@Nicholas.Lee I told Dr. Eleftheriades about your objection to a visible light version of his cloak--that cancelling visible light would just create a black target, rather than transparency as most people interpret "invisibility," and this is what he replied:
No I do not agree with this: What is cancelled out is not the total field surrounding the object to hide. Only the scattered field is cancelled. Hence you will not see anything black around the object. You will just see what light was there when the object was absent.
What is true though is that for light, the cloak cannot respond instantly. Short but finite time is needed (size_of_object/speed of light) for the information to flow behind the object. Hence momentarily the object will show up before it disappears. However, this is true for any electromagnetic cloaks, including the metamaterial based ones (where the situation is worse since they are thicker and light will need even more time to travel around it).
You're actually correct in everything you said, and I was incomplete in what I wrote as the message was getting too long.
For an "invisibility cloak" you actually have to do both things, what you mentioned and what I mentioned. I took the leap of how to apply this concept to make an invisibility cloak, which is to detect what's coming in on one side, send a cancellation pulse backwards, AND send a propagating pulse forward as though your object didn't exist.
So, as you said, what's described in the article has all the issues you correctly identified. However, a logical extension of the same concept (calculate and project what is behind the object) gets around the whole "black on black" issue for the visible range.
Hope this clears things up a bit an sorry that my reply was a bit incomplete. I would consider this extension a part of the "concept" of utilizing detector/emitter combo to make something invisible, but definitely see that what's described in the article is not at this level of complexity.
For the backwards cancellation, the phase matching is absolutely necessary, but for the forward propagation, it's not unless you're trying to fool an actual sensor which can detect phases instead of the human eye, so once again, you are in the strictest sense correct.
The idea propsed in the article is phase cancellation, (not a "camera on one side, +display on the other" type of invisibility.), so that is the only concept I am commenting on. I grant you that other solutions may indeed be viable.
I was specifically criticising the impossibility of the solution they proposed when it was glibly extended to visible wavelengths and a non-emitting observer.
To say "Phase matching is not an issue for practical applications as the eye is not a single event detector and is a continuous amplitude integrating sensor.", is not a correct statement in this case, as to phase cancel out light you need phase coherent cancellation and the non-phase sensitivity of the eye is irrevlevant to the physics of this.
Their idea is to have "destructive interference" of the signal at the receiver, and this requires phase coherency. This approach is easy for mono-static radar, and impractical for light.
I agree with this statement for what can be done with the level of sophistication currently in this paper and that the whole "cancellation/distortion" of the source pulse is about as much as what is shown here (which has a lot of very interesting applications, so it's not to be trivialized).
In principal though, this "concept" can work. For instance, a simple analogy would be that if you had a thin opaque wall that you wanted to cloak. It is possible to put detector/emitter combo on the surface of the wall and directly emit whatever is detected on the other side and it will even be practically invisible on the either side (I think there was a demo of this where someone was wearing a set of camera's on his back and had a white screen that was in front of him that had a projected image of what was behind him on there... some university in Tokyo?). Phase matching is not an issue for practical applications as the eye is not a single event detector and is a continuous amplitude integrating sensor. If you were in a room lighted by a scanning coherent RGB lasers, or watching some display that utilized phase matching to create the image, that's a different matter, but in normal diffuse light situations, you don't need to match the phase, but would probably need it to calculate what would need to be projected on the other side.
However, in practice, I think this is beyond the processing capability to do on anything beyond a very thin sheet and for looking at things from relatively far away. The number of antenna's and the processing that would be necessary to do to translate input from one antenna to the output of the others real time in a more complex 3D object is huge and impractical as anything beyond a thin sheet would not be a direct translation and for visible spectrum, you would need emitter/detectors at the resolution limit of your closest target observer (which could be a spacing of every 500um of less if you wanted to cloak at something at arms length (calculate how many of these you would need to cover even a computer mouse...).
With that said, for psuedo stationary objects that needed portable cloaking from detection far away where a moderate time lag between what is projected and what is detected on the other side doesn't matter, I think it is possible. Think of setting up a military field station tent and wanting to reduce it's detection from binoculars... a "dome" like this would be ideal to make it very hard to detect from 500 yards away.
I am just imagining a scenario when both the attacking planes and the target warships are cloaked. This could be a virtual war where no war-photographers will be able to show it live on their TV channels!
Instead of using metamaterials, Eleftheriades's approach surrounds the object to be cloaked with tiny antennas tuned to the frequency band in which the cloaking is to occur, such as radar.
Now, that's interesting. Colin, you've been writing about this "invisibility cloak" topic for many years. It's such a tantalizing idea. We are all hoping that this will come to life within our life time!
Having been a fan of both Star Trek and Harry Potter, I leanred from Star Trek how to analyze the events for what made sense and what didn't. There are plenty of instances on both that don't make sense or are never explained. One place where the HP story breaks down is right at the very end when Ron suddenly speaks Parsel Tongue. It comes out of nowhere and IMO is a cheap way out.
You can apply the same analysis to the Wizard of Oz books as well as the Wicked series.
I really need to read/see The Hunger Games Series.