[Halc]

Clearly, the video shows different x values so it is wrong.
That guy made a mistake!

You're right. From C's perspective, no matter where the origin of that frame is placed, C should be at the same X coordinate at both the II and III events since he's present at both of these, but he's not.
There are errors in both videos in the OP.

A question.
How do we call 'jumping' from one inertial frame to another?
Changing an inertial reference frame?

I think the latter phrase says it well.
It correctly doesn't imply that one is exiting some frame, or that one is necessarily accelerating. A frame change simply needs a Lorentz transform.

Here is a demonstration of what went wrong in the OP video.

The discussion of the train/platform setup seems to have little to do with the video. I didn't watch it all, so maybe it comes up elsewhere.

It is impossible to align the following 4 events due to disagreement on the simultaneity between the inertial frames.
Imaging a 'space train car' and a 'space platform'.

It would help if you said what speed one was going relative to the other. The pictures don't say, but I see proper lengths of 3.46 platform and twice that for the train car, so I'm guessing train speed of .+.866c to the right.

The events F', B', P, B cannot be aligned though.

'Aligned' isn't really a physics term. You mean they are not the same events? At time zero, apparently F' and P are the same event, and B' and B are not.

Then the invariance of the space-time intervals of a light round-trip defines stationary preferred frame with the slowest time.

This comment would be false if it made any sense, but it's not even wrong. The spacetime interval of light is always zero. That fact does not in any way define a preferred reference frame.

[Eternal Student]

Hi.

   I would concur, this bit looks an error...

From C's perspective, ...., C should be at the same X coordinate at both the II and III events...

    Assuming Don (the presenter) establishes the frame C so that event I is the origin, then he should have listed the x co-ordinate for event II   as     2γL   and  not   γL.    That does seem to be at least a typing error.
    Fortunately, he didn't need the x co-ordinates for the subsequent calculations.

Crumbs,  Halc,  how many times have you watched this video to spot that error?

There are errors in both videos in the OP.

      Both videos were basically Pop Sci videos or at best what is sometimes called "edu-tainment".   I think both presenters up-sold their videos and made fairly grand claims about how their video fixed some mis-understanding.   Neither of them stayed around or did some after-market research to investigate any new mis-understanding that may have arisen in people after they watched the videos - and there are some:  See the contents of this thread and the various people who have been involved.
     I have more of an axe to grind for Don's video.  You can legitimately argue that he hasn't got any one person to make the round trip and was therefore never really answering the main issue.   There was only an exchange of information at event II between two people when they flew past each other. 
    Some of the people in this thread (including some that have now been split into a second related thread) have interpreted the video as if it resolves the twin paradox without assuming any acceleration of a twin - but it doesn't.   
    It's apparent the calculation you can perform to determine the total elapsed times would be independant of the acceleration at event II.   That is not in dispute, there are numerous videos and textbooks that will show that result.  Don's method is an accpetable one you could use to calculate the total elapsed times for various people.   However, Don's method does not magically prevent a twin from needing an acceleration to change their motion at event II if, in fact, one person really was making the round trip.
    A major mis-conception that has arisen in some people after Don's video is that the acceleration (the change in motion) that occurred at the turn-around point isn't important at all.  This is a serious problem when people start asking questions like exactly where, when or as a result of which things, some discrepancy in allocation of times start to appear between the twins.   If you start trying to consider how the different twins would allocate co-ordinates to events, there is an abrupt shift in the allocation of co-ordinates for one of the twins at the turn around point.   It's not that the time dilation was only happening during the acceleration (that would also be a mis-conception) - but just that a change in the alllocation of co-ordinates for events at the turn-around is a contributing issue which cannot be ignored.

Best Wishes.

[Halc]

Assuming Don (the presenter) establishes the frame C so that event I is the origin, then he should have listed the x co-ordinate for event II   as     2γL   and  not   γL.    That does seem to be at least a typing error.

Yes,  x = 2γL for both events II and III if the origin of the C frame is event I.

Crumbs,  Halc,  how many times have you watched this video to spot that error?

That was actually the first time for me, and then only from minutes 8 to 11. I don't watch most videos just because somebody posts them. I would have thought that a video endorsed by Fermilab would be at least peer reviewed, but apparently neither video in the OP was.

As for how I spotted that, Jano gave the exact timestamp (10:36) with the errors. I had to backtrack a bit from there to get the context, where I spotted at least one more.
8:23 shows a picture identifying the entire system as a space-like state being identified as events I through III which is just wrong. They treated as events (with an x coordinate) later on, but that picture is beyond wrong. The events should have an arrow pointing to the actual events, which is the left of the top line, but Don says that event I is "when they all start" instead of the event where A and B part company. C is not present at event I, but the picture tells a different story.

The title of the video identified it as crank from the start, since it claims to be "the real explanation" like the other valid ones are wrong. That's a big reason why I didn't bother to watch it. Somebody who knows their physics wouldn't insist on just one way being the only correct way. Of course that might be just self-promotional hype, but then at least get it right.
His story also doesn't involve twins, which arguably disqualifies it as an explanation of the twins paradox. He's trying to show how acceleration isn't necessarily involved, but as Alan points out, if twins never accelerate, then they can never be at different locations. You noticed this as well.

Both videos were basically Pop Sci videos or at best what is sometimes called "edu-tainment"

Yes, but a video put out by something like Fermilab or somebody with a reputation to protect (Hossenfelder, college physics professors) shouldn't be putting out stuff that is blatantly wrong, and so many of them do.
I kind of expect it from all the uncredentialed yahoos making you-tubes from their mom's basement, but sometimes those guys actually hit the mark and put out something quality.

It's apparent the calculation you can perform to determine the total elapsed times would be independant of the acceleration at event II.

It shows that you can do the computation without knowing the acceleration at any time, anywhere. You just have to know the speed of everybody, or if you're going to complicate your process with coordinate changes, you need to know the velocity of everybody at all times. If the velocity is known, then the accelerations can be deduced, but since the result can be computed from just speeds, it isn't a function of accelerations.

My simple method (which does not worry about acceleration) was posted in the first reply.
My most general explanation (which works in all cases, not just the twins scenario) involved summing up intervals, and also wasn't a function of accelerations. It was posted in one of the other threads that I can no longer find.

[paul cotter]

Emphasising yet again the hazards of pop science education by youtube.

[alancalverd]

The problem with any derivation that ignores acceleration is that it must presume prior synchronisation, and thus becomes a circular argument.

Identical clocks are seen as ticking at different rates if they are moving relative to one another. So their initial relative velocity must have been zero in order to establish synchrony, demonstrate identity, be born as twins, or whatever initial condition you choose to make the subsequent observation paradoxical.  So at least one of them must have accelerated to produce a relative velocity.

[Eternal Student]

Hi.

   We had blackboards,  they were awesome.   A wall of blackboards and you could slide one up and another down.   It meant the lecturer could write half-a-dozen formulae and draw a diagram and then slide them up while continuing on another board.   Awesome.
    So when the lecturer continued for another half-a-dozen lines and said   "we can see from the definition of Z that....",   you actually could gaze upwards and see the definition of Z.   You don't realise how good something is until it's gone.

That's all I wanted to say, thank you for your time and have a good day.

Best Wishes.

    (There is some relevance:    By comparison,  You Tube videos don't have anything slide up the wall.   Formulae and diagrams are gone and you can never fiddle with the time slider adequately to get them back.   If they say "You can see from the definition of Z",  I'm just going to go with it.   I also don't recall people coming into the lecture hall and advertising stuff but I suppose that's a separate issue and a moan for another day).

[hamdani yusuf]

Emphasising yet again the hazards of pop science education by youtube.

It's not specifically YouTube's fault. Any other media can be used to produce misleading or incomplete information. That's including textbooks, websites, podcast, tiktok, and live lectures.

[alancalverd]

Textbooks get read by editors and reviewers.

Live lectures are subject to all sorts of heckling and barracking (unless you are talking absolute bollocks from a pulpit, in which case the audience will swallow anything with utter reverence).

Peer review isn't perfect, but it's better than the nothing you get from online media.

[paul cotter]

Oh dear, here we go, again. Word salad.

[Halc]

Oh dear, here we go, again. Word salad.

Indeed.  The prior posts have been removed for several reasons.

1) Nonsense in the main sections of the forum
2) Contained personal information
3) Had nothing to do with the topic title, hence is a hijack of a topic
4) Contained a bibliography which was nowhere referenced
5) Contained no content from any of said bibliography
6) Word salad as Paul points out
7) Duplicate posts

At least the first three of those violate forum rules.

[paul cotter]

Thanks, Halc. I suggest you may delete my post, #48, as it serves no purpose now as the offending material has been removed.

[varsigma]

Einstein's special theory is really like the law of thermodynamics: if your theory denies the reality of SR, there is no hope for it.

The twin paradox is a problem--a physics problem. But apply a bit of logic and it's obvious that the acceleration/continuous coordinate change is over a distance which is much shorter than the distances covered at constant velocity.

Hence the distance for acceleration(s) is negligible, hence ignoring that part of the journey has a negligible effect. Logic.

[alancalverd]

Hence the distance for acceleration(s) is negligible, hence ignoring that part of the journey has a negligible effect. Logic.

Brilliant!

The time spent accelerating a bullet in a gun barrel is negligible compared with the time it spends in flight at constant velocity, so it doesn't matter whether you fire the gun or just send the bullet by post, the lethal effect is for all practical purposes identical.

[hamdani yusuf]

The twin paradox is a problem--a physics problem. But apply a bit of logic and it's obvious that the acceleration/continuous coordinate change is over a distance which is much shorter than the distances covered at constant velocity.

Some  people said that acceleration is relative, thus it's undefined without a frame of reference. Some others said that it's absolute, thus acceleration of an object can be determined without a frame of reference. This difference showed up in many YouTube videos trying to solve the twin paradox. What do you think is the correct one?

[paul cotter]

Acceleration is absolute. The "paradox" arises when overcomplicating the analysis. All one needs is the Lorentz factor γ.

[alancalverd]

Some  people said that acceleration is relative

An accelerometer consists of a mass attached to one end of some kind of spring, the other end of which is attached to the object being accelerated. An observer within any accelerating object will see a deflection on his accelerometer. Acceleration is therefore absolute.

[varsigma]

Brilliant!

Aw shucks. Not really.

The time spent accelerating a bullet in a gun barrel is negligible compared with the time it spends in flight at constant velocity, so it doesn't matter whether you fire the gun or just send the bullet by post, the lethal effect is for all practical purposes identical.

To be physically pedantic, that ignores the atmosphere and the earth's gravity; and that it's rotating. But as they say, your solution is accurate over small enough distances. "lethal effect" isn't really physics, though. There will be a transfer of momentum, I think.

And so we see that firing the gun will result in a larger transfer of momentum to the bullet, over small enough distances, than posting it in a letter. You could ask the postie to throw it, perhaps.

[alancalverd]

"lethal effect" isn't really physics, though.

It is mostly physics, with a bit of anatomy, followed by physiology.

[alancalverd]

The twin paradox is a problem--a physics problem.

It isn't a paradox at all. It is a measurable effect, fully explained by relativity, which is of little significance in most people's lives. Unless they use GPS navigation, of course. 

As for "problem", it's simply a question of solving a slightly more complicated equation involving v/c if v is not zero.

[alancalverd]

Now let's consider the "fly-by" question.

A and B set their clocks to zero as they whizz past each other - i.e. as their position vectors diverge. Some time later, they pass again and compare their clocks.

For simplicity, let A remain at rest or travelling in a straight line at constant velocity. He knows this because his accelerometer reads zero throughout the experiment.

How, then, can B fly past him again? Not if he is travelling at a constant velocity because their position vectors would continue to diverge. So he must....accelerate.