage' as a type of 'water pressure in a hose'. As you say, it is only accurate to a degree.
Like giving a presentation on 'cloud' computing, and the power point has clouds, and some cohort of IT people are seething in the back of the room, "those aren't really clouds, they are servers".
Guess that is the problem, all analogies are only accurate to some degree, or else they would be the original. Nothing can be fully explained by something else, without both models being the same. So at some point, analogies are just helpful to communicate some concepts, but are not complete.
Yet, having something 'similar' can help get over some hump in understanding the 'new thing'.
That link is pay walled, do you have another copy?
Got it.
Nice.
So you are talking about taking this concept of 'analogies' to the extreme of an actual physical 'analogy' that can be studied. Here acoustics, to mimic the event horizon (i'm simplifying).
This is pretty extreme case. I was just saying generally, that metaphors and analogies can help explain concepts. This is finding actual real correspondence between different 'model's.
Were you saying earlier that there was a way to model this correspondence between models generally, for other cases? Maybe there is a category theory or something, to relate different models, so they can be analogies of each other, but both real so can have experiments?
I'm not thinking of anything as formal as that, more in terms of trying to get a handle on the difference between the words in ordinary language.
Metaphors, analogies and models often help us understand things, but they can mislead unless they are, to a reasonable degree, models of the thing we we are trying to understand. The physicist Matt Strassler calls the latter 'phibs', and has written a recently-published book motivated by one in particular, an attempt to explain the relevance of the Higgs field. https://profmattstrassler.com/2024/04/16/why-the-higgs-field...
Yes, this is good article showing the downsides of analogies. Thank You.
Interesting that while critiquing the common analogy, it did suggest another one.
I wonder if the problem isn't analogies, it is just finding 'good ones'.
"As an analogue, consider air pressure (which is itself an example of an ordinary field.) Air is a substance; it is made of molecules, and has density and weight. But air’s pressure is not a thing; it is a property of air, , and is not itself a substance. Pressure has no density or weight, and is not made from anything. It just tells you what the molecules of air are doing.
The Higgs field is much more like air pressure than it is like air itself. It simply is not a substance, despite what the phib suggests."
> When, where is the outrage when describing 'voltage' as a type of 'water pressure in a hose'. As you say, it is only accurate to a degree.
I was a bit sloppy in distinguishing model and analogy. I guess I meant physical analogy like pressure/voltage (which does outrage me!) vs model being an abstract description of the phenomenon built from the ground up, and was thinking in the context of physics or engineering.
The problem with the "bad" kind of analogies/models is they come with a list of conditions under which they don't apply, which is huge and usually not specified at all. For the "good" kind (Ohm's law/etc.), they come with a list of conditions under which they do apply which is finite and explicitly stated along with the equations so you can actually know if the model correctly describes a particular case.
Everything computes in a general sense, even atoms. But one could just as easily say everything is "just" mathematics because all models of reality are mathematical. In general I think it's important to be wary of totalizing ontologies and metaphysics of reality that reduce everything to a single universal substance (monadology) or activity like computation (computationalism).
It's a vacuous statement. All physical systems are computers because the logic of computationalism is circular. Everything is a computer so there is no meaning in the statement and the danger is that as more people start believing in the circular logic of computationalism they'll be more willing to delegate their cognition to computers even in cases where they should not, e.g. social media and algorithmic feeds designed to increase engagement and profits for advertisers.
Oh it was just an off the cuff comment. I was trying to be clever by finding a phrase that is the opposite of "the universe is a simulation". IE, the universe is "real".
In the articles I’ve read claiming that there is no free will, they seem to be assuming that “free” means that it would arise spontaneously without any interaction with the prior environment that could influence the decisions. They claim, rightly that that cannot happen but then assume that means that nothing short of that can be free will. That seems an excessively restricted definition.
See... but I would say that experiential consciousness is unexplained and impossible to measure, and yet, if I were to rank everything based on how certain I am that it exists, experiential consciousness would be at the top of the list. If you take away all of my senses, what am I left with? Just the ability to think and experience my own existence.
Well... you are certain that consciousness exists because you experience it: that is measuring.
As for explaining it, I guess it's more of a definitional problem than a physical one. I'm still confident that one day we'll be able to look at something and approximate how conscious it is.
Free will, on the other hand, requires supernatural phenomenons coming from outside the material world (dualism) to explain how we, humans, can make decisions free of any (or some) of the influences of the physical world. I don't buy it, and I believe the burden of proof falls on the ones who do.
> I kind of do believe in a deterministic universe, but I thought quantum physics disagree with that.
Not necessarily.
Most people learn the "Copenhagen interpretation" in intro quantum mechanics. The quantum wave function then evolves deterministically (Schrodinger equation), but collapses probabilistically (on observation). In that sense, the world is fundamentally random – although I personally wouldn't describe a robot driven by a real RNG as having "free will", even though it is indeed non-deterministic.
However, there is a major caveat here: This is just one interpretation of quantum mechanics, and there are fully deterministic alternatives. For instance, in "non-local hidden variable" and "superdeterministic" interpretations, the wave function collapse is believed to be a deterministic process. In the "many worlds" interpretation, the wave function doesn't collapse at all (it only entangles), and you're left with only the deterministic Schrodinger equation. It has also been shown that quantum-like behavior can arise in deterministic systems, see e.g. the "bouncing droplet" experiments on YouTube if you're interested (which is a beautiful macroscopic analogue to the "pilot wave" interpretation, which is a viable "non-local hidden variable" theory).
The reason it's commonly stated that quantum mechanics requires non-determinism is that (i) the philosophically most appealing "local hidden variable" theory has been falsified, (ii) the Copenhagen interpretation is easy to teach, and (iii) for practical calculations it actually doesn't matter.
Quantum phenomenons appear non-deterministic, but there is no definitive proof that they are not.
But I think this is irrelevant to the question of free will: having quantum states randomly collapse insinde my brain would have nothing to do with freedom.
That's just moving the problem around. Say the RNG is controlled by a soul of some kind; next up, how does the soul work? It's either random or an algorithm, and either way we're right back where we started.
Statements about free will aren't even wrong without some indication what's meant by the term. Typically it's not even a well-enough formed concept to discuss.
Metaphors help communicate ideas. I think everyone knows they are imperfect. Nobody thinks because we say the brain is a 'computer' that there are chips in there.
On other hand, scientist/philosophers that attempt to makeup 'new' words just to avoid this, sometimes make it worse and their ideas become even more obfuscated.
> It wasn't that long ago that we considered the brain and a computer to also be quite similar
Many people in tech still think this. It will take a while to dislodge the idea that if you squint enough, cells, brains, etc. are all basically computers, because it flatters the egos of computer programmers.
There are genuine connections between specific biological systems and certain ideas in computer science, but most of the casual analogies you hear from tech people only serve to inflate their own perceived authority ("I know a lot about computers, and cells/brains are like computers, so here is my take about cells/brains...")
Wet lab. It was horrible. 6 years of dishwashing and monotony. I write software now and am much happier.
The numbers were very difficult to reproduce and sensitive to tiny variations in experimental conditions, to the point of mystery. My MSc took me 6 years rather than 2, and that’s with a >90% scope cut.
I can say that there was a “comparable” proportion of freeMitos to mitoMPs in the samples I was studying, which were induced from 786-O and 786-O/VHL cell cultures using a variety of agonists (atypically low concentrations of A23187, compared to literature, and serum-starved cell culture medium, seemed to work best; usual concentrations of A23187 seemed very violent on the cells, which is relevant if you’re trying to ascertain anything about physiologic conditions). Shockingly, a very sizeable proportion (10-50% of EVs above a certain detection threshold, probably near 100-150nm) of total EVs were either mitoMPs or freeMitos.
I did not study mitoribosomes.
I was trying to optimize conditions for the generation of EVs (literature is lacking here), enrichment of samples, and flow cytometric techniques for measuring EV subpopulations. Mitochondria make everything much more complicated and even gum up the tubes, requiring special care. Experiments that measure EVs without controlling for the presence of mitochondria are IMO inherently suspect. Their prevalence will depend on sample types, but from what I’ve seen, I believe the default assumption should be that they are present in relevant amounts. I’ve observed similar subpopulations in other cell lines, but with fewer controls and validations (I was focused on 786-O cells).
There is a really great paper published by L Boudreau in the journal Blood with better results than anything I ever achieved, it’s very interesting stuff: https://pubmed.ncbi.nlm.nih.gov/25082876/
I don't know about that but mitochondria can and are modulated by a variety of factors and of course, they include dietary components.
Mitochondrial Modulators: The Defender
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9953029/
This paper refers to '61 mitochondrial modulators which are able to protect the mitochondria from toxic insults and/or improve mitochondrial function.'
It is my impression that extracellular vesicles are now considered to be one of the prime suspects in the aging process, including its potential reversibility.
Calling them a “prime suspect” is both true and the wrong way around.
They are ubiquitous and necessary to maintain normal physiological conditions / normal function, across all cell types and in all systems.
This is the reason why we are discovering that they play important roles in pathology: it’s because they’re an important part of the machinery, and things can go wrong with any part of machinery.
In many ways, they don't merely get attacked by immune cells, but also participate in immune function (and pathology).
At the same time, intracellular components found outside the cell, generally, tend to activate the innate immune system, in a similar way that foreign bodies do.
Some years ago, one of those "Consciousness Studies" or "Intelligence Studies" people unveiled a new model for thinking about the brain: as an "ant farm" for mitochondria.
The notion being that ant colonies work as large scale organisms, and can do some pretty complicated stuff. Well, the bigger notion is that the brain, viewed holistically, really does look like an ant farm when it's in action. I wish I could find the cite . . . oh here we go
https://www.quantamagazine.org/social-mitochondria-whisperin...