The challenge in quantum computing lies in scaling; while achieving high fidelity with a small number of qubits is possible, creating a robust system that consistently performs well with a larger number of qubits is a significant engineering problem.

We plan to build a modular quantum computer made of small quantum modules that, when linked, can work together to form a unified quantum state.

The initial speed of our modular quantum computer will be limited by the connection rate between modules, but for small problems, the computation speed won't be a major concern.

Network approaches to quantum computing enable any qubit to link with any other qubit, enhancing connectivity and power, despite the increased risk of errors if not managed correctly.

To determine if a quantum computer can be modular, we need to assess whether we can achieve good quality entanglements between two modules, which would prove they are part of the same quantum state.

The timeline for building a modular quantum computer depends on interest and investment, but demonstrating two fully linked modules is an immediate goal for the next year.

Quantum computing offers a faster and cheaper path to achieving high levels of computing power compared to traditional methods.

Microsoft is pursuing topological quantum computing, aiming to create qubits that can be controlled much better than current technologies, although it's still in the early stages of development. Scaling quantum technologies is challenging due to control engineering and other factors.

Quantum computing is fundamentally different from classical computing, and if you can determine which computational path was taken, you lose the speed advantage that quantum computing offers. It's essential to sum all possible computations without knowing which path was followed to leverage quantum mechanics effectively.