Top-level navigation for the Qubit Zoo through non-exclusive facets. An entry may appear in more than one map, so facet memberships can sum to more than the unique-entry count. These counts describe editorial organization, not platform importance or maturity.
Family MOCs
- classical-hardware-moc (2 entries)
- codes-moc (6 entries)
- color-center-moc (4 entries)
- cross-platform-moc (10 entries)
- floating-electron-moc (1 entries)
- molecular-moc (3 entries)
- neutral-atom-moc (4 entries)
- photonic-moc (9 entries)
- semiconducting-moc (12 entries)
- spin-photon-moc (5 entries)
- super-semi-moc (6 entries)
- superconducting-moc (22 entries)
- Topological (3 entries)
- trapped-ion-moc (5 entries)
Editorial policy
How to use this map
- Start with a family MOC when you want modality-local comparisons.
- Jump to Evergreen notes when you want mechanism-level synthesis that cuts across families.
- Treat Reference notes as evidence nodes, not the main conceptual layer.
High-value cross-family slice: defect-spin networking
- Start in color-center-moc for the benchmark room-temperature defect-spin story centered on nv-center-qubit.
- Move to spin-photon-moc when the real comparison is remote entanglement, photon indistinguishability, or telecom-native links via siv-color-center-qubit and t-center-qubit.
- Use divincenzo-criteria as the common scorecard, then coherence-time-hierarchy to track where each platform parks the cost: local coherence, cryogenic overhead, or photonic interface quality.
High-value cross-family slice: stationary-to-flying interfaces
- Start in superconducting-moc or semiconducting-moc while the information is still stored and controlled in a microwave-native stationary qubit. Local resonators and exchange links can move excitations around a processor, but they do not by themselves produce a low-loss fibre channel.
- Move through cross-platform-moc and quantum-transduction when the architecture must convert microwave quantum information into optical photons. Judge this route by end-to-end efficiency, added noise, bandwidth, and refrigerator heat load together; optimizing any one metric in isolation can hide the real systems cost.
- Take the native-emitter alternative through spin-photon-moc when the stationary qubit already has a useful optical transition. This can remove a frequency-conversion stage, but it moves the burden to photon collection, indistinguishability, spectral stability, and the emitter’s cryogenic envelope.
- End in photonic-moc once the question becomes which flying encoding or optical computation model carries the information onward. Use divincenzo-criteria to compare the two interface strategies without confusing a good local qubit with a complete networking stack.
High-value cross-family slice: semiconductor-superconductor boundary
- Use super-semi-moc when the same InAs/Al or Josephson-weak-link hardware is being used for tunable circuit physics, Andreev-spin control, or protected-circuit variants like ferbo-qubit.
- Switch to topological-moc when the decisive claim is nonlocal parity encoding or a topological superconducting channel, as in majorana-topological-qubit, planar-josephson-junction-qubit, and tetron-qubit.
- Keep josephson-junction-as-nonlinear-element and spin-orbit-coupling-for-qubit-control as the shared mechanism notes, then use threshold-theorem to judge whether the topological branch actually earns its overhead-reduction promise.
High-value cross-family slice: two-qubit interaction carriers
- Start in semiconducting-moc when neighboring spins entangle through electrically pulsed exchange. Read exchange-interaction-in-quantum-dots for how gate voltages manufacture , then heisenberg-exchange-in-quantum-dots for the logical unitary after the charge sector has been projected out. The scaling debt is local wiring, charge sensitivity, and routing beyond a nearest-neighbor exchange graph.
- Move to superconducting-moc when a microwave resonator or tunable coupler mediates the interaction. resonator-as-quantum-bus explains the shared-mode route; the architectural comparison should track parasitic , idle isolation, frequency crowding, and coupler calibration rather than treating every connected edge as equivalent.
- Use trapped-ion-moc when the carrier is a collective phonon mode. motional-mode-coupling-in-ion-traps explains why ions can interact across a chain and why that advantage eventually creates mode-crowding, heating, and closure constraints that motivate shuttling into smaller interaction zones.
- Use neutral-atom-moc when the interaction is borrowed transiently from Rydberg excitation rather than stored in a persistent bus. rydberg-blockade-mechanism ties gate range to blockade strength and geometry, while tweezer rearrangement changes which pairs are brought into that interaction neighborhood.
- Compare these branches by asking three questions in order: what carries the interaction, how completely can it be turned off or disentangled at gate end, and what resource must be routed or recalibrated as the processor grows? Raw two-qubit fidelity alone hides the architectural cost that distinguishes direct exchange, virtual photons, collective motion, and blockade.
High-value cross-family slice: storage manifold versus interaction manifold
- Start in neutral-atom-moc for the cleanest deliberate separation: information idles in hyperfine, clock, or nuclear-spin states, then temporarily borrows a Rydberg manifold for entanglement. coherence-time-hierarchy explains the storage advantage; rydberg-blockade-mechanism explains the decay, leakage, and laser-control debt introduced during gates.
- Move to semiconducting-moc when encoded spin information is protected from uniform magnetic noise but electrical exchange control still borrows charge admixture. Read decoherence-free-subspace for what the encoding removes and charge-noise-sweet-spot for the residual cost of making that protected information electrically steerable.
- Use trapped-ion-moc when internal electronic or hyperfine states store the qubit while collective motion is populated only as a gate bus. motional-mode-coupling-in-ion-traps makes the key success condition explicit: the phonon trajectory must close so the logical state disentangles from motion at gate end.
- Use superconducting-moc when the separation is engineered through dressed modes rather than distinct atomic manifolds. A data qubit can stay detuned while virtual cavity photons or a tunable coupler activate an interaction, but purcell-protection-via-detuning and resonator-as-quantum-bus show that the borrowed mode still leaves lifetime and residual-coupling costs.
- Compare these branches by asking: where does information wait, what auxiliary degree of freedom is occupied or admixed during a gate, and what evidence shows that degree of freedom is empty or disentangled afterward? This catches leakage and bus-reset failure modes that a two-qubit fidelity number can conceal.
High-value cross-family slice: hardware-assisted fault tolerance
- Start in superconducting-moc when the win is being claimed at the device layer: bosonic encodings, protected circuits, cat-bias engineering, or erasure-converting superconducting encodings all live there first.
- Move through codes-moc when the next question is which check structure, measurement schedule, or connectivity tradeoff best matches that hardware; this is where Bacon-Shor, Floquet, qLDPC, surface-code, and color-code choices are compared as code families.
- Move to cross-platform-moc when the comparison becomes hardware-wide overhead, decoder assumptions, feedback latency, or modular systems architecture via erasure-qubit, surface-code-logical-qubit, color-code-logical-qubit, classical-control, and quantum-transduction.
- Use bosonic-code-hierarchy, noise-bias-and-asymmetric-error-channels, erasure-error-vs-pauli-error, and threshold-theorem as the four-note bridge from hardware mechanism to architectural consequence.