spooky nonlocal communication, or bad abstract? I'm referring to this recent paper, "Experimental Proof of Nonlocal Wavefunction Collapse for a Single Particle Using Homodyne Measurements" by Fuwa et al. published in Nature Communications. Non-paywalled arXiv version here. From the abstract:

...the choice of measurement in one lab really causes a change in the local quantum state in the other lab

My understanding of entanglement (whether it is two particles, or one as in this article) is that Bob's measurement can be affected by the (uncontrollable) result of Alice's measurement, but not her choice of measurement. (the latter would seem to enable a form of non-local communication). So is it just a badly-worded abstract? (the full article uses similar language)
 A: From my initial reading of the paper it looks like the the quantum state Bob has access to is changed by Alice's choice of measurement but in such a way that he can't tell until Alice communicates her choice of measurement to him.
In particular they present a way for Alice to prove to Bob that she has influenced the part of the state he has access to even if he doesn't trust her (so even if Alice is malicious Bob can be confident that they shared an entangled state initially).
The phase that Bob measures looks to be uniformly distributed if he doesn't condition on anything. However if he conditions his measurements on the numbers Alice gives him (basically he splits the data up into groups depending on what settings Alice says she chose) structure emerges as shown in figure 2.
This is quite a general feature of entanglement. Alice can "steer" the quantum state Bob has access to in what appears to be a violation of special relativity but only in such a way that Bob can't tell unless Alice sends him supplementary information over a classical channel (so they can't break causality).
I agree that there is a huge amount of superstition / misinformation around about quantum effects though (especially entanglement effects).  
