From: PD on
On Jun 17, 4:33 pm, colp <c...(a)solder.ath.cx> wrote:
> On Jun 18, 8:14 am, PD <thedraperfam...(a)gmail.com> wrote:
>
>
>
> > On Jun 16, 1:25 am, colp <c...(a)solder.ath.cx> wrote:
>
> > > The classic twin paradox is asymmetric in that one twin remains on
> > > Earth while the other leaves (i.e. only one of them accelerates and
> > > deaccelerates). In the symmetric twin paradox both twins leave Earth,
> > > setting out in opposite directions and returning to Earth at the same
> > > time. The conventional explanation for the classic twin paradox is
> > > since only one twin accelerates, the ages of the twins will be
> > > different. In the symmetric case this argument cannot be applied.
>
> > > The paradox of the symmetric twins is that according to special
> > > relativity (SR) each twin observes the other twin to age more slowly
> > > both on the outgoing leg
> > > and the return leg, so SR paradoxically predicts that each twin will
> > > be younger than
> > > the other when they return to Earth.
>
> > No. This is a basic misunderstanding due to oversimplification, and it
> > is exactly the kind of thing that the original puzzle was intended to
> > highlight for learners of relativity.
>
> It is true that I haven't discussed what happens at turnaround, but
> only for the reason that turnaround cannot possibly compensate for the
> SR time dilation.

This is where you may be surprised. :>)

> Imagine what would happen if a twin's observation of
> the other twin's clock was based on counting the number of pulses in a
> radio wave. When the twins return to Earth the number of pulses
> received by a twin must equal the number sent in order to avoid a
> paradox. SR predicts that the observed pulse rate will decrease during
> the outbound and inbound legs (ignoring the complication of the
> transit time of the signals), and SR does not describe an increased
> pulse rate which would reconcile the paradox.

Yes, it does.

> So, for the paradox to
> be resolved the observed pulse rate must increase according to the
> local rotation at turnaround, which could involve partial reversals.

Yes.

> The problem is that that signals exchanged by the twins can also be
> observed on Earth, and you have the illogical situation where, for an
> observer on Earth, the signal rate of one twin depends on the rotation
> of the other twin - remember that it is the same signals that pass
> Earth that are received by the twins.

Yes. There is no illogic.

http://crib.corepower.com:8080/~relfaq/twin_paradox.html

Try reading through all of this carefully.
It's free, openly available.

>
>
>
> > The novice says, "The moving clock runs slow" and considers that
> > effect during the outbound and incoming legs of the travel, but no
> > consideration is given for what happens to the relative reading of
> > clocks during the turnaround. And in fact, the analysis of the
> > original puzzle reveals some pretty amazing things about what happens
> > during that turnaround.
>
> Yep, so amazing that the faith of the relativist makes all possibility
> of a paradox disappear.

No faith required. Just a little deeper understanding.

I think you have put together a stance that says, "If there is no
paradox, then this should be obvious from even a superficial
inspection. If someone tells me that I need to look at it harder, then
this is really just an invitation to stare at it long enough until I
believe it." Nothing could be further from the truth. You REALLY CAN
fool yourself with a shallow grip on a subject, and it leads precisely
to this kind of problem.

>
>
>
> > So SR does not in fact say that the twins in the symmetric case will
> > both be younger than the other.
>
> Wrong. In no case does SR predict a time compression which would
> compensate for the time dilation that is predicted by SR.

I'm sorry, but you're flat wrong, and it comes from the superficial
statement that "moving clocks run slow". It's more complicated than
that. The proper time experienced by a clock depends on the *shape* of
the worldline the clock follows.

> Without such
> compensation the local time of the twin can never agree with the
> observed time of the other twin. But the observed time must agree with
> the twin's local time when the twins return to Earth otherwise a
> paradox ensues.

From: PD on
On Jun 18, 5:15 am, "k...(a)nventure.com" <k...(a)nventure.com> wrote:

>
> You are correct in stating twin paradox is a gedanken, and not
> an empirical experiment.
> Everone understands (or should understand) that a thought
> experiment is very susceptible to errors.
> However, just because an empirical experiment is conducted,
> this does not mean the researchers arrive at the correct
> conclusions.

The only conclusion arrived from an experiment is whether its results
agree with the predictions of a theory or not. Very often, competing
theories make different predictions about what will be observed, and
in such cases, the conclusion from the experiment is which theory
agrees with the measurement and which theory does not.

As for your claim that experiments are not to be trusted, this is the
reason that experimental results are taken with a grain of salt until
confirmed by an independent investigator, usually with complementary
methods, which tends to isolate sources of error not already accounted
for. If however you are taking the stand that experimental results are
untrustworthy notwithstanding this verification, then my only response
is that you don't like the scientific method.

>
> Nevertheless, what you do not seem to realize and be able to
> accept is that the whole of Einstein's SR and GR are based
> on gedankens, and not a bit of these are based on empirical
> experimentation!

This is a freshman mistake. Einstein used gedankens to explain his
reasoning with SR and GR. They are a pedagogical device only. However,
relativity wasn't taken very seriously until its real experimental
predictions started getting confirmed in real measurements.

>
> You and most relativists also cannot seem realize that post
> Einsteinian Special Relativity and General Relativity are not
> Einstein's SR or GR (e.g., relativistic mass is now invariant
> mass, time contracrion is now dilated time, etc.,). So many
> relativists cannot seem to accept that both are whole new
> theories of relativity, wherein any and all the reasoning and
> mathematics pertaining to mass and contracted time within
> Einstein's SR and GR just do not apply in these 2 new
> theories.

Physicists do not INVENT laws of physics. They *discover* them. They
are not creations like poems or paintings. They exist outside of us,
and our description of them changes a little bit as we get to know
them more. Relativity is an aspect of nature, first described by
Galileo, then the description was expanded quite a bit by Einstein
with SR, and then expanded quite a bit more by Einstein with GR.

Likewise, the original discoverers of dinosaurs mis-strung the found
fossilized leg bones, giving the dinosaurs a stance like a lizard and
therefore led to their characterization as prehistoric reptiles. It
wasn't until much later that it was found that the correct orientation
of the bones made them more closely related to birds, so that a T-Rex
is a closer cousin to a chicken than to crocodile. Nevertheless,
dinosaurs are dinosaurs and nothing about them has changed even though
our understanding of them has slowly modified over time.

>
> Furthermore, the twin paradox cannot be empirically
> modeled or duplicated.

This is simply incorrect. The GPS system includes the results of the
twin paradox in its operation.

>
> So all are open to the paradox of very rapidly rotating,
> perfectly round, RIGID discs (e.g., a Frisbee) moving
> near the speed of light wherein length, but not height
> and width contracts as tensors (so the physical form
> must continually change depending on the speed and
> rotation).

Lorentz contraction is not a mechanical deformation. If you thought it
was, then you need to learn again what relativity in fact says.

>
> Furthermore, all these continual changes of form (that
> depends of both the rate of spin and change of position)
> must occur without generating any heat and/or sound.
> (Sound is the physical aural result of vibration and is a
> form of energy). So if the deformations responsible for
> sound requires energy, how does this affect the tensor
> mechanics without also violating the Laws of
> Thermodynamics. Moreover, the need for this
> deformation energy would then prevent Einstein’s uniform
> motion that SR is all about.
>
> In other words; the deformations would also ‘tire’ inertia,
> and the motion would not be uniform, but nonuniform,
> i.e., deceleration, so only explainable by GR, because SR
> would be invalid if uniform motion is a fallacy. But how
> can GR be valid if both SR and the Lorentz Transformation
> are both invalid?
>
> So are not the whole ideas that Nature is time symmetrical
> and the tensors of Einstein in the same category as the
> notion of the relativity of mass?
>
> D.Y. Kadoshima

From: colp on
On Jun 19, 3:27 am, Tom Roberts <tjroberts...(a)sbcglobal.net> wrote:
> colp wrote:
> > It is not necessary for me to showing you the math in order for you to
> > identify the errors in the article.
>
> The basic error in that article is that they DID NOT use the math of SR.

That isn't necessarily an error. Can you show how their math resulted
in coming to an incorrect conclusion?

> Instead
> they used a comic-book description of SR such as "moving clocks run slow" -- SR
> does NOT say that;

The truth is not determined by what SR says because of the paradoxes
which may arise from it's application.
Is their description innacurate, or are you resorting to an ad-
hominem?

> what SR says is that if a moving clock's tick rate is
> measured by using two SYNCHRONIZED clocks at rest in an inertial frame and
> pre-positioned along the moving clock's path, then between meetings the moving
> clock ticks fewer times than the inertial-frame clocks. In particular, the
> METHOD of comparing the clocks is important.

How is the method important in the context of the experiment described
in the paper?
From: colp on
On Jun 19, 3:57 am, PD <thedraperfam...(a)gmail.com> wrote:
> On Jun 17, 4:33 pm, colp <c...(a)solder.ath.cx> wrote:
>
>
>
> > On Jun 18, 8:14 am, PD <thedraperfam...(a)gmail.com> wrote:
>
> > > On Jun 16, 1:25 am, colp <c...(a)solder.ath.cx> wrote:
>
> > > > The classic twin paradox is asymmetric in that one twin remains on
> > > > Earth while the other leaves (i.e. only one of them accelerates and
> > > > deaccelerates). In the symmetric twin paradox both twins leave Earth,
> > > > setting out in opposite directions and returning to Earth at the same
> > > > time. The conventional explanation for the classic twin paradox is
> > > > since only one twin accelerates, the ages of the twins will be
> > > > different. In the symmetric case this argument cannot be applied.
>
> > > > The paradox of the symmetric twins is that according to special
> > > > relativity (SR) each twin observes the other twin to age more slowly
> > > > both on the outgoing leg
> > > > and the return leg, so SR paradoxically predicts that each twin will
> > > > be younger than
> > > > the other when they return to Earth.
>
> > > No. This is a basic misunderstanding due to oversimplification, and it
> > > is exactly the kind of thing that the original puzzle was intended to
> > > highlight for learners of relativity.
>
> > It is true that I haven't discussed what happens at turnaround, but
> > only for the reason that turnaround cannot possibly compensate for the
> > SR time dilation.
>
> This is where you may be surprised. :>)

So surprise me.

>
> > Imagine what would happen if a twin's observation of
> > the other twin's clock was based on counting the number of pulses in a
> > radio wave. When the twins return to Earth the number of pulses
> > received by a twin must equal the number sent in order to avoid a
> > paradox. SR predicts that the observed pulse rate will decrease during
> > the outbound and inbound legs (ignoring the complication of the
> > transit time of the signals), and SR does not describe an increased
> > pulse rate which would reconcile the paradox.
>
> Yes, it does.

An unsupported assertion from a relativist in the face of a paradox is
hardly a surprise.

And you are most definitely wrong in asserting that SR describes an
increased pulse rate.

>
> > So, for the paradox to
> > be resolved the observed pulse rate must increase according to the
> > local rotation at turnaround, which could involve partial reversals.
>
> Yes.
>
> > The problem is that that signals exchanged by the twins can also be
> > observed on Earth, and you have the illogical situation where, for an
> > observer on Earth, the signal rate of one twin depends on the rotation
> > of the other twin - remember that it is the same signals that pass
> > Earth that are received by the twins.
>
> Yes. There is no illogic.

Your statement is self-contradictory. You are agreeing with my
statement which describes the illogic, and yet you also deny that it
exists.

>
> http://crib.corepower.com:8080/~relfaq/twin_paradox.html
>
> Try reading through all of this carefully.
> It's free, openly available.
>
>
>
> > > The novice says, "The moving clock runs slow" and considers that
> > > effect during the outbound and incoming legs of the travel, but no
> > > consideration is given for what happens to the relative reading of
> > > clocks during the turnaround. And in fact, the analysis of the
> > > original puzzle reveals some pretty amazing things about what happens
> > > during that turnaround.
>
> > Yep, so amazing that the faith of the relativist makes all possibility
> > of a paradox disappear.
>
> No faith required. Just a little deeper understanding.

How do you tell the difference between faith and a deep understanding?

>
> I think you have put together a stance that says, "If there is no
> paradox, then this should be obvious from even a superficial
> inspection. If someone tells me that I need to look at it harder, then
> this is really just an invitation to stare at it long enough until I
> believe it." Nothing could be further from the truth. You REALLY CAN
> fool yourself with a shallow grip on a subject, and it leads precisely
> to this kind of problem.

So why can't you explain what I am missing?

>
>
>
> > > So SR does not in fact say that the twins in the symmetric case will
> > > both be younger than the other.
>
> > Wrong. In no case does SR predict a time compression which would
> > compensate for the time dilation that is predicted by SR.
>
> I'm sorry, but you're flat wrong, and it comes from the superficial
> statement that "moving clocks run slow".

Non sequitur. The statement is not superficial, although it does lack
context.

> It's more complicated than
> that. The proper time experienced by a clock depends on the *shape* of
> the worldline the clock follows.

Can you give an example where SR predicts that an observer will
observe a moving clock to be running fast?

>
> > Without such
> > compensation the local time of the twin can never agree with the
> > observed time of the other twin. But the observed time must agree with
> > the twin's local time when the twins return to Earth otherwise a
> > paradox ensues.

From: PD on
On Jun 18, 4:04 pm, colp <c...(a)solder.ath.cx> wrote:
> On Jun 19, 3:57 am, PD <thedraperfam...(a)gmail.com> wrote:
>
>
>
> > On Jun 17, 4:33 pm, colp <c...(a)solder.ath.cx> wrote:
>
> > > On Jun 18, 8:14 am, PD <thedraperfam...(a)gmail.com> wrote:
>
> > > > On Jun 16, 1:25 am, colp <c...(a)solder.ath.cx> wrote:
>
> > > > > The classic twin paradox is asymmetric in that one twin remains on
> > > > > Earth while the other leaves (i.e. only one of them accelerates and
> > > > > deaccelerates). In the symmetric twin paradox both twins leave Earth,
> > > > > setting out in opposite directions and returning to Earth at the same
> > > > > time. The conventional explanation for the classic twin paradox is
> > > > > since only one twin accelerates, the ages of the twins will be
> > > > > different. In the symmetric case this argument cannot be applied.
>
> > > > > The paradox of the symmetric twins is that according to special
> > > > > relativity (SR) each twin observes the other twin to age more slowly
> > > > > both on the outgoing leg
> > > > > and the return leg, so SR paradoxically predicts that each twin will
> > > > > be younger than
> > > > > the other when they return to Earth.
>
> > > > No. This is a basic misunderstanding due to oversimplification, and it
> > > > is exactly the kind of thing that the original puzzle was intended to
> > > > highlight for learners of relativity.
>
> > > It is true that I haven't discussed what happens at turnaround, but
> > > only for the reason that turnaround cannot possibly compensate for the
> > > SR time dilation.
>
> > This is where you may be surprised. :>)
>
> So surprise me.

I gave you a link to a collection of pages about the twin puzzle that
deals with this precisely. Read that. Then we can discuss.

>
>
>
> > > Imagine what would happen if a twin's observation of
> > > the other twin's clock was based on counting the number of pulses in a
> > > radio wave. When the twins return to Earth the number of pulses
> > > received by a twin must equal the number sent in order to avoid a
> > > paradox. SR predicts that the observed pulse rate will decrease during
> > > the outbound and inbound legs (ignoring the complication of the
> > > transit time of the signals), and SR does not describe an increased
> > > pulse rate which would reconcile the paradox.
>
> > Yes, it does.
>
> An unsupported assertion from a relativist in the face of a paradox is
> hardly a surprise.
>
> And you are most definitely wrong in asserting that SR describes an
> increased pulse rate.

Read the pages at the link I provided.

>
>
>
> > > So, for the paradox to
> > > be resolved the observed pulse rate must increase according to the
> > > local rotation at turnaround, which could involve partial reversals.
>
> > Yes.
>
> > > The problem is that that signals exchanged by the twins can also be
> > > observed on Earth, and you have the illogical situation where, for an
> > > observer on Earth, the signal rate of one twin depends on the rotation
> > > of the other twin - remember that it is the same signals that pass
> > > Earth that are received by the twins.
>
> > Yes. There is no illogic.
>
> Your statement is self-contradictory. You are agreeing with my
> statement which describes the illogic, and yet you also deny that it
> exists.
>
>
>
>
>
> >http://crib.corepower.com:8080/~relfaq/twin_paradox.html

The above is the link I keep referring to. Read it.

>
> > Try reading through all of this carefully.
> > It's free, openly available.
>
> > > > The novice says, "The moving clock runs slow" and considers that
> > > > effect during the outbound and incoming legs of the travel, but no
> > > > consideration is given for what happens to the relative reading of
> > > > clocks during the turnaround. And in fact, the analysis of the
> > > > original puzzle reveals some pretty amazing things about what happens
> > > > during that turnaround.
>
> > > Yep, so amazing that the faith of the relativist makes all possibility
> > > of a paradox disappear.
>
> > No faith required. Just a little deeper understanding.
>
> How do you tell the difference between faith and a deep understanding?
>
>
>
> > I think you have put together a stance that says, "If there is no
> > paradox, then this should be obvious from even a superficial
> > inspection. If someone tells me that I need to look at it harder, then
> > this is really just an invitation to stare at it long enough until I
> > believe it." Nothing could be further from the truth. You REALLY CAN
> > fool yourself with a shallow grip on a subject, and it leads precisely
> > to this kind of problem.
>
> So why can't you explain what I am missing?

I referred you to some free, open-access online reading, prepared
precisely for people like you. If you have difficulty understanding
it, then ask about the particular places where you have trouble.

>
>
>
> > > > So SR does not in fact say that the twins in the symmetric case will
> > > > both be younger than the other.
>
> > > Wrong. In no case does SR predict a time compression which would
> > > compensate for the time dilation that is predicted by SR.
>
> > I'm sorry, but you're flat wrong, and it comes from the superficial
> > statement that "moving clocks run slow".
>
> Non sequitur. The statement is not superficial, although it does lack
> context.
>
> > It's more complicated than
> > that. The proper time experienced by a clock depends on the *shape* of
> > the worldline the clock follows.
>
> Can you give an example where SR predicts that an observer will
> observe a moving clock to be running fast?
>
>
>
> > > Without such
> > > compensation the local time of the twin can never agree with the
> > > observed time of the other twin. But the observed time must agree with
> > > the twin's local time when the twins return to Earth otherwise a
> > > paradox ensues.
>
>

First  |  Prev  |  Next  |  Last
Pages: 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Prev: Black Hole is Black Day for Earth
Next: n-stars.