Abstract
In 1917, Willem de SItter published the first description of an FLRW model for a universe that is expanding at an exponential rate. That FLRW model has come to be known as
de Sitter space.
In 1917, it was philosophically fashionable to assume the universe in which we live is static, so de Sitter described the space using a particular set of static coordinates. Those static coordinates present a misleading impression of the space, because the metric form of de Sitter's equation (8B) is not spatially homogeneous: the g
tt component of the metric tensor depends upon one of the spatial coordinates.
The author and sole proponent of Helland physics has been citing and quoting two different metric forms for de Sitter space, without understanding how those two metric forms assume two distinct systems of static coordinates.
He has also been failing to understand that the metric tensor described by those two metric forms is an FLRW metric. The homogeneity and isotropy of de Sitter space becomes obvious when that metric is written as an FLRW metric form.
Warning
What follows is mostly first year calculus. The author and sole proponent of Helland physics won't understand it, but scientifically literate readers who have been reading this thread for laughs might enjoy it.
What I've read since then also casts doubts on your coordinate system claims.
http://www.bourbaphy.fr/moschella.pdf
....snip....
Ok, as for choosing time coordinates, the "de Sitter tour" goes on:
[qimg]http://www.internationalskeptics.com/forums/imagehosting/thum_762186490b4c275093.png[/qimg]
(click thumbnail to read the caption)
A couple other sources say similar things:
https://astro.uchicago.edu/~kent/fnal/effect.html
https://physics.stackexchange.com/q...bble-redshift-be-interpreted-as-time-dilation
And here's de Sitter's 1917 paper:
So I guess as far as the map territory stuff goes, choosing how to make your time coordinate seems like a normal thing to do.
And de Sitter's original version, seems to give a time dilated past, when the appropriate part of the manifold is covered, which in this case should be less than or equal to zero but greater than -1. This represents the present and the past.
As for the sentence I highlighted: Yes, the theory of relativity allows you to select a coordinate system you prefer to use. That's why it's called the theory of relativity.
As this thread and its predecessor have demonstrated again and again, some coordinate systems are less misleading than others. Although all admissible coordinate systems can be used to reason correctly about the physics, a poor choice of coordinate system makes the physics considerably harder to understand. If your knowledge of general relativity and cosmology is marginal to begin with, so you struggle to understand even the best choices of coordinate system, then a poor choice of coordinate system is going to make the physics even harder for you to understand, to such an extent that you will almost certainly make mistakes.
Which explains why individuals whose hobby horses require them to make such mistakes are especially likely to prefer a poor choice of coordinate system.
For example:
It kind of looks like de Sitter's 1917 model with closed static spatial coordinates might have been a better fit all along.
And while we’re about it could you do the converse and explain what a relative time would be in a homogeneous and isotropic universe?
Wouldn't it be this:
http://latex.codecogs.com/gif.latex?ds^2 = -\left(1-\frac{r^2}{\alpha^2}\right)dt^2 + \left(1-\frac{r^2}{\alpha^2}\right)^{-1}dr^2 + r^2 d\Omega_{2}^2.
?
I think this covers a part of the manifold centered on an observer.
That’s the Schwarzschild metric which describes the effect of gravity outside a spherically symmmetric central mass, so it’s obviously not homogeneous and is completely inappropriate for cosmology.
No, that is not the Schwarzschild metric. It is a metric form for de Sitter space, expressed using a certain set of static coordinates that (misleadingly) makes the space appear to be inhomogeneous: the g
tt component depends upon the spatial coordinate r.
Because de Sitter space really is homogeneous and isotropic, and that particular choice of coordinates really does make the progress of time depend on one's spatial location,
Mike Helland actually gave a fairly good answer to
hecd2's challenge.
But
Mike Helland came up with that metric form via his usual cargo cult process of searching the web for something he can misinterpret as support for his mistakes. He doesn't understand how that particular coordinate system and that particular metric form are related to the coordinate system and metric form in de Sitter's paper from 1917:
That equation is from:
https://physics.stackexchange.com/q...bble-redshift-be-interpreted-as-time-dilation
W. de Sitter's line element (8B) is:
http://latex.codecogs.com/gif.latex?ds^2 = -dr^2-R^2\sin^2\frac{r}{R}[d\psi^2+\sin^2\psi d\theta^2]+\cos^2\frac{r}{R}c^2dt^2
I suppose through trigonometry they're equivalent? Haven't fully worked that part out yet.
I can help with that.
Theorem
All four of the following metric forms describe exactly the same metric tensor, which is the metric tensor of de Sitter space:
http://latex.codecogs.com/gif.latex?ds^2 = -dr^2-R^2\sin^2\frac{r}{R}[d\psi^2+\sin^2\psi d\theta^2]+\cos^2\frac{r}{R}c^2dt^2
http://latex.codecogs.com/gif.latex?ds^2 = -\left(1-\frac{r^2}{\alpha^2}\right)dt^2 + \left(1-\frac{r^2}{\alpha^2}\right)^{-1}dr^2 + r^2 d\Omega_{2}^2.
http://latex.codecogs.com/gif.latex?ds^2 = -dt^2 + \alpha^2 \cosh^2(t / \alpha) d\Omega_{3}^2
http://latex.codecogs.com/gif.latex?ds^2 = - dt^2 + \left(e^{(t / \alpha)}\right)^2 [dx^2 + dy^2 + dz^2]
Metric form 1 is expressed using de Sitter's static coordinates from his 1917 paper, and is equation (8B) of that paper. Its sign convention is (− − − +). The g
tt component depends on the spatial coordinate r, so these static coordinates obscure the homogeneity and isotropy of the model.
Metric form 2 is expressed using a different set of static coordinates, a different sign convention (− + + +), and the abbreviation
http://latex.codecogs.com/gif.latex?d\Omega_{2}^2 = d\psi^2+\sin^2\psi d\theta^2
Once again, the g
tt component depends on the spatial coordinate r, so this second set of static coordinates also obscures the homogeneity and isotropy of the model.
Metric form 3 is expressed using spherical spatial coordinates in which the homogeneity and isotropy of the model become evident.
Metric form 4 is expressed using Cartesian spatial coordinates and the (− + + +) sign convention. Once again, the homogeneity and isotropy of the model are evident.
Metric forms 3 and 4 are instances of the familiar FLRW metric form. With metric form 3, the spatial slices are hyperspherical, with constant positive curvature. With metric form 4, the spatial slices are flat (Euclidean), with zero curvature. How can they be describing the same metric tensor?
The Ricci and Riemann curvature tensors are invariants, independent of the coordinate system you choose to use, but their numerical components depend on that coordinate system. Furthermore, they are defined on the 4-dimensional spacetime manifold. In relativity, the division of 4-dimensional spacetime into one time dimension and 3 spatial dimensions is not absolute. (That's part of why we call it the theory of relativity.) In some cases, that limited choice of what counts as time and what counts as space allows your choice of coordinate system to push curvature into spatial dimensions if you so desire, or into the time dimension if that is what you prefer.
I will now sketch a proof that metric form 1 (de Sitter's equation (8B) from 1917) describes the same metric tensor as metric form 2. (
Mike Helland guessed that their equivalence could be proved using trigonometry, but he guessed wrong.) Their equivalence is proved by a change of variables, just as in my proofs that the (revised) Helland metric form and the TDP metric form are nothing more than obfuscations of the familiar Minkowski metric.
Sketch of proof
Rewrite de Sitter's equation (8B) to use the (− + + +) sign convention.
Rename de Sitter's R to α (alpha).
Define a new variable u by
http://latex.codecogs.com/gif.latex?u \equiv \alpha \sin \frac{r}{\alpha} = R \sin \frac{r}{R}
Then
cos2 (r/R) = 1 − sin2 (r/R) = 1 − (u2/α2)
and
du2 = (cos2 (r/R)) dr2 = (1 − (u2 / α2))
so
dr2 = (1 − (u2 / α2))−1 du2
Use those equations to replace R and r and dr by expressions involving α, u, and du.
That gives you a metric form that is the same as metric form 2 except it is expressed in terms of u and du instead of r and dr. Simply rename u to r, and you get metric form 2.
To prove the rest of the theorem, I'm going to let Wikipedia do the work.
Metric form 2 is the metric form of
Wikipedia's static coordinates for de Sitter space. Wikipedia gives equations that define another set of coordinates x
i in terms of those static coordinates. (Wikipedia's presentation actually starts with the x
i coordinates and defines the static coordinates in terms of those x
i, but we can go in the other direction.)
Metric form 3 is obtained from metric form 2 via transformation to
Wikipedia's closed slicing coordinates, which are defined (implicitly) by equations that have the x
i on the left hand side.
Metric form 4 is obtained from metric form 2 via transformation to
Wikipedia's flat slicing coordinates, which are also defined (implicitly) by equations that have the x
i on the left hand side.
Metric form 4 is the metric form that makes it easiest to see that de Sitter space describes a universe that is expanding at an exponentially increasing rate. It is also the metric form that makes it easiest to see that the relationship between redshift and distance in a de Sitter universe coincides with the relationship postulated by Helland physics, for an allegedly non-expanding universe (with absolutely no attempt to explain how a non-expanding universe could possibly give rise to the same cosmological redshift as the exponentially expanding de Sitter universe).