While the idea itself is worthwhile, I didn't see any explanation why these books and not others were selected, and no underlying structure to the choices made by the author. He also frequently lists several books covering the same topic without mentioning how they differ from each other.
I have previously seen two similar lists that are in my opinion better than this one:
John Baez rocks. In general. All of his stuff is great and his writing style will cheer you up when you feel like you need to go back to freshman calculus.
I read some of his pieces on Lie algebras and A-D-E classifications. Very helpful.
Baez is awesome. I used to follow his news-in-mathematical-physics digests back in the 90ies (basically a blog years before there were blogs); sadly almost everything went over my head. I dream of finding time and stamina to study enough background material to understand his Rosetta stone paper well.
Indeed, I should elaborate on the differences between the books. I'm quite keen to expand the post based on the comments relayed here. It's all very useful critique!
I also agree that those two lists are excellent - thank you for bringing them to my attention.
One important aspect I have neglected is Statistical Mechanics. I'm not familiar with that area, being more of an applied maths guy.
Mary Boas' "Mathematical Methods in the Physical Sciences" is also excellent, I'm using it for a course I'm teaching. At a level where a student is comfortable with calculus it presents an overview of the many areas needed to make graduate physics courses accessible. Although intended to be self contained in principle, it also might serve as a good jumping off point to further study in a particular area.
For general relativity, I really recommend Sean Carroll's book Spacetime and Geometry http://preposterousuniverse.com/spacetimeandgeometry/. If it seems like a long road ahead just take it one step at a time. You'll get there.
I personally found that after a solid grounding in tensor analysis, GR become far more accessible. The problem that many GR authors face is that they have to teach tensors simultaneously with spacetime principles. Quite tricky for a student to grasp on the first go.
Carroll's book looks interesting. I'll add it to the Amazon wishlist!
Something like this implicitly misses the point that one of the harder things to overcome when trying to do theoretical physics are the details of how research is done. This is the main point of having a PhD-level adviser and research group: to be able to get through the difficult parts that require intuition and very-specific domain expertise. These are not necessarily difficult or too technical, but they are obscure (e.g. how to get MadEvent and MadGraph up and running).
If all you care about is learning some physics, then this is great. But to really do theoretical physics the best way still seems to be to have an apprenticeship with someone who already does it.
Thanks for the kind words - although many people have commented that I've missed out some discussion on Statistical Physics. That is the first thing to correct.
As it happens, Physics is not my main discipline. I did Maths, then Aero (CFD). I did get by with enough Cosmology/Quantum Mechanics at undergrad to be able to pick up some advanced undergrad/grad level texts however.
I can certainly agree that some of the more obscure aspects of compressible flow solvers would have been very difficult to grasp, if I hadn't been able to question my supervisor on the topic.
I visited the MadEvent/MadGraph wiki. I can see it is related to particle physics and symmetry models, but what does it do, out of interest?
You've got to get some statistical mechanics on this list. One would have a very difficult understanding quantum field theory without at least some experience considering phase transitions in more familiar settings.
I used to be an experimental particle physicist, so I've certainly got some bias here, but I really think learning particle physics and the Standard Model is worth while. Halzen and Martin is a VERY good textbook and an excellent preparation for studying quantum field theory.
Yes, it is expensive. But it is very well written and thorough.
Also, I think it's always important to mention in these conversations that most physicists don't take string theory seriously. From the outside, it looks like the exciting frontier of modern physics. But that's more about Brian Greene's skill in marketing himself than string theory's explanatory value.
That being said, the Fabric of the Cosmos is a good layman's survey of the modern physics landscape:
Would it be possible to make such a list with free and/or libre books? The list they made is not extraordinarily expensive but not cheap either. Some are outrageous: an introduction to partial differential equations for 110 GBP?!
One is that these books are not computing hardware. If you buy them used and take care of them they won't depreciate quickly. Just sell them a couple years later. If you find you need them again several years after that, Abe Books/Amazon will find them for you again.
The other is: Theoretical physics is hard, evil hard for most people. Don't handicap yourself with crappy tools. The difference between a lousy book [1] and a great book may be hundreds of hours, it may be frustration with the material that makes you want to cry, or it may be the difference between learning the material and hitting an impenetrable wall.
I should also warn you that I and my colleagues in physics routinely found that we had to sample a dozen highly-rated books in a given subfield before we found the one that gave us the insight into our particular problems. I picked up that many books on semiconductor optics - and my better-read colleagues tried many, many more papers than that - before finding Coldren and Corzine, which is a revelatory text. (And, for all I know, out of print now. The hell of becoming a world expert in something is that you wind up being part of a global audience that would fit in one or two buses.)
---
[1] For you, at a particular time in your career, for a particular set of things you are trying to learn. Every book is is lousy to someone. The fact that there is no one-size-fits-all standard for books is just another reason why your physics education is going to require a bit of trial and error, fueled by a bit of cash.
I agree - some of the textbooks on there are extortionately priced.
I did put some free resources at the bottom of the article, although admittedly I could have added more. If you look at the link in the first comment there are some great lectures in there.
Leonard Susskind's Stanford Modern Physics lectures are fantastic. If you search on YouTube for Stanford/Susskind/General Relativity etc, you should be able to dig them out.
I didn't go through the whole list but I'd make sure to not miss out on the Feynman Lectures. Once you learn the material it's not really worth it to go back. They're nice because he gives you the story behind the equations.
Further down the road Laundau & Lifshitz is apparently the Bible.
I've been meaning to post my reading list for, maybe, 2 years now? I've never gotten it to a state where I'd consider it too helpful. Along with the list should come some comments on the books recommended: when to read them, how to use them, strengths/weaknesses i.e. rationale. It is silly to expect that someone will just go buy all the books you say and read them.
Basically I haven't posted mine so that the top-voted comment on my submission to HN doesn't read like this: http://news.ycombinator.com/item?id=1889946 ;-) (which I upvoted and entirely agree with, by the way).
Becoming a theoretical physicist was one of my aspirations after reading Stephen Hawking's A Brief History of Time (understood half of it, give or take) and being good at Physics in high school. Really appreciate the effort to categorise every steps and resources necessary to attempt this field.
But at the same time this is starting to seem daunting... how long would it take to go through all this material on a part-time basis? Well, my original plan was to do it after I semi-retire.
It's an amazing goal and you should definitely give it a try. My advice is start at the beginning and see how you get on. After a while you'll probably be further along than you realise.
Also, it depends on how much you really want to understand. If you want to be at the level of taking exam papers in all the material, then you would need to study a lot. If you just want a broad (but mathematical) introduction to the topics, you can go much faster.
To make things ultra-exciting, you can try and generate computer experiments around the material you're reading. Build a differential equation solver, use some Python libraries to visualise a pendulum swing, etc. It really makes it come alive and gives you a far better understanding of what is really happening.
Get 'Road to Reality' -- massive 1400 page book which builds up from basic algebra all the way to string theory. Unfortunately, it is still sitting unread on my book shelf. Completing that book is one of the must-do activities in my life!
I strongly advise against trying to learn physics from this book. Trust me, while the idea looks magnificent in theory, it's a horrible book to try and learn from for someone who doesn't have the mathematical and physical background. It just isn't possible, and yes, even if you're very, very smart and talented. Despite the fact that theoretically it's self-contained, you really need an equivalent of a PhD in math or mathematical physics to approach this book; but if you do have that background, there are better specialized textbooks.
The whole thing is basically a massive exercise in vanity. I wouldn't be surprised if it actually were the most unread book published recently, in relative terms (that is, the most percent of people who bought it didn't end up reading any substantial part of it).
Nope, no voting ring - I'm pretty surprised it hit the front page to be honest. In fact, I was debating whether to post it at all as it is obviously not startup/tech related. Then again, I'd seen a few other physics related posts pop up so I thought it might have some appeal.
Given that you're a physics PhD, I'd love to hear your thoughts on how the post could be improved. I'm keen to expand it and "fill in the gaps" that others have suggested.
I don't see the problem is that there are gaps, it's that it's just too much stuff. The reason it seems spammy to me is that it's a big dump of stuff that isn't well categorized.
To really get on top of all that stuff is a huge project, and the best way you're going to do it is by solving problem sets for a few years.
From my current viewpoint, I'd pick something really specific (say, modern models of quantum black holes) and try to develop the straightest line to it. Of course, I blew 9 years getting an undergrad and grad degree so I've got that base.
I have previously seen two similar lists that are in my opinion better than this one:
John Baez, How To Learn Math And Physics http://math.ucr.edu/home/baez/books.html
Gerard 't Hooft, How To Become A Good Theoretical Physicist http://www.phys.uu.nl/~thooft/theorist.html