Technology: Fundament Of Our Civilization's Wealth
Is democracy the reason for our current prosperity? I do not think so.
I think it’s the ever more sophisticated technology
available on the market. And more specific microelectronics.
I furthermore deem capitalism as the backbone of industrialization
and computerization. In this article I want to focus on a ca. 70 year civilization
time scope not on the 1000 year time scope. Thus this article focuses on the price of computing.
The applications of microelectronics have lead to many products that we enjoy nowadays: flat-screens, smartphones, ATMs and the Internet. Self-Driving
cars are also hitting mainstream adoption fairly soon in ca. 1-3 years.
Each decade had its share of new electronic tools that drove
productivity, increased consumer choice or life quality. The 60s had communication satellites,
the 70s had for example ATMs, the 80s home gaming consoles, the 90s the internet and the 00s the smartphone to name just a few.
Based on my interpretation of the predictions of Ray Kurzweil we might have to wait some time until another innovative tool, powered
by microelectronic advancements comes to market.
Kurzweil analyzed and predicted the advancement of the last
35 years shockingly accurate, beginning with his first book “The Age of Intelligent Machines” from 1990.
Would Kurzweil or Intel tell you in their forecasts that
the years from ca. 2017 onward will be a decade-long desert of microelectronic innovation?
Would either Kurzweil or Intel receive any benefit from spelling out this innovation
drought that looms?
Kurzweil would paint himself as a technology
doomsayer, which would clash with his image as entrepreneur and inventor at
Google in the field of artificial intelligence. Intel would basically have to
say that one of their trademarks, Moore’s Law, is outdated and that no big innovations should be expected from them in the next 10 years.
If I apply the basic assumptions of Kurzweil, that technological
innovations, also called paradigms, are improving the price performance of
computation in an s-curve, then the current paradigm (integrated circuits) is
currently nearing its end, while the next paradigm (probably quantum computing) is not yet ready to be used for productivity advancements.
Our current paradigm was marked by substantial increases of
price performance of computing by a factor of ca. 100.000 between ca. 1973 and 2015.
End Of Moore's Law
The size of the tiniest transistor is approaching the size
of a molecule (Wired: “World's tiniest transistor is the size of a singlemolecule”).
Shrinking the size down any further is almost impossible say experts since
quantum effects and other limiting natural laws take over to form an insurmountable
wall of increasing costs and fuzzy computation.
That might be the case but anybody who compares the performance
of a PC you could buy for $1000 in 2012 with 2016's PCs, recognizes that the price
performance has not increased by a lot.
That is a unscientific remark by me, but sometimes common sense can tells us things that the Intel investor relations brochure does not: New chips nowadays improve foremost in terms of energy consumption and rarely in terms of price of computation.
That is a unscientific remark by me, but sometimes common sense can tells us things that the Intel investor relations brochure does not: New chips nowadays improve foremost in terms of energy consumption and rarely in terms of price of computation.
The next s-curve paradigm will again bring our civilization to a new level
of productivity, prosperity and life quality. Kurzweil makes convincing
arguments that this will happen. My estimate is that quantum
computer technology will be the next paradigm, but it is not known when this will
happen. As a
critical and enthusiastic reader of Kurzweil, Wired and Intel publications I estimate that we might have 10
years to wait.
Conclusion
If my assessment of economy and technology are correct, these are the logical conclusions:
The major innovation engine (computing price performance) is
taking a long break, though the minor innovation and productivity engines will continue to churn globally:
- Software improvement
- Software has time to mature since the underlying hardware changes slower
- Company restructuring
- CEO's will be under pressure to find news ways to optimize their company
- Biotechnology
- Genetics, Medicine, Crop Science
- Privatization of state monopolies
- More wellness services to keep you fit for productive work
- More & improved education focused on job skills
- Modernizing old infrastructure
- Space exploration
- Solar energy
Blaming China
Politicians and journalist are going to blame politicians, immigrants or China for the stagnation, since most do not acknowledge that technological innovation is largely responsible for increased wealth and economic growth.
Impact On Our Lives
If you have been a budget shopper, it is now more sensible to invest into top of the line electronic devices. A fast premium smartphone or PC will still rank as a fast device in 5 years.
Next year’s new device will probably offer only ca. 5% longer battery, 5% more pixels and 5% more speed. The main improvements are small yearly functionality increases through software updates.
Impact On Intel
Classical microprocessors are becoming a commodity. Intel needs to rebrand itself. From a leading innovation company to a lifestyle brand selling a commodity. The playbook for commodity marketing has been written by Coca-Cola, so expect similar marketing.
Impact On Kurzweil
Kurzweil will face a new generation of people doubting his singularity predictions during the time of stagnation. He is currently writing the second installment to his 2005 Book "The Singularity Is Near". I predict that his new book "The Singularity Is Nearer" scheduled for 2017 will probably have a chapter specifically focusing on the public doubts emerging at the end of Moore's Law.
Outlook
If you zoom out mentally and temporally you see a new kind of transitional period. We are not just switching a paradigm as seen with jumping from transistor to integrated circuit.
We are going to cease using miniaturization. Next time we use a new technique, probably
quantum entanglement effects.
I estimate the end of the age of miniaturization will arrive within the next ca. 1-5 years. Let's take the oldest surviving mechanical clock that strikes the hour as a starting point of the miniaturization trend: the Salisbury Cathedral clock (build in 1386). We thus will witness the end of an age that lasted 630 years.
Do we even need to speed up anymore? As long as there are a few
ambitious, visionary humans left, one will always yearn for building a faster,
cheaper computer.
I hope you found this article interesting. I am open to your feedback and corrections.
Stephen Forrest, Vice President for Research, University of Michigan, "The End of Moore's Law"
Wally Rhines, CEO, Mentor Graphics, "The End of Moore's Law"
Eric Ladizinsky, Co-Founder/Chief Scientist, D-Wave, "The Future of Quantum Computing"
I hope you found this article interesting. I am open to your feedback and corrections.
Further Reading
- Washingtonpost.com -Why Google’s new quantum computer could launch an artificial intelligence arms race
- Wired.com - For Google, Quantum Computing Is Like Learning to Fly
- Google Research Blog - When can Quantum Annealing win?
- Independent.co.uk - The end of Moore's Law?
- Ray Kurzweil - The future of Moore’s law
- Wikipedia - The predictions of Ray Kurzweil
- Hackaday.com - Exponential Growth In Linear Time: The End Of Moore’s Law
Further Watching
Stephen Forrest, Vice President for Research, University of Michigan, "The End of Moore's Law"
No comments:
Post a Comment