Reimagining of Schrödinger’s cat breaks quantum mechanics — and stumps physicists

They were both wrong, Superpositions are allowed in QM. But rather than alive 'and' dead, the cat is alive 'or' dead, because QM determines the probabilities (and some other things) when the wave functions are added.

So in a superposition, the particle doesn't become anything specifically so nothing happens and the cat is still alive.  Excuse my linear logic but I still can't see what everybody else seems to see! 

But in order for an illustration to be useful, it has to make sense as if it were a real event.  If it were real, the position that doesn't break the vial doesn't prevent the vial from being broken so the cat dies.  Also, I believe you are wrong about Schrödinger's motive.  My understanding is that he was trying to show how untenable the Copenhagen Interpretation was, something that Einstein later congratulated him for, after all, a cat cannot be both dead and alive!

Correct. Absurd isn't it. Here's a bit more detail http://wtamu.edu/~cbaird/sq/mobile/2013/...ent-prove/

There is somebody who has problems with QM...

He seems sound enough on the question of How does quantum theory allow a rock to turn suddenly into a duck?

The authors don't seem to accept this for the assumptions they make in their experiment, they see the 2 systems as somehow isolated, and that the systems are not affected by each of the 4 measurements made by each of the 4 observers. But in Copenhagen, the systems initial states *are* changed for each of the other observers, after one of them makes an observation/measurement.

This blogger, Lubos Motl, seems to be making essentially the same criticism - that when the first of the two observers outside the boxes makes his observation, it affects the systems inside the boxes, and the authors are not allowing for that. He presents it as a crass, obvious error:
https://motls.blogspot.com/2018/09/frauc...stent.html

Motl seems to be something of an outcast in the physics community, and to call his blogging style abrasive would be a severe understatement, but I'm not sure his intelligence is in question. But it does seem odd, if the error is so obvious, that the paper has ended up being published after two years' debate about the preprint.

There is somebody who has problems with QM...

What problem specifically?

What is it you think you see?

A cat that is either dead or alive depending on whether a superposition is literal or potential.