On Wednesday, January 27, 2016 at 11:49:06 PM UTC-8, Koobee Wublee wrote:
> On January 27, 2016, JanPB the relativistic moron[sic] wrote:
>
> > Show us that the Einstein equation after transforming to spherical
> > coordinates produces g_ij independent of time and no cross terms.
> > Good luck.
>
> Once again, obviously, Jan the relativistic moron has never solved any differential equations. To solve such, you have to start with some educated guessing. One such assumption is the static case hoping to find one such solution that will degenerate into Newtonian gravity at weak curvature of spacetime. <shrug>
Yes, correct so far.
> Besides, the assumption that space and time can possibly intertwine lie what you are so fvcking gung ho over has never shown such a necessary merit of doing so.
Stop! Here is the problem: you simply _assume_ that in a spherically symmetric static vacuum
the coordinate space t=const. and the coordinate time (xyz)=const. are orthogonal. But there
is simply no reason for that. And you cannot deduce this from the Einstein equation either:
without _assuming_ diagonal metric that equation is a very complicated system of PDEs, so
you simply don't know if its solution (assuming you had it) would NOT be analogous to
something like:
ds^2 = dt dx (cross-terms present)
or:
ds^2 = -t^2 dt^2 + dx^2 (a time-dependent g_tt)
...both of which are _static spherically symmetric vacua_ (in fact they are both the flat
Minkowski space or a portion thereof).
In fact, with hindsight, we know that this is precisely what happens: the curvature of
spacetime in the large is such that _insisting_ on [1] coordinate space and coordinate time
orthogonality (i.e. no cross-terms) and insisting on [2] labelling the spheres of symmetry
by (essentially) the square root of their areas FORCES the coordinate spaces to bend
near the horizon _so much_ that they are not able to cross it (they look like graphs of
2M * ln|r - 2M| + (arbitrary constants) on the Eddington-Finkelstein diagram).
You are free to _change_ the coordinates so that there are no cross terms and g_ij are
time-independent but this means forfeiting the standard spherical coordinates in R^3. From
that point on you cannot claim that your system's domain covers the entire manifold anymore,
you cannot claim any singularity of the coefficients g_ij is a real singularity of the
manifold, and you cannot claim any coordinate-values-specific implication for the geometry
like "(0,0,0) is the centre" or "(0,0,0) is pointlike", etc. etc. etc.
BTW, if you are willing to give up the constraint [2] above, then you can get a diagonal
metric without coordinate singularities, that's how the Kruskal-Szekeres does it.
> It is not found in the Minkowski spacetime.
It is:
ds^2 = dt dx
or:
ds^2 = -t^2 dt^2 + dx^2
...or countless others. How do you know that a complicated mess of Christoffel symbols, each
of which is a sum involving g_ij and their derivatives, does _not_ yield a solution of
the above type?? You need to _prove_ it if you want to use it. Plus, you are also assuming
that the solution would label the spheres of symmetry by sqrt(their area/4pi). In other
words, you are assuming the existence of the term r^2*dO^2. How do you know it's not
some _other_ function of r?? In fact, as Kruskal-Szekeres coordinates show, you _do_ need
a rather complicated function f(r) in that term, so it's equal to f(r)*dO^2 there.
> Why do you expect it to find it in the real world?
Because you _don't know_ otherwise. One cannot just assume things without proof. If you
don't know something, you keep the option open. If it turns out after you're done that
the assumption was in fact correct, fine. But you cannot _start_ by assuming stuff willy-nilly.
> The equation describing a segment of spacetime geometry can very well be defined as follows. <shrug>
>
> ** ds^2 = c^2 [g]_00 dt^2 - [g]_ij d[q]^i d[q]^j
>
> Where
>
> ** i, j = 1, 2, 3 of special dimensions only
> ** [g] = 3x3 matrix that represents the metric
> ** [q] = 3x1 (or 1x3) matrix that represents the coordinate system
It can be defined that way after the aforementioned coordinate change. Here you are already
past the step of aligning the original coordinate system with the hypersurfaces that define
the word "static". This alignment is accomplished by a coordinate change.
> Koobee Wublee does not have to prove the stupidity of the silly assumptions done by the mystic generations to follow after the founding fathers of GR. So, the burden is up to Jan the relativistic moron to prove time and space can intertwine. Good luck with that. <shrug>
The burden is on you though. I _don't_ assume something, you _do_ assume something.
> > > You have to first to try using the polar coordinate system because
> > > under this coordinate system, the metric becomes diagonal.
> >
> > It doesn't. It becomes diagonal only after a _second_ coordinate
> > change.
>
> No, Schwarzschild started with the Cartesian coordinate system and found the metric cannot possibly be diagonal. Thus, being a person (very unlike Jan the relativistic moron) that had solved many differential equations in the past, instinct just told him rightfully that the polar coordinate system will certainly yield a diagonal metric.
It doesn't. The change from the Cartesians to the sphericals will simply change one complicated
mess of PDEs into another. You still don't know if that other mess does _not_ produce solutions
of the form I quoted above for the 2D Minkowski flat spacetime.
Schwarzschild quotes Einstein for that step, I haven't read his paper, it would be
of historical interest to see how Einstein describes the process (although his opinion on
the subject is not relevant as mathematics does not depend on _who_ is saying stuff).
> Solving such a complex set of differential equations require some trials and errors in which Jan the relativistic has no experiences in doing so. <shrug>
Except ages ago I posted a derivation of the solution:
https://drive.google.com/file/d/0B2N1X7SgQnLRQ2ctaXA4aTVId28
> > That second coordinate change is the root cause of all the
> > misunderstandings regarding Schwarzschild's paper.
>
> No, there is no mysticism to the second coordinate change.
Nobody says there is any "mysticism" there. Just an easily overlooked step.
> > No, the determinant is equal to -1/4.
>
> You are correct.
Nice of you to say this!
> Koobee Wublee was thrown astray by your expertise in bringing up irrelevant topics.
Meaning: we were discussing whether a static spherically symmetric metric must be diagonal,
and I wrote down a couple of easy counterexamples.
How that is "astray" I shall never know :-)
> > If you change to the spherical system, all you do is you change the
> > Einstein equation to the spherical system. You _don't_ get the
> > diagonality yet, simply because there is no reason for it. The
> > transformed field equations might as well produce something like
> > the example of the Minkowski (flat) metric I gave:
> >
> > ds^2 = -T^2 dT^2 + dX^2
>
> Notice there is no hints of time and space can possibly be intertwined to create non-diagonal metric.
Point is, there are no hints of time and space NOT intertwined. You cannot just assume
stuff - if you don't know something, you have to keep the option open.
> However, that is a topic worth of philosophic discussions not really a scientific one.
This is not a philosophy problem but a mathematical one (or logical).
> With that said, why doesn't Jan the relativistic moron show how Jan the relativistic moron would lay out these differential equations of the field equations into whatever the God coordinate system that Jan the relativistic moron has in mind. Koobee Wublee thinks this is a fair challenge since Jan the relativistic moron has been tossing divine scriptures from the religion of relativity for quite some time without getting its hands dirty. <shrug>
The definition of "static"(*) includes a construction of such coordinate system. This system
is thus not given _explicitly_. The assumption is that such coordinate change _exists_.
Ultimately the justification for this assumption is that the metric found using this
approach does in fact satisfy Ricci=0.
(*) Spacetime is static if there exists a timelike Killing field(**) in it (that's the definition
of "stationary") AND this field is hypersurface-orthogonal. The required coordinate system
is then set up by aligning these hypersurfaces with coordinate spaces of t=const. See e.g.
Wald's "General Relativity" or MTW or many others.
(**) In other words, translating the metric along this field's flow lines does not change it.
> Just for comparison, Koobee Wublee would follow Schwarzschild's footsteps and declare the polar coordinate system as the choice of coordinate system to describe the metric solved as solution in later differential equations derived from the field equations, and no other coordinate system can possibly fvcking contaminate with this decision made by Koobee Wublee.
Perhaps you are following Schwarzschild, point is neither Schwarzschild nor Einstein nor
anybody else can just assume that the polar system implies the Einstein equation magically
will yield solutions in the diagonal form, labelling the spheres of symmetry by sqrt(area/4pi)
to boot. Anything else you care to assume? Why not just assume that the solution is
the flat space and be done with it? This would be an equally arbitrary nonsense assumption.
> This means the metric must be diagonal.
No, it's just wishful thinking.
> So, only 4 differential equations are involved. Of course, the extra assumption of spherical symmetry reduces to just 3 differential equations. Then, the Christoffel symbols of the second kind are to be derived from the polar coordinate system and no other fvcking coordinate systems are fvcking allowed. They are followed by the Ricci tensor also described in the polar coordinate system and nothing else. Only then, Koobee Wublee following Schwarzschild's footsteps would be able to attempt to solve these 3 differential equations. Again, the solution can only be applied to the polar coordinate system and no fvcking others. <shrug>
Footsteps don't matter. Correctness matters.
--
Jan