# I just want to be able to read physics equations

1. Nov 1, 2014

### bugmagnet

I am a layman in physics. I am a programmer.
I am not looking for a career in physics.
But I do want to understand physics.
I know it's not really possible if I can't read the math.
I feel that if I could just read the equations, ie at least know what the symbols represent then I could begin to sit with them and begin to actually understand them.
Like reading philosophy texts... You don't necessarily have to understand it on the first read, but you begin by first at least knowing the language it is written in.

So I want to at least understand what mathematicians write, or read about string theory and try to understand what is really meant in the words by reading the accompanying equations. I realize this might take time staring at an equation - maybe even for several days. That is OK.
What I want to know is: is this even practical/possible for a layman?
If so, where to begin to learn?

2. Nov 2, 2014

### atyy

Do you understand ma=mg?

3. Nov 2, 2014

### 462chevelle

Even if you could read some of the symbols, the operations that some of them represent aren't understood so simply.

4. Nov 2, 2014

### PeroK

I believe you are approaching this the wrong way. In general, the equations are what emerge from an understanding of the physics. Not the other way round.

Special Relativity, for example, is not postulated on the basis of a set of equations. It has two (very simple) physical postulates. You can understand a lot of Special Relativity with only basic maths.

5. Nov 2, 2014

### bugmagnet

Thanks for all the responses :)

I understand it algebraically.
But I don't know what it means in physics because I don't know what the variables represent.
I presume I could gain insight if I knew.
How/where could I look it up?

Perhaps I am.
Then where to begin?

6. Nov 2, 2014

### PeroK

It depends what you want to achieve. You could ask yourself:

Do you want to understand the history of physics? Who discovered what when? Gravity, electricty and magnetism, the nature of the atom etc.

Do you actually want to study physics (so you can solve problems)? Then you either need to enrol on a course or possibility self-study using the Internet and this forum.

Do you want to gain an insight into modern physics: the expanding universe, quantum mechanics, relativity, particle physics, string theory? Unless you have studied physics and maths up to a certain level, you are probably limited to "popular science" level of understanding.

You need to decide what you want to do or learn.

7. Nov 2, 2014

### Kafzilla

Well, basic vector math and the concepts and notation of calculus (derivatives and integrals) do wonders with motion physics with respect to applying Newton's laws.

8. Nov 2, 2014

### PeroK

Perhaps Google "introduction to physics" and see what you get.

9. Nov 2, 2014

### bugmagnet

My goal is close to what PeroK suggested:
I want to understand the problems at both a metaphorical level and mathematical level, to the point where it is at least possible for me to solve them.

I am a game developer, so I have modeled force/mass/momentum physics (even though physics are mostly faked in games). Vectors are bread and butter to a game coder. I can use quaternions, but I have never been able to visualize them. Maybe this gives an idea of where I am at. I know physics math goes much much deeper than this. I am not afraid of that.

Given that my biggest interests in physics are cosmology and the problems of combining relativity with QED:
Is there a must-read book? A must-read site? A reference or a paper that I should determine to understand first?
Clearly, I must understand the related fields first, before ever understanding the problems of combining them.

My goal is to reach that level.
I know it will take years of self study. I am not afraid of that, I am just looking for directions. I thank you in advance :)

10. Nov 2, 2014

### PeroK

It seems to me you need to do a University Physics degree. Have you looked into that? The course providers will know what prerequisites you need and whether you'll need to do some preliminary courses.

In the UK we have the Open University, which is designed for those in work who cannot enrol on a full-time or campus-based course.

11. Nov 2, 2014

### homeomorphic

You might want to try The Road to Reality, by Penrose. It's pop-sci, but not the usual pop-sci. Or a graduate level text book masquerading as pop-sci, as someone put it (maybe Peter Woit?). As Penrose points out in the intro you can read it on different levels. It's too difficult to get much out of it without reading other things, but it's a good guide to what to study. The way I've used it is to get previews of subjects I need to study, reinforce subjects that I have studied, and get extra insights that you won't find elsewhere because Penrose has good intuition. Definitely just a supplement, though.

Books are generally better than internet stuff, still, but you'll find a lot of stuff online. I'm not sure that there's any one that's a must, but World Science U is a good one. Susskind has his lecture videos posted.

I would steer clear of most papers, if you are just doing it for the sake of it, although I probably shouldn't discount the possibility that some older papers might have some value at some point. They are definitely not the place to begin.

12. Nov 3, 2014

### atyy

Is this too basic or too advanced?

A good set of free books are these by Ben Crowell: http://www.lightandmatter.com/books.html.

Last edited: Nov 3, 2014
13. Nov 3, 2014

### bugmagnet

Thank you guys for the responses, I really appreciate it.
I have MIT near me, I have thought about maybe auditing classes there. But they also have opencourseware, which I will look into more seriously as per your suggestion.

I will read it. Thank you for the recommendation!

I have watched some classes that Susskind has on youtube. I really liked them but I couldn't follow the whole way. I understood maybe 60% when it was starting, which dropped to 1% after maybe half way.
as far as World Science U it looks really good. I didn't know about it. The "master" classes seem especially in-depth/ promising. Thank you!

Too basic. Like I just mentioned, I have tried watching Susskind's classes on QED, but they were a little bit too advanced.
I could follow Feynman lectures I have found, fwiw.

Thank you guys for the resources! I really appreciate it!
They will keep me busy for the short term to begin. Thank you ^_^

14. Nov 3, 2014

### atyy

The Feynman Lectures are great. Volumes 1 and 2 are still excellent, and volume 3 (my favourite) has become outdated because there were some basics of quantum mechanics that Feynman didn't understand (I think Feynman's postulates for quantum mechanics are idiosyncratic, and he did not understand the possibility of hidden variables, which was only widely understood after the work of Bohm in the 1950s and Bell in the 1970s). Nonetheless, volume 3 is still valuable, especially because it starts with the simple case of spin 1/2, which does build the correct intuition for all of quantum mechanics. But I would follow it up with the quantum mechanics texts of Landau and Lifshitz, Shankar, Sakurai & Napolitano, Weinberg, and Nielsen & Chuang. Shankar is an all round good text. The modern postulates are given especially well by Sakurai & Napolitano, and Nielsen & Chuang. Landau & Lifshitz, and Weinberg are especially good for interpretation.

Since you are able to follow the Feynman lectures, here's a very good reading list that was drawn up by John Baez: http://math.ucr.edu/home/baez/books.html.

Another useful list is 't Hooft's http://www.staff.science.uu.nl/~Gadda001/goodtheorist/.

Last edited: Nov 3, 2014
15. Nov 3, 2014

### bugmagnet

Wow, atyy thank you!
Hooft's list seems to be exactly what I have been looking for!

Thank you all so much. I definitely have enough resources now to get serious.
Thank you thank you! :D

16. Nov 8, 2014

### specdude

17. Nov 9, 2014

### carllooper

I think this is a really really good question.

The language in which physics is expressed is mathematics and at a fundamental level, so too are algorithms in computer programming, however a programmer will typically learn mathematics via another language, such as C# (to quote but one of many languages). And indeed they can acquire a pretty good understanding of mathematics and physics in this way. Despite not being able to read mathematical notation. In a sense they already understand the concepts (vectors, matrices, etc) just not how to write it down in native form.

But most books on physics in computer programming only scratch the surface of physics, and only to the extent of how such concepts in physics might be used in, for example, a computer game.

For physicists the opposite can often be the case. They will, of course understand the physics and the concepts and how to express that in native mathematical notation, but not necessarily how to translate that into, for example, a C# program.

I believe there is a branch of physics, called computational physics, but from what I understand it assumes the reader is already a physicist and/or understands mathematical notation.

What would be great is the reverse. Physics for computer programmers. Not physics for games, but physics in itself. And instead of expressing physics in mathematical notation, it would be expressed in computer language terms instead - either using an established language such as C#, or what's otherwise known as "pseudo-code" (where programmers already know how to translate such into a specific computer language).

And if the book can provide a way to translate the C# back into mathematical notation, the programmer would then have a way to eventually reach an understanding of the concepts expressed in their native form, and eventually be able to translate any concepts expressed in such native form into their preferred language and a resulting computer program.

Carl

18. Nov 9, 2014

### bugmagnet

This would be pretty cool! But I wonder if there would be some dead ends... for example set math. Can arrays replicate the behavior of sets in mathematical notation? What about irrational numbers? Divide by 0 (which almost universally causes a crash)? It's a good thought though i wonder where it would lead in practice.

19. Nov 10, 2014

### homeomorphic

Actually, in Java, there's a Set interface and data structures like HashSet, which are like lists, except they don't allow duplicates. So, I don't think finite sets are a big deal. Infinite sets are another story. So, there are some aspects that need to be studied separately from computers, but there's a lot of stuff you can do with a computer. Irrational numbers can be approximated by floating point numbers, so not being able to deal with irrational numbers isn't the end of the world. Physics doesn't depend fundamentally on those points, anyway, although the math that expresses it might at times. Dividing by zero also causes a "crash" in pure mathematics.

I'm not able to fully imagine what "math and physics for programmers" would look like, but some form of it would be possible.

There are a lot of things that are not constructive in math, so I don't know how you'd code them up, and even if they are constructive, mathematics has a lot of nasty combinatorial explosions lurking, which would prevent many algorithms from running in a reasonable time. I've been thinking about graph algorithms lately and it seems like there are a lot of exponential time ones that would involve something computationally horrible, like checking every subgraph for a certain property (this is the sort of algorithm directly suggested by Kuratowski's theorem on planar graphs--you have to be very clever to come up with better ways). But it's easy enough to reason about it mathematically. There are certain advantages to not depending on a computer, although you can learn a lot by teaching dumb machines how to do math.