Rejected by Best Fit School - Reapply, Transfer? Seeking Advice

[Arsenic&Lace]

Well it hasn't shown itself to be useful, and if my limited understanding of the Maldacena paper + a conversation with the resident string theorist are anything to go by, the theory is supported with post-hoc aestheticism, not robust physical insight. My also limited experience with theory in various fields but in particular biophysics has turned me to the opinion that theory is almost certainly science fiction until it is verified by experiment, and so as long as you're far away from experiments, you're just indulging in fantasy. There's no magic trick to even remotely know ahead of time that a theory is going to be verified, just myopic interpretations of historical anecdotes, and given the massive gulf in time between AdS/CFT and experiment, it's definitely headed towards the scrap heap.

 

[atyy]

Well it hasn't shown itself to be useful, and if my limited understanding of the Maldacena paper + a conversation with the resident string theorist are anything to go by, the theory is supported with post-hoc aestheticism, not robust physical insight. My also limited experience with theory in various fields but in particular biophysics has turned me to the opinion that theory is almost certainly science fiction until it is verified by experiment, and so as long as you're far away from experiments, you're just indulging in fantasy. There's no magic trick to even remotely know ahead of time that a theory is going to be verified, just myopic interpretations of historical anecdotes, and given the massive gulf in time between AdS/CFT and experiment, it's definitely headed towards the scrap heap.

String theory and AdS/CFT are backed by robust physical insight. They are backed by
(1) general relativity, quantum mechanics, the Wilsonian framework and the Weinberg-Witten theorem.
(2) thermodynamics, general relativity, quantum mechanics and the Bekenstein-Hawking formula

Also, biophysics is not the only culture of biology. There are those who remember classical Mendelian genetics, a theory that is wrong, but certainly not on the scrap heap.

 

[Arsenic&Lace]

Neither Weinberg-Witten nor Beckenstein-Hawking have anything to do whatsoever with experiment, so at least two of the supporting components are probably science fiction. But I don't really follow the field at the moment so if that's changed experimentally then I take that back.

The remaining points you make would need some kind of elaboration. Again my experience with string theory is that it usually seems to connect with other theoretical physics in a post-hoc way. You postulate something which is not grounded in any kind of physical reasoning, then after the fact viola, you miraculously find a spin 2 field incorporated naturally into your theoretical framework. Aesthetically pleasing and post-hoc "justified" but still just as likely to be science fiction as any other empirically unsupported idea.

 

[radium]

You were just accepted to grad school correct? If so, I really don't think you have the knowledge to be making such a strong claim about AdS/CFT. I am a beginning grad just starting to learn about the topic as I am interested in the application to entanglement. I have had conversations with some of the leading people in the field (who were the first to use it in condensed matter as well as Juan Maldacena himself) but will admit that I certainly don't have the knowledge to argue for it either way. I really don't think that anyone can currently say with absolutely certain that AdS/CFT is a valid approach. Even the most accomplished physicists in the world are divided on the issue.

AdS/CFT is a pretty new thing and people don't really know if it will be useful. It has however reproduced a lot of results from hydrodynamics for transport in strongly correlated systems. Recently people also have started to think it might be helpful in studying disordered systems.

Also, the claim that AdS/CFT is not physically motivated is wrong. I attended a talk by one of the people I previous mentioned and it actually gave a pretty interesting and physical explanation as to why people would think about looking at the AdS metric in the first place. In fact, a lot of the reason people thought about black holes is because they believe that there is a similarity between the dissipative behavior between black holes and strongly correlated systems. I find that quite interesting since my favorite thing in physics is being able to find connections across subfields.

 

[Arsenic&Lace]

Sure, nobody really knows, but there is a staggering ratio between the number of ideas which have failed and the number of ideas which succeed, to the point where it is hilarious and even a little sad that something without a pinch of empirical verification has consumed so much energy, since merely judging from that staggering ratio it is almost certainly entirely or at the very least mostly false.

I'm no expert on AdS/CFT but I'd be delighted to have a crack at some papers which furnish its physical foundations, but even strong physical foundations are a far distant third to good phenomenology.

I'd be interested in some of the papers where it reproduces results for strongly correlated systems, although even reproducing old results is a far distant second to agreement with experiment.

At this point it may be better to simply create a new thread to examine it.

 

[radium]

I mean I think it's pretty overconfident to much such a blanket statement about a field that you admit you do not know much about. There are some of the most accomplished physicists in the world (at schools like Harvard, MIT, and Stanford for example) who are currently working in holography. Even many people in CMT who may be more skeptical about it have told me its a valid thing but it may be hard to find a realistic system it can be used for. They say the field is in it's infancy, so a lot of then are just watching on the sidelines to see where it goes even if they are not actively participating now.

 

[Arsenic&Lace]

I could care less how accomplished they are or where they work; if even the most basic tenets of empirical science, i.e. that there is nothing that can safely be said about a theory until it is at least partially verified empirically, elude them, how good at theoretical physics can they truly be? Some individuals hail this stuff as quite plausible when they have essentially no idea, which is what I find dubious. Even more worryingly there's a huge gulf between phenomenology and theory where there appears to be less interest in phenomenology; even a disregard for it's importance (a friend of mine had a long conversation with Kitaev, as an example, where Kitaev more or less bluntly implied that he cared not even slightly for experiments). The cold truth is that the human mind is woefully limited and is capable of hardly anything without observational insight.

Sometimes we are lucky and the next portion of the world of ideas we are exploring looks like the previous portion in some way; the jump from QED to QCD is sort of like this, where one gets a great result by building something which is made in the image of the last great result. But then that landscape fundamentally changes and it's all guess work, and nobody knows anything at all about what could work; speculating that many miles ahead of where you're actually at is effectively pointless unless you're very lucky and the next part is quite familiar to the current part.

I suppose AdS/CFT has a better shot than some pet theory about unicorns and ghosts, but not by much. It's just a feature of nature, where the history of science does not repeat itself save only in very general details.

 

[berkeman]

Thread closed temporarily for Moderation...