Hard-Nosed Predictions

It seems appropriate to follow up Mike's post about prediction with a post on actual predictions.

These predictions are not the work of futurists. They are predictions made by 700 IEEE Fellows -- about half of them academic researchers, the rest working in industry -- asked to forecast trends within their area of expertise over the next 50 years.

They are not overly optimistic. For example, while 67% said that fuel cell would likely be widely used in mobile devices, only 19% said that they would likely be used as a source of household electricity globally, and only 29% said it was likely that photocells over 50% efficiency would be in commercial production.

So, let's see what these hard-nosed, bubble-bursting experts had to say about more advanced technologies:

• Almost half said that it was likely that implantable brain-machine interfaces would be widely adopted.

• Over half said that most individuals in developed countries would have documented personal genetic profiles -- and within only 20 years.

• 64% said that a universal language translator would be commercially available, and almost 70% thought that this would happen in 20 years or less.

• Nanoelectromechanical systems will likely go commercial (58%), probably in less than 20 years.

And of special interest to CRN:

Only 26% thought that molecular self-assembly would likely be commonly used to build integrated circuits; nevertheless, 50% thought that five-nanometer processors would likely become commercially viable. So, one wonders, how will they be built?

Well, 56% of experts thought it was likely that it will be "commercially viable to manufacture nanostructured materials to exact specifications without machining." And of those, over 75% thought that this would happen within 20 years or less.

Meanwhile, almost 2/3 of experts expected "robust design tools for fabrication at the nanoscale" to become available.

They weren't asked directly about molecular manufacturing, but enabling technologies are certainly looking plausible. If you can do NEMS, five-nanometer commercial lithography, robust design, and built-to-order nanostructured materials, then it's not a very big step from there to NEMS-building-NEMS.

The paradigm is shifting. The nanoscale is rapidly moving from the domain of scientists to the domain of engineers -- and the engineers know it, and are looking forward to it.

Molecular manufacturing is going to be more about engineering than about science. Even if it's developed first along lines that are different from the ones we are analyzing, equivalent capabilities are certainly coming -- and, as we've been saying all along, they are coming very quickly.

Chris Phoenix

(Hat tip to Foresight/Nanodot for the story.)

Tags: nanotechnology nanotech nano science technology ethics weblog blog

Comments

[Tom Craver]

One possibility for making 5nm feature semiconductors is direct contact printing.

[NanoEnthusiast]

Sub-wavelength photolithography maybe another candidate. Is direct contact printing related to dip pen nanolithography?

[Chris Phoenix, CRN]

No, direct-contact printing is essentially a high-tech rubber stamp. DPN is more like writing with a quill pen.

Chris