Quantum computer
Any computer operates on two principle components, the information and what you can do with that information. In classical computing we have the bit and logical operators to operate on one or two of these bits. The bit is either a 1 or a 0. Adding two of those bits (logical AND operator) gives you a new bit. In quantum computing the computing part is exactly the same (operations on information) but the components are fundamentally different. A quantum computer has qubits (or qubit registers) that are operated on according to the laws of quantum mechanics.
Computing complexity
One of the fundamental differences between a bit (or a bit register, for example a byte) and a qubit is that a bit has one state and a qubit has multiple states at the same time. This means you can encode information that is much more dense and/or uncertain. And you can compute with these complex numbers (vectors to be precise) with the speed (relatively) of a classical computation. One way of describing this is that quantum computation is massively parallel. I prefer to talk about complexity over parallelism. But all this comes at a cost.
How to deal with uncertainty?!
A qubit register (a byte is a bit register) of 32 qubits can have 232 states at the same time. When calculating this superposition of states has meaning. But when you want to ‘know’ what the value of the register is a measurement has to take place. This measurement is probabilistic. And you have measure the same computation enough times to be sure enough of the result. This makes unpredictability a fundamental principle of computation in a quantum system. In the research you see we are trying to ‘fix’ this, because we don’t know how to use this property.
New characteristics, new applications
The consequence of this property is that cloning of information is impossible. But quantum mechanics offers the compensation of reversibility of computation. There are no known novel applications of these characteristics, other than optimizations of our existing applications (like cryptography.) I guess that the impossibility of cloning will lead us to regard the concept of identity differently (in the broad sense.) And the reversibility of computation will have a huge impact on the concept of time. We implement ‘undo’ by storing the states of the past in memory. A quantum computer does not need memory and consequently is not limited by memory. (For example, reversibility means, in theory, no backups anymore.)
Two waves of innovation
The first wave of innovation relating to this research field is, what I would call, Quantum Enhanced Classic Computing. Classic computing can continue to improve with ‘the speed of Moore’ for decades to come. The other is quantum computing itself. There are already quantum computing applications, for example quantum cryptography, but special purpose applications.
Quantum compared to classical
If Wal-Mart (the most recent phase of industrialization) is not possible with classical computing what would be the example for quantum computing? The problem that needed the computer as a solution was the encryption of the german military in WOII. What is the problem that needs quantum compution? The climate and how to proceed dealing with it? Climate modeling is too complex to properly compute with classical computers. It is also unpredictable in its very nature, not only because we don’t understand it well enough. Perhaps if we need to manage huge migrations of people due to rising (or falling) water levels we need a good prediction on where it is safe?!
Hype or hope
Quantum computing is not hype. It is just one of the fields where governments and research institutes around world spend money on. And for quantum computing the budgets are growing. But because we percieve the ‘end of Moore’s Law’ is near we are looking at alternatives. But this research is still fundamental and not applicable. Based on the charters of european and US research budgets I think it will take at least another 10 years for commercial quantum computers to emerge. But in 5 years time we might expect classical computing to be quantumly enhanced. These two phases will go hand in hand for 5 to 10 years. And perhaps in 15 to 20 years we might have a quantum desktop. But perhaps this is too much conceived from the current context of the application of classical computing. Definatelly not hype and certainly hope. But the timeframes in this field are not ‘compatible’ with expectations of instant progress.
