Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Re: Ritz's law and Galilei's principle.

266 views
Skip to first unread message
Message has been deleted

Alan Folmsbee

unread,
Aug 24, 2016, 12:05:49 PM8/24/16
to
On Tuesday, August 23, 2016 at 8:59:56 PM UTC-10, V. Trovò wrote:
> In www.vitotrovo.it site, I published an important result: Ritz's law is a necessary condition for the validity of Galilei's principle.
>
> V. Trovò

I did not look at the link. What is Ritz' Law?
Message has been deleted

Paul B. Andersen

unread,
Aug 24, 2016, 3:11:52 PM8/24/16
to
On 24.08.2016 19:44, V. Trovò wrote:
> The law of W.Ritz and M. La Rosa, known in Italy as ballistic law.
> - On the properties of light- it contains the demonstration.

> www.vitotrovo.it

From where I quote:
"I prove that Einstein postulate and Galileo's principle
of Relativity are incosistent from the logical viewpoint."

"I invite the reader to study carefully this work and to find,
if there is a hypothetic logic mistake. If that is discovered,
I commit to give the first person, an award of ten thousand euro."

Why won't I bother to study this work? :-D

--
Paul

https://paulba.no/

Gary Harnagel

unread,
Aug 24, 2016, 10:20:23 PM8/24/16
to
Thanks for the heads up, Paul, it's obvious that "logic" has nothing to do
with whether the Principle of Relativity is correct. The fact is that it
has NEVER been falsified and it has always been confirmed.

Gary
Message has been deleted
Message has been deleted
Message has been deleted

Gary Harnagel

unread,
Aug 25, 2016, 7:43:28 AM8/25/16
to
On Thursday, August 25, 2016 at 1:01:58 AM UTC-6, V. Trovò wrote:
>
> ............. I prove that Einstein postulate and Galileo's principle of Relativity are INCONGRUENT (no inconsistent) from the logical viewpoint.

Then your logic is flawed.
Message has been deleted

Dono,

unread,
Aug 25, 2016, 9:52:47 AM8/25/16
to
On Thursday, August 25, 2016 at 12:01:58 AM UTC-7, V. Trovò wrote:

> ............. I prove that Einstein postulate and Galileo's principle of Relativity are INCONGRUENT (no inconsistent) from the logical viewpoint.


Another day, another delusional nutter.

Dono,

unread,
Aug 25, 2016, 9:53:30 AM8/25/16
to
On Thursday, August 25, 2016 at 6:43:33 AM UTC-7, V. Trovò wrote:
> then I will lose ten thousand euro.

No, crooks like you never pay up. You are just trolling, mr. Trovo.
Message has been deleted

Python

unread,
Aug 25, 2016, 10:01:56 AM8/25/16
to
Le 25/08/2016 à 16:00, V. Trovò a écrit :
> In Italy, the attached declaration has legal force.

How could a law convince a mentally ill person that he is wrong, even
in presence of a rational and convincing refutation?



Message has been deleted

Python

unread,
Aug 25, 2016, 10:12:17 AM8/25/16
to
Le 25/08/2016 à 16:06, V. Trovò a écrit :
> Talk about the Italian courts?

Parlando di Matematica e Fisica in tribunale? Stai scherzando?

Message has been deleted

Python

unread,
Aug 25, 2016, 10:56:34 AM8/25/16
to
Le 25/08/2016 à 16:20, V. Trovò a écrit :
> No mathematics and physics, but logical!
> Do not underestimate those who have made humanities.

This is not underestimation, it is just none of their business...



Odd Bodkin

unread,
Aug 25, 2016, 11:17:57 AM8/25/16
to
On 8/25/2016 9:56 AM, Python wrote:
>> No mathematics and physics, but logical!
>> Do not underestimate those who have made humanities.
>
> This is not underestimation, it is just none of their business...

This is where I think there's a fundamental disconnect.
Intelligence and logic and analytical thinking are common to all
disciplines, but they are not sufficient for expertise in the
discipline. To be an expert in the humanities requires background study
and ancillary skills that are special to the humanities.
To be an expert in physics requires background study and ancillary
skills that are special to physics.

To say, "I have done well in humanities and so I have logic and
intelligence and analytical thinking, and so that should be sufficient
to contribute to physics," is to demean the importance of the things
beyond logic and intelligence and analytical thinking that are required
to do that in reality.

Sure a brain surgeon is smart. This does not make him a competent
economist. Sure a software developer is smart. This does not make him a
competent physicist.

--
Odd Bodkin --- maker of fine toys, tools, tables
Message has been deleted

Dono,

unread,
Aug 25, 2016, 11:52:30 AM8/25/16
to
On Thursday, August 25, 2016 at 8:41:27 AM UTC-7, V. Trovò wrote:
>
> Read the statement.
> let it translate.
> The judges are the scientific committees:
> S.I.F. (Società Italiana di Fisica)
> European Journal of Physics
> Foundations of Physics
> Electronic Journal of Theoretical Physics

But all these forums treat you as a lunatic.

Python

unread,
Aug 25, 2016, 11:57:19 AM8/25/16
to

As a matter of fact, your kind of arguments has been seen
seen in a paper long ago:

On the consistency of the postulates of special relativity
James M. Kingsley
http://link.springer.com/article/10.1007/BF00717444

and has been debunked too:

http://link.springer.com/article/10.1007/BF00708946

Both articles are quite easy to find without giving a penny
to the Springer's gang I guess...


Message has been deleted

Python

unread,
Aug 25, 2016, 12:53:37 PM8/25/16
to
Le 25/08/2016 à 18:50, V. Trovò a écrit :
> Thanks for the link, I'll try the articles.
> Meanwhile, warn James M. Kingsley, to get his opinion is.
> He is certainly authoritative person, more than me.

Authoritative or not, he was pretending nonsense, no more,
no less than you are.



Gary Harnagel

unread,
Aug 25, 2016, 3:26:29 PM8/25/16
to
On Thursday, August 25, 2016 at 7:43:33 AM UTC-6, V. Trovò wrote:
> Il giorno giovedì 25 agosto 2016 13:43:28 UTC+2, Gary Harnagel ha scritto:
> then I will lose ten thousand euro.

And how does that matter to me?

Rafael Valls Hidalgo-Gato

unread,
Aug 25, 2016, 4:17:23 PM8/25/16
to
El miércoles, 24 de agosto de 2016, 2:59:56 (UTC-4), V. Trovò escribió:
> In www.vitotrovo.it site, I published an important result: Ritz's law is a necessary condition for the validity of Galilei's principle.
>
> V. Trovò

I took already a first rapid glance at your very interesting paper, sufficient
to understand that it seems involving a contradiction between the two original
postulates of 1905 Einstein Relativity, what I denote as 1905R to distinguish
it from the very different Special Relativity (SR) of our days; the possibility
of that contradiction is already addressed by himself since the Introduction of
his first Relativity paper.

Following the scientific method, the postulates of any theory can be objected
only with the available adverse experimental evidence against them.

However, you seem ignoring that since already almost four decades, the very
accurate Global Positioning System (GPS) operation is giving to the original
1905R theory a huge very favorable experimental support. In particular, all GPS
receptors are resolving continuously systems of very simple 1686 Newtonian
equations, with Euclidean geometry, Cartesian coordinates and space separated
from the relative time carefully defined by 1905 Einstein for what he denotes a
stationary system, as the Earth-Centered Inertial (ECI) one used in the GPS
(the same one already introduced by himself at the end of Section 4 in his
Jun30 paper).

Fast moving satellites in the GPS are periodically broadcasting signals in all
directions with a constant isotropic speed c in the corresponding stationary
space of the ECI system, in total agreement with 1905R two postulates.

Besides, I note that you are assuming that the Lorentz transform applies
between any two relative moving entities, taking for granted that anyone of
them can be considered at rest with equal right. You also seem then ignoring
than in 1905R a clear distinction is made between a stationary system and a
moving one in it, being the relative time defined only to a stationary system,
being also the unique one where the second postulate applies.

A concrete physical example can help you to understand better what I am saying.
The Solar System (SS) is moving in the Galaxy as part of it, but the
Galaxy is not moving in the SS, what would imply a flagrant violation of the
Newton's Laws that apply always in any stationary system (like the Galaxy, SS
and GPS ECI ones). In a similar way, the GPS ECI system is moving in the SS as
part of it, but the SS is NOT moving in the ECI one.

Before debating about the nature of movement, space and time, I recommend you
strongly to take a glance at the 1905 Einstein's Jun30 paper (at least the
Introduction and Section 1). I am pretty sure that doing that you will be
curious enough to know also in detail 1686 Newton's three Axioms or Laws of
motion. By the way, Corollary V is Galileo's Principle of Relativity,
generalized from zero acceleration (uniform velocity ship) to any variable
acceleration with the Corollary VI. Already in Corollary IV, the Center of
Gravity (today Center of Mass) system of an interacting body set is introduced,
with very interesting properties.

RVHG (Rafael Valls Hidalgo-Gato)
Message has been deleted
Message has been deleted

Python

unread,
Aug 26, 2016, 5:58:34 AM8/26/16
to
Le 26/08/2016 à 11:18, V. Trovò a écrit :
> ... Questo risultato non avrebbe alcun valore se prima non avessi dimostrato che la “legge balistica”
> condizione necessaria per la validità del principio di Galilei.

É totalmente impossibile che questa proposizione e vera, é
matematicalmente impossibile.

Ha stato demonstrato sule unique basi della teoria del grupos e del
principio di Galilei che la unica familia de transformazioni possibile
e questa:

|t'| | 1 kv | |t|
| | = 1/sqrt(1+kv^2) * | | * | |
|x'| |-v 1 | |x|

k is a free parameter k <= 0

con k = 0 questo é Galilean Transformations
con k < 0 questo é Lorentz Transformations, e facile de vedere che
sqrt(-k) é, alora, la massima velocita.







Rafael Valls Hidalgo-Gato

unread,
Aug 26, 2016, 10:15:41 AM8/26/16
to
El viernes, 26 de agosto de 2016, 5:18:58 (UTC-4), V. Trovò escribió:
> Dear Rafael
> thank you for your intervention, aimed at confrontation and not to the offense.
> Your intervention is structured, and requires an articulated response. Please, allow me to answer you in Italian, my native language.
>
> Il punto centrale della questione è il ruolo della logica nella scoperta scientifica;
> più precisamente, il ruolo degli esperimenti concettuali.
> Affronterò il problema rispondendo alle seguenti domande.
> - Si deve privilegiare il dato empirico o l'esperimento concettuale?
> - Il fatto che una legge empirica possa essere inquadrata in un dato ambito teorico, è sufficiente per affermare che la teoria rappresenta la realtà fisica?
>
> La prima domanda può essere parafrasata nel seguente modo: il fatto che tutte le macchine termiche reali non riproducano perfettamente il ciclo di Carnot (esempio supremo di esperimento concettuale) significa che il teorema di Carnot, con tutti i suoi corollari, è falso? A parte i pochi che ancora oggi si esercitano nella ricerca del moto perpetuo, penso che nessuno si disposto a mettere in discussione il teorema di Carnot. Ma come tu sai, gli esperimenti concettuali sono stati strumenti formidabili per isolare i principi cardine della fisica. Oggi non c’è alcun motivo perché il loro ruolo sia sminuito. Tutti hanno usato esperimenti concettuali, da Galilei, passando per Kirchhoff, fino a Faymann.
> In definitiva, sulla risposta a questa prima domanda non ho dubbi: le verità che si conseguono attraverso esperimenti concettuali devono prevalere sull’interpretazione dei dati empirici.
> Vengo ora alla seconda domanda.
> Anche in questo caso userò un esempio concreto, esplicativo.
> Tutti sanno che le due teorie oggi in vigore, quella di Einstein e la Meccanica Quantistica, sono teorie antitetiche. La prima è una teoria deterministica, per certi aspetti ricade nell’ambito della fisica classica. La secondo è una teoria probabilistica, completamente estranea alla fisica classica. Ciò che le rende antitetiche è il “principio di indeterminazione”: nell’ambito delle teoria di Einstein il moto di una particella che si muove di moto rettilineo uniforme, può essere determinato ( in linea di principio) con assoluta certezza. Questo invece non è possibile nell’ambito della Meccanica Quantistica. Ora, sono molti le leggi empiriche che hanno un duplice inquadramento teorico: relativistico e quantistico. L’effetto Compton è una di queste.
> La questione che si pone è quindi la seguente: se le due teorie sono antitetiche, qual è la vera natura dell’effetto Compton? Relativistica o Quantistica?
> Rispondere “entrambe” è come attribuire un comportamento schizofrenico alla natura. É possibile fidarsi delle risposte di uno schizofrenico?
> Quindi, il fatto che una legge empirica possa essere derivata nell’ambito di una data teoria non significa che la teoria rappresenti la realtà.
> D’altra parte, ho mostrato che dalla legge di Ritz-La Rosa si possono derivare le formule dell’ “effetto Doppler” e dell’ “effetto Sagnac”. Questo risultato non avrebbe alcun valore se prima non avessi dimostrato che la “legge balistica” è condizione necessaria per la validità del principio di Galilei. La legge di Ritz-La Rosa è logicamente coerente con tutti i principi e le leggi della meccanica di Newton. A questo punto immagino che la tua obiezione sia: e le equazioni di Maxwell? Se mi sarà data l’opportunità, dopo che la comunità scientifica si sarà espressa positivamente rispetto alla legge di Ritz-La Rosa renderò pubblica la parte del mio lavoro sulle equazioni di Maxwell.
> In definitiva, la questione è la seguente: nel mio esperimento concettuale c’è qualche errore logico? il funzionamento del dispositivo è quello da me descritto ?
> Spero di aver risposto alle tue domande, malgrado non sia entrato nel merito delle singole questioni che tu mi hai posto.
>
> Grazie, per le tue obiezioni, mi hanno offerto l’occasione di esprimere il mio pensiero.
>
> Cordialmente, V. Trovò.

My native language is Spanish, fortunately sharing with Italian the Latin as a
common origin, making me able to understand the essential of your post (I think
so), but not without some effort. By the way, I was on 1971 at the ICTP
(Trieste) attending a one month scientific activity, being then the director
the future Nobel Prize Abdus Salam (in those days P. Dirac was making a visit
to the Center).

I do not think that in the scientific method (Galileo is a pioneer in its
development) exists any priority between theory and practice, both are the
faces of the same coin. The point is not if you have or not a logic error in
your derivation (more probable it does not exist at all), but if from the
premises that you take for granted, an agreement is reached or not between
theory and practice.

The not existence of a logic error does not assure the validity of any theory
derivation; and the simple accumulation of empirical data means absolutely
nothing (from the scientific point of view) if you have not a theory in
agreement with them.

In my comment I include many specific assertions about 1686 Newton and 1905
Einstein theories. I expect from you to express an opinion about them, clearing
if you accept or not them, pointing at least one that you consider false or not
valid for any reason.

From my part I pointed already my not agreement with the symmetric character
that you take for granted in the Lorentz transform mentioned on your
derivation, referring the available huge experimental evidence to support my
position.

I appreciate a lot the respectful character of our dialogue, not common at all
in this group when debating any thing.

RVHG (Rafael Valls Hidalgo-Gato)

nero

unread,
Aug 26, 2016, 10:50:15 AM8/26/16
to
Il giorno mercoledì 24 agosto 2016 08:59:56 UTC+2, V. Trovò ha scritto:
> In www.vitotrovo.it site, I published an important result: Ritz's law is a necessary condition for the validity of Galilei's principle.
>
> V. Trovò

...perhaps you can find interesting ( a confirmation !? ) toread the topic in this group , named ' Lessons of English language ' ...
oldogf by yahoo dot it
Message has been deleted
Message has been deleted

Python

unread,
Aug 27, 2016, 5:41:40 AM8/27/16
to
Le 27/08/2016 à 10:14, V. Trovò a écrit :
> I can not decipher the formula well.

This is Lorentz Transformation where c is written as sqrt(-k)
and k<=0

You do know linear algebra and matrix product, don't you?



Message has been deleted

Python

unread,
Aug 27, 2016, 6:57:50 AM8/27/16
to
Le 27/08/2016 à 12:38, V. Trovò a écrit :
> Yes! I could not decipher the writing.
> The proof of which you speak, I know her.
> It is built on the condition that the Maxwell equations must
> be invariant in shape.

No. Maxwell equations are not involved at all in the proof. You
could have guessed, as 'c' (electromagnetic waves speed) is not
the parameter. I told you, the 'k' parameter is *free*.

ONLY group theory and Galilean relativity principle are involved
in the proof.

This mathematical fact completely condemn the validity of your
"theorem", hence of your "proof".



Message has been deleted

Python

unread,
Aug 27, 2016, 10:55:16 AM8/27/16
to
Le 27/08/2016 à 16:43, V. Trovò a écrit :
> That c = sqrt (-k) with k free parameter, it does not matter, k is
> a constant within the group of transformations. For each value of k
> corresponds to a fixed value of the speed of light

No. Light plays NO ROLE in the demonstration. The only thing you
could deduce is that sqrt(-k) is a maximum and invariant speed.
Nothing linked it to the speed of light.

> (no matter its value); this is equivalent to saying that the speed
> of light is independent of the speed of the source; that is to deny
> a priori the ballistic law; i.e., a priori assume as true the Einstein
> postulated.

Not in this demonstration, it is NOT based on any postulate about light
(or any other kind of propagation).

The link between c (as the speed of light) and k, can be done but only
later on experimental basis.

You're still trapped: proving that non-ballistic light is in
contradiction with relativity is mathematically impossible.

Sorry, but this is a simple fact.



Message has been deleted

Tom Roberts

unread,
Aug 28, 2016, 1:14:52 PM8/28/16
to
On 8/27/16 8/27/16 10:32 AM, V. Trovò wrote:
> Inside the Lorentz transformations is the implicit postulate of Einstein: c
> is an absolute constant.

Yes. But what does "c" represent?
A) In Einstein's original derivation it represented the vacuum speed of
light.
B) In a modern derivation, in which geometry is separated from
electrodynamics, it represents an invariant speed which is also the
maximum speed of massive objects (relative to any inertial frame).

Python has been trying to explain to you that (B) is DIFFERENT from (A). The
resulting transforms are identical, but the basis from which they are derived is
DIFFERENT.

In particular, (B) requires only the following, without any mention of light or
electrodynamics:
1. Einstein's principle of relativity (PoR)
2. the definition of inertial frames
3. Einstein's "hidden" postulates:
3a clocks and rulers have no memory (for differential measurements)
3b space is isotropic and homogeneous (implied by 2)
3c time is homogeneous (implied by 2)
4. any experimental result that distinguishes SR from Galilean relativity,
such as pion beams exist longer than 1 kilometer

1-3 alone restrict the transformation group among inertial frames to one of
three: the Poincare' group (SR), the Galilei group (NM), and the Euclid group
(in 4D space and time). Experiments unequivocally refute the latter two.

This seems to have been displayed earlier in this thread,
but without recognizing the Euclid group for k>0.

Note that for any of these to form a group, c must be constant (in your words,
"an absolute constant").


> Nothing changes if "c" is replaced by the parameter
> "k"; it normalizes the speed of the bodies to that of light. When k = 0 there
> is no limit to the speed of the bodies; and in fact they obtain the
> transformations of Galilei. But I would like to make you see that in my
> thought experiment mathematics is almost superfluous. It comes to the
> conclusion without the use of algebra. In some respects it is a very similar
> demonstration to those who are in elementary geometry: where the reasoning by
> contradiction is important.

I have ignored this thread as it contains too much Italian. But it's absolutely
certain that any claim that an invariant speed of light is inconsistent with the
Lorentz transform is just plain wrong. And if light should happen to travel with
that invariant speed, and it is OBSERVED to do so (in vacuum), then light speed
cannot possibly depend on the speed of its source. This is a CONCLUSION based on
observations, not an "a priori postulate". It is also an experimental FACT.


Tom Roberts

Message has been deleted

Dono,

unread,
Aug 30, 2016, 9:44:03 AM8/30/16
to
On Tuesday, August 30, 2016 at 3:31:16 AM UTC-7, V. Trovò wrote:
>
> You speak of experimental proof of Einstein's postulate.
> The experimental proof of ballistic law are many: stellar aberration; Doppler effect; Sagnac effect; Michelson and Morley.
> See my paper.
Imbecile,

The ballistic theory fails the Sagnac experiment.

Tom Roberts

unread,
Aug 31, 2016, 9:32:19 PM8/31/16
to
On 8/30/16 8/30/16 5:31 AM, V. Trovò wrote:
> Einstein's principle of relativity (case B): all laws of physics must have
> the same form in all inertial reference systems; then also Maxwell's laws.

Yes, in that Maxwell's equations are "laws of physics" in an appropriate domain.


> The laws of Maxwell respect the principle of Einstein only for Lorentz
> transformations.

Yes, but that's OK as the Lorentz (Poincare') group is the only transformation
group not refuted by experiments.


> In the Lorentz transformations there is the constant "c". It
> expresses a limit speed for the dynamic of the bodies: bodies with mass
> (ponderable bodies), or without (light-electromagnetic waves). No entity
> (physical) may then exceed the limit speed "c = sqrt (-k).

The usual definition is k = 1/c^2. So for k=0, one gets "c ~ infinity" and
obtains the Galilean transforms. (The usual derivation uses k, not c, so there
is no actual problem with "infinity".)


> When k = 0: the
> laws of Maxwell are not invariant in shape;

Yes (for my definition of k, not yours, which I assume is a typo). No matter, as
a theory with that value is solidly refuted experimentally.


> therefore falls on the principle
> of Einstein.

I cannot parse this into sensible English.


> Ultimately; also in case B ) inside there is the postulate of
> Einstein if k is not zero (postulate of Einstein: the speed of light is an
> absolute constant; it does not depend on speed of the source). If k = 0,
> there is no limit to the speed of bodies = transformations of Galilei.
>
> Looks well my thought experiment: If you replace the light pulses with small
> pebbles what changes? ...........

Ditto. I have no idea what you are trying to say.

But we do know from experiments that k<0 (my definition above), and also that
the c in the Lorentz transform has the same value as the c in Maxwell's
equations, and the c in QED. That is, c is the vacuum speed of light.


> You speak of experimental proof of Einstein's postulate.

No. Never. It simply is not possible to "prove" a physical theory; nor is it
possible to "prove" a postulate (which is only part of a theory).

One can only test a theory, not a postulate. There are literally hundreds of
tests of Special Relativity, and none of them refute it. So SR is known to be
valid -- not "proven" -- within its domain of applicability.

And there are domains in which SR is not valid, such as
the curved manifolds of GR used to model gravitation.


> The experimental
> proof of ballistic law are many: stellar aberration; Doppler effect; Sagnac
> effect; Michelson and Morley. See my paper.

Again: it is not possible to "prove" a physical theory [@]. None of those
phenomena come "prove" any sort of ballistic theory. They are CONSISTENT with
some/most ballistic theories, but that does not "prove" anything. Moreover,
those same phenomena are also consistent with SR (plus Maxwell's equations or QED).

[#] Proof is possible only in the province of mathematics,
where the quantities involved are abstract and exactly
defined -- no phenomenon in the real world is exactly known.

But there are OTHER measurements and experiments that refute all ballistic
theories I have seen [#]. But all of these measurements and experiments are
consistent with SR.

[#] Around here, "HGWilson" attempts to advocate a mish-mash
he calls "BaTh", which seems to have some ballistic parts; it
is not a physical theory, as it changes with his whim and
acquires a special plead for each and every experiment with
which he is challenged.


Tom Roberts

mlwo...@wp.pl

unread,
Sep 1, 2016, 2:04:08 AM9/1/16
to
W dniu czwartek, 1 września 2016 03:32:19 UTC+2 użytkownik tjrob137 napisał:
> One can only test a theory, not a postulate. There are literally hundreds of
> tests of Special Relativity, and none of them refute it.

A lie, as expected from fanatic trash.
GPS shows time (as defined by your idiot guru)
is galilean with the precision of an acceptable
error.

Message has been deleted

Tom Roberts

unread,
Sep 1, 2016, 8:35:05 PM9/1/16
to
On 9/1/16 9/1/16 6:01 AM, V. Trovò wrote:
> When k=0, Maxwell equation are not invariable in shape for trasformation di
> Lorentz; therafore, in the case B), Einstein principle is not true: then,
> there is not any limit speed for bodies, with or without mass; therefore,in
> B) is not true the Einstein postulate.

Hmmmm. This means that k=0 is not consistent with Einstein's first postulate. We
also know k=0 (Galilean relativity) is inconsistent with many experiments -- SO
IT DOES NOT MATTER.


> If k is not zero, for all inertial
> frames,

In order for these transforms to form a group, k must be the same for all
members of the group, so different values of k yield different groups. The
behaviors of those groups depend only on the sign of k: k<0 => Poincare' group;
k=0 => Galilei group; k>0 => Euclid group. Only k<0 is consistent with
experiment, and we usually write k=-1/c^2 to obtain the usual formulas.


> there is a limit speed for bodies, with or without mass; It is a
> universal costant: but this is the Einstein's postulate! Therefore, when k is
> not zero, the Einstein principle contain the postulate Einstein: there is a
> limit at the speed of bodies in all inertial frames.

Sure. Well known.


> I speak of thought experiment (Gedankenexperiment) Many physicists of the
> past have used concetuali experiments to discover the principles and laws of
> physics: starting from Galileo, through Kirchhoff, ........... until
> Feynman.

OK.


> These experiments are based on logic; they show the internal contradictions
> to some assumptions; they show the true and false assumptions.

They can show internal inconsistencies and contradictions. They cannot possibly
show which assumptions are "true and false", because that depends on experiment
in the real world, not abstract logic as in such gedanken experiments.

And as I have said before, "true and false" are INAPPROPRIATE. Say, instead,
"valid" or "not valid"; the former imply proof, which is not possible for
physical theories and their postulates.


> I have thoroughly studied the problems generated by Einstein's
> theory. I came to the conclusion that it is not a general theory. It only
> applies to some physical systems. The GPS system, mentioned by Rafael Valls,
> is one of them. Unfortunately, in this place, I can not give the proof of
> this fact.

If by "Einstein's theory" you mean special relativity, this is true and well
known. Yes, it does not describe the GPS very well at all. Nor any other system
in which gravitation is significant.

Einstein's other major theory, General Relativity, is of considerably wider
domain, and it does describe the GPS extremely well.


Tom Roberts

Message has been deleted

Python

unread,
Sep 4, 2016, 6:14:48 AM9/4/16
to
Le 04/09/2016 à 11:01, V. Trovò a écrit :
> G = principle of Galilei
> E = Einstein postulated
>
> If (G) = 1, then (E) = 0
> If (G) = 0 then (E) = 1

This is NOT true. G and E are NOT in contradiction.

End of story.

Tom Roberts

unread,
Sep 6, 2016, 1:27:45 PM9/6/16
to
On 9/4/16 9/4/16 - 4:01 AM, V. Trovò wrote:
> [...]

I have no idea why you think that Galileo's principle and Einsteins Principle of
Relativity are mutually exclusive. They are not, and Einstein's first postulate
is merely a more modern restatement of Galileo's original principle.

So everything you said is nonsense.

Tom Roberts

Message has been deleted

Dono,

unread,
Sep 7, 2016, 9:09:52 AM9/7/16
to
On Tuesday, September 6, 2016 at 11:08:38 PM UTC-7, V. Trovò wrote:
> The principle of Galileo is more general principle of Einstein. The principle of Einstein says that the laws of physics must be invariant in shape.
> As you know, there are many laws of physics are not invariant in shape. For example, all the laws of thermodynamics.
>
> Vito Trovò

You are an imbecile, R.C. Tolman formulated the laws of thermodynamics on covariant form about 85 years ago.

Tom Roberts

unread,
Sep 7, 2016, 7:14:23 PM9/7/16
to
On 9/7/16 9/7/16 1:08 AM, V. Trovò wrote:
> The principle of Galileo is more general principle of Einstein.

Only because the principle of Galileo is not stated very precisely. Einstein was
MUCH more precise.


> The principle of Einstein says that the laws of physics must be invariant in
> shape.

Yes. Though your "in shape" is unusual English phrasing and is normally omitted.
And it is best to state under which operations they are invariant, as Einstein did.


> As you know, there are many laws of physics are not invariant in shape. For
> example, all the laws of thermodynamics.

The laws of thermodynamics are NOT "laws by which the states of physical systems
undergo change". So the PoR does not apply to them. In a very real sense, except
for the first they are not "laws", they are approximations. You'll find the same
applies to any "law" that cannot be expressed in invariant form.

Not even close -- to specify the state of 1 mole of an ideal
gas would require 6*6.02E23 real numbers.

The first law of thermodynamics (conservation of energy) can be expressed
invariantly, and is a direct consequence of the Einstein field equation of GR.
The 0th, 2nd, and 3rd laws cannot, in part because temperature is an important
aspect of them, and it cannot be defined covariantly -- temperature is basically
an average over a large number of (unknown) variables.


Tom Roberts

Message has been deleted
0 new messages