Yup, the threshold theorem is the holy grail. Going back to the number of qubits: IBM and some other research institutes have been working on an idea of connec

Yup, the threshold theorem is the holy grail. Going back to the number of qubits: IBM and some other research institutes have been working on an idea of connecting multiple quantum computers in a grid. This would potentially help alleviate the problem of "how to fit 1000s of qubits, with all the wiring, into a single fridge", as well as "how to fabricate high-quality single chip reliably". In theory, we could fabricate and wire-up medium sized-chips, setup full quantum computers and connect them with some sort of quantum memory and quantum communication channels. Fabrication is tough; the company I work at does it all, from fabrication of quantum chips to final on-premise installation, and fabrication is as difficult as one can imagine.