This proposition sounds nice and full of good desires; however we seem to forget that when we call the semi electronic industry strategic we are really referring to an industry based on one technology "Planar Lithography" that so clearly seems to be running the end of its course rather quick.
We also forget that there is no correlation between high tech and economic growth; in fact a mature high tech industry is just as bad as a mature low tech industry. None of them grows, none of them creates good jobs and none of them attracts capital.
Economic growth is correlated to business growth, and that is not only function of markets but of skillful exploitation of market opportunities. A small market growing fast attracts capital, creates jobs, generates ROI and delivers investable capital surpluses. A stagnant market consumes capital, and if corporate welfare kicks in, it also consumes taxes.
EU got into the semi current disaster by trying to interfere in market processes and by trying to catalyze business expansion through subsidies and targeted investments.
Once again the EU is headed downwards, this time by trying to "correctly" guess the growth industries of the future, and once again will end up in another Quimonda, AMD+Globalfoundries and Infineon disasters.
The semi industry is now mature, the game is now played through cost cutting, by extending plant life-cycles, and by living with diminished profits and with decaying R&D expenditures.
There will be fewer growth segments in the microelectronics sector; however they may come from new technologies, perhaps bio inspired, perhaps from architectures or even perhaps from growing networks or older systems. Only the market can signal the right course of action, signal the path towards higher ROI, towards higher rates of VC deployments and towards higher standards of living.
The EU commission, in all of its infinite wisdom, cannot replace the outcome of the most complex social game ever organized; that is the game of the competitive markets.
Blog Doing Math in FPGAs Tom Burke 13 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...