Science Forums: Do you have a new theory? - Science Forums

Jump to content

Welcome to ScienceForums.Net!

Welcome to ScienceForums.Net! We welcome science discussion at all levels — from beginners to researchers, covering topics from biology to computer science, and much more. Registration is fast and free, and allows you to post on the forums, so register now and join the discussions!
  
After you've registered, come in and introduce yourself, or visit the forum index. If you need any help  registering, posting, or if you just have some questions about our site, please feel free to contact us at staff at scienceforums dot net.

  • Start new topics and reply to others
  • Subscribe to topics and forums to get automatic updates
  • Create a ScienceForums.Net Blog!
Guest Message © 2012 DevFuse
  • 11 Pages +
  • 1
  • 2
  • 3
  • Last »
  • You cannot start a new topic
  • This topic is locked

Do you have a new theory? Rate Topic: ****- 3 Votes

#1 25 August 2005 - 10:21 AM Severian 


Scientist
Very often people come to these fora with a belief that our current theories of physics, such as the Standard Model or relativity, are flawed and present some alternative of their own. On the whole, this is a fine attitude to take - we should always be skeptical, and it is good if people can think a little 'out of the box' and generate ideas which more standard thinkers may not have come up with. I have always thought that genius was not an ability to think 'better' than everyone else - it is an ability to think differently from everyone else.

However, when coming up with a new theory it is important that it should be better than the old one. Therefore the first step of coming up with a new theory is a sufficient understanding of the old one. You have to make sure that your new theory does everything at least as well as the old theory, otherwise the old theory remains more attractive. This is very difficult mainly because our current theories are so spectacularly good in their predictions.

Let me give an example: the magnetic moment of the electron.

If we look at the energy (Hamiltonian) of an electron in an electromagnetic field, we find that there is a contribution from the interaction of the electron's angular momentum and the magnetic field. For an orbital angular momentum L, this is \vec{\mu}_L \cdot \vec{B} with a magnetic moment

\vec{\mu}_L = - \frac{e \hbar}{2mc} \vec{L}

(The charge of an electron is -e and its mass is m.)

However, if the particle has 'spin' (intrinsic angular momentum) \vec{s}, we also have a contribution to the magnetic moment of

\vec{\mu}_s = - g \frac{e \hbar}{2mc} \vec{s}

g is known as the gyromagnetic ratio, and its value depends on the theory. Since this can be measured in experiment very accurately, it is a good test of a theory to check if it predicts the correct gyromagnetic ratio.

For example, simple QM (the Dirac equation in an em field) predicts a gyromagnetic ratio g=2. Experiments shows that g is very close to 2, so this is good news, but since experiment shows that it is not quite 2, the Dirac equation cannot be the whole answer.

Quantum Field Theory, in the form of the Standard Model, predicts a deviation from 2. It is usual to write down the prediction for this deviation from 2 rather than the gyromagnetic ratio itself. For the SM this is:

\frac{g_{\rm th}-2}{2} = 1159652140(28) \times 10^{-12}

The experimantal result is:

\frac{g_{\rm exp}-2}{2} = 1159652186.9(4.1) \times 10^{-12}

(A note on errors: the numbers in brackets denote the error on the prediction/measurement at the same precision to which the value is specified. For example 1159652140(28) means 1159652140 \pm 28 and 1159652186.9(4.1) means 1159652186.9 \pm 4.1.)

You can see that the theory predicts the correct experimental value to incredible precision (although the experimental error is still better than the theory one). If you want to persuade scientists that the Standard Model is wrong, then you have to explain why this is a coincidence or show that your new theory predicts g-2 to at least this accuracy.
0

#2 25 August 2005 - 01:03 PM Locrian 


Baryon
So true.
0

#3 25 August 2005 - 01:59 PM Jacques 


Molecule
I was wondering if the Standard Model was able to predict the instability of the neutron ? Is it able to predict 15 minutes life of the neutron ?
If yes what are the equations and parameters used to achive that ?
0

#4 25 August 2005 - 02:08 PM Severian 


Scientist

Quote

I was wondering if the Standard Model was able to predict the instability of the neutron ? Is it able to predict 15 minutes life of the neutron ?
If yes what are the equations and parameters used to achive that ?


Yes it does. This is a weak decay. The neutron is made up of three quarks: one 'up' quark and two 'down' quarks. One of the down quarks decays to an up quark by emitting a 'W-boson' (which then decays to lepton and a neutrino). Since the proton is two up quarks and a down quarks, the neutron has decayed into a proton.

The actual calculation is a bit beyond the scope of a web forum though....
0

#5 25 August 2005 - 03:23 PM LucidDreamer 


Protist
But going to the trouble to really understand a current scientific theory isn't as much fun as learning just a tiny bit and then coming up with my own crazy theory that appears to be better than the original because of my lack of understanding.
0

#6 25 August 2005 - 03:48 PM Royston 


Señor Butt Monkey

LucidDreamer said:

But going to the trouble to really understand a current scientific theory isn't as much fun as learning just a tiny bit and then coming up with my own crazy theory that appears to be better than the original because of my lack of understanding.


Perhaps you should start another thread, this one's for the experts...which I'm clearly not either, but it won't stop me reading and researching the terminology used and trying to get a slight understanding of the subjects raised...that also can be fun.

I think Severian deserves a break from advising people like me all the time.
0

#7 25 August 2005 - 03:54 PM Jacques 


Molecule
Thanks Severian
So the standard model can compute the life time (or should I say the half-life time ?) of the neutron! I asked because the only thing I saw on the standard model is little table with the "fundamental particle" and some Feynman diagram .
What would be a good start to go deeper than that ? What level of math is necesary ?

Quote

Therefore the first step of coming up with a new theory is a sufficient understanding of the old one.

I completly agree and I am at that step.
0

#8 25 August 2005 - 04:04 PM Severian 


Scientist
To really understand the Standard Model, you really need to understand Quantum Field Theory. The best book on QFT that I have seen is called "An Indtroduction to Quantum Field Theory" by Peskin and Schroeder. It starts off quite advanced though.

What level are you at? Would you understand this:
http://www.phys.ualb...html/node1.html
or is that too advanced?
0

#9 25 August 2005 - 05:08 PM Jacques 


Molecule
Thanks for the link ! It will be a lot of work for me to understand, but with the help of Wikipedia I think I will be able to grasp the essentials things. My goal is not to do any calculations, but to understand the basic principle (more than the usual analogy).
0

#10 25 August 2005 - 05:24 PM bascule 


Genius
So does this mean thefinaltheory.com is wrong? *gasp*
0

#11 25 August 2005 - 07:56 PM nameta9 


Suspended
Severian, you seem to have some knowledge of modern physics. I have some too, but not very technical. My question is actually how on earth can any normal person hope to completely understand and calculate things as hard as quantum electrodynamics ? The theory is hard, the calculations are harder and in many cases very few scientists can even cross check many results, you need a supercomputer and need to work at CERN or with an equipe. At this point how much can we really trust the results that are being given to us since it becomes increasingly difficult to REALLY control them ?

How many years of studying tensors and integrals and all are necessary ? How smart in the technical sense must a person really be ? It has always been a question in the back of my mind. Thanks for any clue.
0

#12 25 August 2005 - 08:10 PM MetaFrizzics 


Banned

Quote

How many years of studying tensors and integrals and all are necessary ? How smart in the technical sense must a person really be ? It has always been a question in the back of my mind. Thanks for any clue.


Actually there is now hope for almost anyone to quickly pick up and more deeply understand the basic concepts of modern physics than ever before:

Quote

Would you like to begin the study of Newtonian mechanics using NO vectors and NO (F = ma)? How about starting Quantum Mechanics with NO complex numbers and NO Schroedinger equation? Would you and your students enjoy exploring General Relativity deeply and widely with NO tensors and NO field equations?

Suppose further, that along the way your students learn concepts and methods central to contemporary physics research. Finally, what if particle motions described by Newtonian mechanics, quantum mechanics, and general relativity were summarized in three brief commands of Nature that turn out to be variations of the same command?

Physics is already being taught this way, and students respond with enthusiasm and understanding. The theoretical background for this curriculum has been around for a long time and is well developed and deeply understood by the physics and math communities.

MIT (Manchester Institute of Technology Advanced Learning)


Here are some links:

E.F. Taylor Edu. Lagrange
Anyone can master Newtonian Gravitational Theory..
0

#13 25 August 2005 - 08:31 PM nameta9 


Suspended
OK, I know you can learn it and all, and I know some of it. I mean how hard is it to really understand and calculate feynman diagrams and cross check the results knowing exactly what you are doing ? I guess I am suggesting at the PHD level where you review peer papers regarding new results and corss checking all the old ones.

It can be done and all but I have the impression that you really have to gifted to be able to start calculating these things seriously.
0

#14 26 August 2005 - 12:24 AM J.C.MacSwell 


Organism

LucidDreamer said:

But going to the trouble to really understand a current scientific theory isn't as much fun as learning just a tiny bit and then coming up with my own crazy theory that appears to be better than the original because of my lack of understanding.


LOL, as much as I agree with Sevarian, I agree with this more!

My problem is with "other people's" crazy theories. :D
0

#15 26 August 2005 - 02:23 AM CPL.Luke 


Primate

Quote

It can be done and all but I have the impression that you really have to gifted to be able to start calculating these things seriously.


yep, some people have trouble with math in general, these people probably won't be able to understand the mathmatics of a feynmann diagram without a large amount of effort.

Then again it would take me a lifetime of classes to learn how to draw anything more than a stick figure :P
if you're ever driving at 49000 miles per second, for the love of god remember that green means stop.
0

#16 26 August 2005 - 09:13 AM Severian 


Scientist

Quote

My question is actually how on earth can any normal person hope to completely understand and calculate things as hard as quantum electrodynamics ? The theory is hard, the calculations are harder and in many cases very few scientists can even cross check many results, you need a supercomputer and need to work at CERN or with an equipe. At this point how much can we really trust the results that are being given to us since it becomes increasingly difficult to REALLY control them ?


They are not that hard! I do particle physics calculations as part of my job and I have never used a supercomputer. The only need for supercomputers in particle physics is in lattice QCD, where the strong interaction is modeled on a space-time lattice. Most calculations are recalculated by other groups, but even if they aren't there are a lot of checks which can be done. The analytic expressions must have certain properties (like gauge invariance) which would be removed by a careless mistake. Mistakes are made in the literature all the time, but they are corrected quickly and it is rare for mistakes to remain for long.

Quote

How many years of studying tensors and integrals and all are necessary ? How smart in the technical sense must a person really be ? It has always been a question in the back of my mind. Thanks for any clue.


Well, simple Feynman diagrams calculations are done by our final year undergraduates, so it doesn't require that much training to get started. To work in the field and publish papers you really need to do a PhD. To get a permanent faculty job you need to have done a few 2-year postdocs. I think perseverance if more important that hyper-intelligence. No physicist understands everything, so we are all still learning.
0

#17 26 August 2005 - 10:56 AM nameta9 


Suspended
So then we can be fairly sure that all the theoretical physics results and calculations are cross checked ? I doubt this because most PHDs who are then the ones capable of controlling all the fine details are working on new theories, new results and new papers. I don't think many are cross checking all the old results. Anyways, if new results are built on all the old calculations and math, then I guess we can be pretty sure there are few mistakes.
0

#18 27 August 2005 - 11:38 AM Severian 


Scientist

nameta9 said:

So then we can be fairly sure that all the theoretical physics results and calculations are cross checked ? I doubt this because most PHDs who are then the ones capable of controlling all the fine details are working on new theories, new results and new papers.


Heh - no. A PhD is research done by the superviser under particularly arduous circumstances.
0

#19 27 August 2005 - 04:47 PM CPL.Luke 


Primate
severian, do people with phd's ever do research? under normal circumstances
if you're ever driving at 49000 miles per second, for the love of god remember that green means stop.
0

#20 27 August 2005 - 06:12 PM Severian 


Scientist
Yes, a PhD is all about research. They do supervised research so that they can learn how to do unsupervised research. I would never trust a grad student's calculation without checking it though (no offense to grad students).

Usually if there is a mistake in a calculation it is really obvious - you find an answer with the wrong properties. But even if a mistake goes unnoticed, it doesn't really matter. A false calculation will never agree with the data (unless the mistake is very small) so you will find a discrepancy with data, and the first thing you do (or someone else does!) is check the calculation.
0

Share this topic:


  • 11 Pages +
  • 1
  • 2
  • 3
  • Last »
  • You cannot start a new topic
  • This topic is locked

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users