Past experiences.
You see a news article of some lab with a new technology. But, past experience says that taking a simple lab test proof of concept and making it a product on the shelf actually takes 8-10 years. This is because scaling up production always has problems. Getting manufacturing always has problems. Mass production never works as well as a single experiment performed with care. There are whole specialities in scaling things up.
I’d interpret that statement as: we know it’s not impossible, and we know roughly how to do it, but we can’t just build it yet…there are lots of problems to solve first.
A good example would be sending an astronaut to Mars and returning them safely home. I don’t think there’s any question it’s possible. However, NASA thinks it doesn’t make sense to jump directly to a manned Mars mission right now, but to first take some incremental steps and learn more along the way. One of those next steps is a sample return mission – there have been dozens of spacecraft sent to Mars and none have ever returned, so that’s a great next milestone.
Another example would be improvements to computer processor speeds. While we have hit certain physical limitations (the size of each transistor can’t physically get much smaller), nevertheless computer processors continue to get faster with every new generation. It is reasonable to assume that trend will continue. The reason we can’t just make a faster processor today is because every increase in speed isn’t just one improvement, it’s the cumulative effect of hundreds of small innovations, many of them taking years of research and development. So we know that there are hundreds of teams working on ideas to improve the next generation of processors. They haven’t run out of ideas. We don’t know exactly which ideas will work and which will be ready next year, but we can reasonably predict that enough of them will work that next year’s processor will be faster. Will a processor be 5x as fast? Yes, most likely, but that’s years away.
There are a lot of good answers here but I feel like one big reason has been omitted which is cost and scale. There are a lot of technologies that have proof of concept however the cost is too great or the technology simply cannot scale at this time due to the need to implement some sort of infrastructure.
Latest Answers