Science Books of 2019 | Hacker News Reader

Science Books of 2019

fivebooks.com

140 pointsAnon8428 comments

Perhaps a slightly different sort of science book, but I highly recommend "Foundations of Applied Mathematics" by by Humpherys, Jarvis, & Evans, the first volume of which was published in 2017, and of which the second supposedly will be published in 2019. Volume 1 is good overview of the Analysis needed in various applications, they give as examples "analysis, optimization, probability, statistics, differential equations, machine learning, and control theory". I found it useful as an overview of the theory one might need in a unified exposition, with a modern and pragmatic choice of topics and notation. Learning the topics previously from many books, all with different notations and ideas of pedagogy, can be confusing (to me at least), especially in understanding how different parts of the theory are interrelated. It even comes with computational "labs", the code for which is available freely online, something often lacking in books on numerical analysis, applied math, etc. My only complaint is that it somewhat leaves out differential equations. Definitely worth looking at! In particular the chapter on matrix calculus.

Thanks for this. Going to check it out.

How would you compare this to "The Princeton Companion to Applied Mathematics" ?

"The Princeton Companion to Applied Mathematics" is an overview of Applied Mathematics, some applications of different areas of it, brief introduction into problems and areas. More for curiosity.

This book is a textbook. It is "Foundations", good proper introduction, it does not go very deep, but it builds solid background. It's is actually going to be 4 books, so tons of knowledge.

Moreover, here: https://foundations-of-applied-mathematics.github.io/ you can find labs and materials in Python to back reading with practicing.

Thank you; The link is informative.

Like many classmates getting our PhDs, I considered my field, physics, the most fundamental of all the sciences, joking that we could derive the rest. That was the 90s. In today's world, I consider the most important role of anyone who understands nature to help spread understanding of our environmental situation, though I consider the most important task now to lead.

As far as science books, I'm going to share three titles I read this year I highly recommend, though they're about ten years old

- Countdown, by Alan Weisman, for covering (over)population, especially for sharing examples of nations that lowered birthrates without coercion or legislation, yielding greater happiness and prosperity.

- The World Without Us, by Alan Weisman, for illustrating our effect on the world colorfully and inspirationally.

- The Once and Future World, by J.B. MacKinnon, for illustrating how much more nature there used to be, how much we've lost, and how much more we stand to lose.

- Limits to Growth, the 30-year Update, almost 20 years old, but the most thorough treatment of us and our environment. Critics criticize it, but all the criticism I've seen actually criticizes straw men and other misunderstandings (if you know of meaningful criticism please let me know). Meanwhile, this paper https://sustainable.unimelb.edu.au/publications/research-pap... from five years ago shows their "business as usual" model continues to be unreasonably accurate nearly 50 years later, suggesting to me LtG's relevance and poignancy.

You can't go wrong reading any of these books, though I'd start with Limits to Growth.

There is an xkcd for that: https://www.xkcd.com/435/

Very disappointed with (pop-)science books these days. They're basically science history books. The formula is something like 60% (How it all started filler) + 5 % (An essay-worth of original relevant content + 35% (personal stories about conferences and papers).

Looking through some of them on the amazon preview, these seem to be in the same league. Meh.

Are there any science books (pop or not) you could recommend that don’t into this pattern?

Atom Land is pretty good, though there are some problems with the analogy. If you use it with the internet as a supplement and are a bit forgiving regarding bad analogies (as with PG essays), it's great.

The Eerie Silence by Paul Davies.

It does cover a history of SETI, but goes into a lot of the theory that is generally beyond a bit of what I've seen discussed before in popsci. I thought I knew the subject very well, but learned a lot of new things.

Some books I enjoyed in the past years:

https://www.goodreads.com/book/show/28256439-the-hidden-life...

https://www.goodreads.com/book/show/34272565-life-3-0

https://www.goodreads.com/book/show/19395553-our-mathematica...

https://www.goodreads.com/book/show/28116739-other-minds

https://www.goodreads.com/book/show/30555486-to-be-a-machine

https://www.goodreads.com/book/show/7972148-proofiness

https://www.goodreads.com/book/show/25817279-door-to-door

https://www.goodreads.com/book/show/23013953-gut

Thank you for sharing — the gut bacterial and two math books (proof ones and universal theory) caught my eye.

Other Minds, about cephalopods, is written by a philosopher, not a biologist or neuroscientist, so take it with a grain of salt[1]. I would also recommend _Octopus_ by Anderson, Mather, and Wood, who are actual biologists.

[1] Specifically, cephalopods are very intelligent compared to other invertebrates, but that's it. It's interesting because they represent a parallel evolution of greater intelligence outside the vertebrates, and because what they can do with that intelligence is greatly enhanced by their physical capabilities. Give a fish a human brain and it will have trouble twisting a lid open.

i highly recommend looking at the used market for the out of print but plentiful scientific american library series.

https://www.thriftbooks.com/series/scientific-american-libra...

the books are heavily illustrated, don’t shy away from the real stuff, but are still the most readable scientific publishing i have seen. whoever was the editor or manager of the series is a genius of publishing. john wheeler, julian schwinger, steven weinberg, peter atkins, and more can be found as authors. i am personally trying to collect them all.

History of science isn't "filler" -- you can't understand a scientific field today and the questions it is asking without understanding how it developed. A historical perspective is also useful because many students and early researchers get caught up in the issue of rival camps favoring Theory A or Theory B. History teaches us that more often than not, the reason for rival camps isn't that one theory is right and the other wrong, but that the experimental data isn't completely explained by either existing theory and that a new theory will come along that will cover both sets of cases.

Let me give you an example. Every pop-sci book I pick up about quantum mechanics is at least 30% the Bohr–Einstein debates and 'spooky action'. Getting tired of reading the same stories about Claude Shannon whenever I pick up a book which touches Information theory and of course, Newton's quirks every time I read a book which touches on gravity. It's boring, tedious and quite frankly irrelevant. If I want to read a book about the history and evolution of a particular branch of science, that's fine. But whenever I start reading a quantum mechanics book and it starts off with 'It was a sunny day in Athens and Epicurus was pondering..' oh.. fuck off.

I agree with you. Practitioners not knowing the history of their fields is a problem. In grad school, we learned the canon of managerial theory over the past century. Knowing the tradeoffs between different schools of thought makes me a better manager.

This is how I feel about most "science" writing. I thought I was the only one.

May I recommend:

https://www.amazon.co.uk/Oxford-Modern-Science-Writing-Landm...

Each chapter is basically that 5% "of original relevant content" on a different subject. So the whole book is "all killer no filler"!

I have a question about:

"Gender and Our Brains: How New Neuroscience Explodes the Myths of the Male and Female Minds"

I read "The Female Brain" by Louann Brizendine and all it makes sense to me but I am beginning to think that I (and this book) are wrong.

¿How is it possible that two opposing theses could be maintained in something like neuroscience?

I can imagine that it is possible in Economics, Philosophy, Psycology and other similar sciences but in neuroscience it does not make sense to me.

Gender and Our Brains cites studies and data that are selectively chosen to support a predetermined conclusion.

As to why:

> After almost 20 years of hearing the same invalid arguments (like Bill Murray in “Groundhog Day” waking up to the same song every day), I have come to see clearly that the real problem is a deeply ingrained, implicit, very powerful yet 100 percent false assumption that if women and men are to be considered “equal,” they have to be “the same.” Conversely, the argument goes, if neuroscience shows that women and men are not the same on average, then it somehow shows that they are not equal on average. Although this assumption is false, it still creates fear of sex differences in those operating on it.

A critique: https://quillette.com/2019/03/29/denying-the-neuroscience-of...

I'm curious: if the male and female brain, from a neuroscientific perspective, are truly the same - can we assume the book's author would be perfectly fine with all neuroscience from now on be done on male brains exclusively and the results then applied to female brains without further research? Somehow I doubt this.

Reading the book's blurbs and Kiser's descriptions I don't see anything about the book claiming m/f/intersex brains are all neurologically _identical_ in their developmental origins and trajectories... just that they aren't "gendered" (e.g. aren't structured in ways that cause tropes or stereotypes) and that there are many old myths related to the genderedness assumption in brains to be debunked.

The Amazon description indeed reads "A breakthrough work in neuroscience—and an incisive corrective to a long history of damaging pseudoscience—that finally debunks the myth that there is a hardwired distinction between male and female brains" -- but it's basically a 0% that the author wrote that bit!

If anybody ever says the brains of men and women are identical, the average brain has

> a volume of around 1260 cm3 in men and 1130 cm3 in women.

It's nonsense, spawned by fear that acknowledging differences will justify discrimination.

> How is it possible that two opposing theses could be maintained in something like neuroscience

It is very common in neuroscience (and also from experience psychology; but I'd bet also econ and Phil). Popular science books often paint results as ironclad, and gloss over small, or convenience samples, methodological issues that might cause researchers to disagree, etc..

Another big one in psychology and neuroscience (and many areas) is that even if you have a clear result, the interpretation often requires buying into certain assumptions. (E.g. interpreting activity in fMRI when no data from methods with better spatial or temporal resolution; intepreting non human animal studies)

What if neuroscience is an unknown problem space where boids are exploring? Technology, experiments, observations are increasing the problem space faster than it's being mapped.

What if theories of neuroscience are like product development? New bugs are being identified faster than they're being closed. So it'll be a while longer before the bug curve settles down.

What if science and engineering are like software design and architecture? It takes a while, plus lots of trial and error, to identify patterns, to generalize, to simplify. It's normal for code bases to "pulse", growing for a while as solutions are explored, then contracting for a while as dead ends are culled. aka technical debt.

Check out introduction to evolutionary informatics. Very readable application of information theory to evolutionary algorithms with deep implications.

Adaptive Markets: Financial Evolution at the Speed of Thought. I believe it is an excellent book written in a format that is accessible by almost anyone interested in economics. If you think that a random walk down the wall Street is a classic, then you should read this, since it suggests a model that expand on the Efficient Market Hypothesis.

Reply on news.ycombinator.com