Two experiments, opposite assignments
Wineland’s trapped-ion work uses light to manipulate matter; Haroche’s cavity experiments use atoms to probe stored light. The 2012 prize joined these methods of controlling individual quantum systems. It did not identify one universal detector that leaves every property untouched.
An experimental claim must name what remains available after interrogation: the particle, a chosen observable, or a particular coherence. Keeping an ion in a trap and preserving an arbitrary superposition are different tests.
Read the evidence: NIST (2012), David J. Wineland wins Nobel Prize in PhysicsNobel Prize in Physics 2012 — presentation speech
Bright and dark are detector records
An ion can scatter many photons on a cycling transition while another internal state remains comparatively dark. A photon-count record can then discriminate internal states. Missing counts also have mundane causes: finite collection efficiency, background, detector noise and an insufficient observation interval.
A telegraph-like trace becomes evidence for quantum jumps only through a model of the levels, driving, detection and dwell times. Readout changes the state being interrogated; repeating fluorescence does not preserve an arbitrary initial superposition.
Read the evidence: Leibfried et al. (2003), Quantum dynamics of single trapped ions, Rev. Mod. Phys. 75, 281
Count photons without absorbing each one
In the 2007 cavity experiment, off-resonant atoms probe microwave light through a phase shift. Subsequent atomic measurements yield information about the stored field. Repeated records reveal transitions in photon number while the light remains available between probes.
Quantum non-demolition refers to the measured observable under suitable dynamics, not to the absence of backaction. Learning photon number constrains the field state; loss and thermal exchange can still change it. A sudden inferred jump must be distinguished from uncertainty in a short sequence of imperfect probe outcomes.
Read the evidence: Gleyzes et al. (2007), Quantum jumps of light recording the birth and death of a photon in a cavityGuerlin et al. (2007), Progressive field-state collapse and quantum non-demolition photon counting
What the teaching models do and do not establish
Compare a fluorescence count trace with the ion sideband spectrum, then inspect a decoherence channel. These are three different data questions: internal-state readout, quantized motion and loss of coherence. One cannot replace the others by changing a label.
The linked Labs are teaching models, not a recreation of Haroche’s microwave cavity apparatus or a digitization of its atomic detection sequence. Their outputs illustrate specified dynamics. Historical measurement fidelity, cavity lifetime and state reconstruction require the experiment’s own calibration and records.
Read the evidence: Gleyzes et al. (2007), Quantum jumps of light recording the birth and death of a photon in a cavityNIST (2012), David J. Wineland wins Nobel Prize in Physics