Hacker News Reader

Comment by danbruc | Hacker News Reader

Tl;dr Biology is applied chemistry, chemistry is applied quantum physics and in some cases chemistry is a leaky abstraction and does not perfectly explain what is going on in organisms. Three examples - photosynthesis being more efficient than expected, birds sensing Earth's magnetic field by an unknown mechanism and a lack of understanding how olfaction works in detail - are presented but besides that there are no new revolutionary ideas in the article.

View on HN

> ... chemistry is a leaky abstraction ...

Well I second you on that. Even stronger:

All of (theoretical) science is a leaky abstraction. When an area is new, people are doing the spherical-cow-approximation willy nilly. After it gets established, corner cases start to surface.

Incidentally, theoretical physicists are the masters of creating spherical-cow-approximations and leaky-abstractions. They're always going for the next big unexplored territory. E.g., once QM got established as a theory, they moved on to high-energy, while there is about 1000x more work to be done picking up the pieces of QM in the areas of solid-state-physics and molecular-chemistry. (largely speaking that is; not everyone is like that).

A meta spherical cow?

relevant xkcd https://www.xkcd.com/435/

Math is a tool. It's orthogonal and used by all, not hierarchical above (or below) all of them..

http://euclid.trentu.ca/math/sb/misc/mathsci.html

    > The natural sciences investigate the physical universe but mathematics does not,
    > so mathematics is not really a natural science. This leaves open the subtler
    > question of whether mathematics is essentially similar in method to the natural
    > sciences in spite of the difference in subject matter. I do not think it is.

I'm not even including that famous Einstein quote about math.

Example: When you are doing math in economics you are not doing something that is even more foundational than physics.

what if the nature of the physical universe has no other reason to be as it is other than mathematical consequences of harmonies, resonances, balances, etc? In other words: if all that exists follows mathematical law, doesnt that make math part of the existential?

    > if all that exists follows mathematical law

As a thought experiment, try the opposite direction: Math mirrors reality, not reality mirrors math. I don't see electrons solving quantum equations - equations describe an electron. The cannonball follows a trajectory that we can describe using math - but it doesn't do math. Sure, you can say that it does, implicitly, but IMHO that's stretching it a bit, I don't see the value in such a metaphor.

From a certain level such questions become completely arbitrary and useless: Remember, it's your brain that interprets everything and you can just say whatever you want when there are no consequences.

> As a thought experiment, try the opposite direction: Math mirrors reality, not reality mirrors math.

If this was true, it would severely limit the usefulness of math. We'd see our predictability break down much sooner than we do actually. Read up on it here:

http://plato.stanford.edu/entries/mathphil-indis/

Why do you interpret my post as an absolute?

http://dilbert.com/strip/2015-06-07

I said - and I have no idea why I have to paraphrase my one sentence - to try a thought experiment, to see what happens when you reverse the direction.

Not to mention that you didn't really say anything? I have no idea what your point even is.

We don't have to make this thought experiment since we already know that it isn't the case - reality is a subset of what's mathematically possible. Finding that subset is what physics is concerned with.

Were you replying to me? I actually upvoted you, if it's meant to be a contradiction than I don't see it. I only just realized you may have been answering me since you used my words, but other than that I have no idea if that's true.

In any case, I should have read more carefully, because

    > reality is a subset of what's mathematically possible

does not mean anything. It's a grammatically correct sentence, sure, and that's all. The number of (grammatically) correct sentences is definitely larger than the number of useful ones though :)

let's just say it was a bit late in my timezone ;-).

I probably replied to the wrong post, but I wanted to say that math is in fact a tool for science - instead of being the most fundamental science itself. I've met some people, including mathematicians, who think that because everything can be described with math, it means that math is all that's needed to find the truth about reality. My point is that the process of appying math to find the model that best fits our own world is something that's outside of math itself.

Alternatively: mathematics can be seen as the science of structure.

http://web.maths.unsw.edu.au/~jim/structmath.html

That sounds like a pretty solid method for making progress. First outline something with spherical cows and refine the theory as data that doesn't quite fit keeps accumulating.

I was under the assumption that the working theory is birds share a similar feature as fruit flies and possibly foxes. The cones that allow us to see blue light actually work as compasses when suspended in a fluid, and they are not attached to the retina of a fruit fly but are suspended in their eye and thus shape incoming visual patterns around magnetic fields

https://en.wikipedia.org/wiki/Magnetoreception

That's fascinating, but birds do have their rods and cones attached to the retina. IIRC birds are though to be in some way detecting either directly, or through secondary effect, polarization of light.

That is essentially the proposed mechanism discussed in the article but the article gives the impression - at least to me - that it is not yet confirmed that this indeed how it works in birds.

EDIT: After skimming the Wikipedia article, both articles agree that it is a possible but not yet confirmed mechanism.

The cryptochrome hypothesis doesn't require detachment from the retina as it's a protein that has directional excitatory preference, not the whole cell.

Interesting. Thanks for the clarification. I find this topic to be of extreme interest

Good summary. Classical physics is not good enough for chemistry. So why are we surprised that without a quantum foundation, biology cannot function at all?

OTOH, we have a theory that classical computation adequately accounts for intelligence. Finding an unavoidable quantum route that nature has picked would certainly be interesting, despite that quantum Turing machines are also classically computable (I think).

>OTOH, we have a theory that classical computation adequately accounts for intelligence. Finding an unavoidable quantum route that nature has picked would certainly be interesting, despite that quantum Turing machines are also classically computable (I think).

I don't think we have a foundation for classical computation without thermodynamics, which is a classical theory whose ultimate foundations probably require quantum information theory.

Can you provide an example of how this rather facile assertion leads to any insight into biology? For example, how does it follow in a coherent fashion that the chemical properties of DNA lead to Mozart, or a lady bug? In short, how does chemistry explain either the origin, function or future trajectories of any of the utterly amazing biological entities that exist today?

I mean... evolution explains the complexity of things, and evolution certainly has a foundation in the chemistry of DNA. Or are you asking why we have the particular configuration of biological entities we have instead of a different one?

Well what does it mean to say that Biology is applied Chemistry? To me it says 2 things. 1 - That there is no 'new' knowledge in biology. 2 - The important aspects of biology can be reduced easily to applied chemistry.

Evolution contradicts this, because it is a genuinely new idea. Some may say it is the ONLY new idea in biology, but that's good enough.

It's often quoted here that "Computer Science is no more about computers than astronomy is about telescopes." You could extend this to "Biological science is no more about chemistry..."

>in some cases chemistry is a leaky abstraction

There has never been "classical" theory for chemistry like there has been classical mechanics that works reasonably well within limits, just weird heuristic rules. Chemistry is quantum chemistry is quantum electrodynamics.

Or: If it can be optimized/exploited by evolution, of course it will be.

Reply on news.ycombinator.com