User talk:Count Iblis/Speed of light

inconsistent units

In the interest of making these additions transparent to the non-expert, can you provide some preamble about what is meant by "inconsistent units"? Are they the same as units that are not natural units? Does the article on natural units appear to you to present the matter adequately for the non-expert, in particular, why are natural units anything more than a notational convenience? Is the natural units article well sourced? Brews ohare (talk) 05:24, 1 August 2010 (UTC)

That's a good idea, I'll expand the introduction to address these issues about natural units. Count Iblis (talk) 16:48, 1 August 2010 (UTC)

Why does light propagate a speed of 1 in natural units?

And not with a speed of 2, pi or with Euler's constant? The answer to this is simple: Because natural units are (partially) defined by the condition that c=1. There is no logical or natural reason to take this value, except convention. This makes the second part of your lead a circular argument.TimothyRias (talk) 09:59, 1 August 2010 (UTC)

I agree that it is a convention, but it is "natural" in a similar sense as why you won't think of adding extra conversion factors for distances in the x, y and z directions relative to some system of axis and then put all these conversion factors in the rotation matrix describing transformations from one coordinate system to another. Count Iblis (talk) 16:55, 1 August 2010 (UTC)

It would also be "natural" to take units in which c=Sqrt(3), this has the cool property of making the (naive) trace of the Minkowski metric zero. Which is pretty awesome, admit ;-), it is at least as awesome as making the absolute values of the diagonal entries of this metric equal (i.e. c=1). My point here is that time is already different in the Minkowski metric, this introduces a fundamental speed in the theory. The value of this speed is not gauge invariant and may by a suitable gauge choice be taken to have any positive value. But this fundamental speed is there and is called the speed of light, and the fact that it isn't zero or infinite has physical meaning. What value it takes not so much, it could 1, sqrt(3), or Pi, whatever is convenient for a calculation. Usually, c=1 is pretty convenient, but its not a more natural choice, than that Lorenz gauge is the natural gauge choice for EM.

Anyway, the last sentence of your intro:

"Since light always propagates at the speed of 1 in natural units, the value of c is identical to the speed of light in the new units, hence the name "speed of light" for c."

Simply, puts things in reverse. Ask yourself the question: "Why does light propagate at the speed 1 in natural units?" and try to come up with an answer that does not reduce to "because c=1".TimothyRias (talk) 11:54, 2 August 2010 (UTC)

I now see better what the objection to the last sentence is. I, of course, agree that any value for c is as good, as the different choices can be mapped to each other. Thinking again at what I wanted to say in the lead, I see that the main point is about leaving an undermined parameter c in the equations and then pointing out that this is a mere rescaling parameter.

I agree that in natural units, one does make the arbitrary choice ds^2 = dT^2 - dX^2 instead of e.g. ds^2 = 3 dT^2 - dX^2 as you point out. However, this sort of freedom is present in any equations of physics, e.g. one can just as well replace dE = T dS - P dV by dE = c T dS - P dV. So, I think there is more to why one thinks that putting a c in equations of relativity is somehow different. I think this originates from classical thinking, because it is only in classical physics that the equations stay invariant under a rescaling of time.

We've gotten used to that classical freedom so much that special relativity could not overturn this. The fact that the speed of light is large when expressed in conventional units and that it took a long time for the metre standard to be redefined using the speed of light, presumably also played a role here. Count Iblis (talk) 15:23, 2 August 2010 (UTC)

Well, there is this other thing about c. It has a clear physical interpretation it is the speed that EM waves propagate at. Even in natural units you can ask the question at what speed do EM waves propagate and symbolically call the answer c, and come to the conclusion that its value must be 1. Of course such a question only makes sense if you have made a time slicing of your spacetime, in which case c is relate to the lapse function of the slicing.

The other thing to say is that the statement you are making here about c can be made about any dimensionful physical quantity. You can always rescale the unit to make the quantity 1. Makes this all dimensionful quantities completely meaningless?TimothyRias (talk) 15:45, 2 August 2010 (UTC)

Yes, in case of the other dimensionful quantities, you also have that the classical equations remain invariant under a rescaling that corresponds to changing units, while the (general) relativistic or quantum equations do not stay invariant. This is explained by recognizing that the classical limit is obtained by an infinite rescaling from the fundamental equations.

A physicist who lives close to the classical limit then has to deal with different physical quantities that cannot be compared to each other in a meaningful way. So, he invents units for these quantities such that the typical values encountered in daily life are of order unity. The different units are assigned different dimensions to distinguish them from each other. As a consequence, when the fundamental equations are discovered, they contain conversion factors that are extremely large or small when expressed in conventional units. These conversion factors are, of course, dimensionful precisely because they have to compensate for whatever dimensions were assigned to the different units. Count Iblis (talk) 16:49, 2 August 2010 (UTC)

Reliable source

You need a reliable source for this. I didn't know what to think of it. I even considered the possibility that it was satire. --Bob K31416 (talk) 13:40, 1 August 2010 (UTC)

If I wanted to add this as written up now to a Wiki article, then I agree. But I now just want to present what I think is the correct interpretation of the constant c. Count Iblis (talk) 16:58, 1 August 2010 (UTC)

You might consider this excerpt from WP:NOR.

"The best practice is to write articles by researching the most reliable sources on the topic and summarizing what they say in your own words..."

Good luck and regards, --Bob K31416 (talk) 17:19, 1 August 2010 (UTC)

Cosmology

Do you want to involve this discussion with the questions of evolution of the constants with the expansion of the universe, the Dirac conjectures and Non-standard cosmology, for example? Brews ohare (talk) 17:16, 2 August 2010 (UTC) Maybe this book is useful? Brews ohare (talk) 12:18, 3 August 2010 (UTC)

I think I'll stick to down to Earth established physics :) . Count Iblis (talk) 15:09, 3 August 2010 (UTC)

Relation to reduction of all base units to time

By suitably choosing the metre, lengths are seconds, and the long-term goal is to have only one base unit, the second. That would introduce a number of reference constants like c =299,792,458 m/s. Is there a connection between these ideas and natural units? Brews ohare (talk) 12:24, 3 August 2010 (UTC)

Yes, although the metrological aspects could just as well cause us to move in the other direction, as far as the practical realization of the standards is concerned. E.g., we may decide to measure time using the signals of a pulsars in the future if that provides for a better standard than a frequency standard based on some atomic transition. We would then move further away from a standard based on fundamental physics. From the theoretical point of view, you could consider a frequency standard that is closely related to a fundamental parameter of the theory, e.g. the Compton frequency of the electron.

From the point of view of fundamental physics, what matters is what the experimental limits on the deviations of the theory are. So, if using pulsars to actually measure time things improves the accuracy of measurements, this will then contribute to stronger experimental limits on any possible violations of the theory, which in turn will make certain theoretical standards that may be completely impractical, better motivated. Count Iblis (talk) 15:28, 3 August 2010 (UTC)