SoylentNews is people
SoylentNews
Sen. John McCain pens an opinion piece on Wired:
America's military technological advantage is eroding—and fast.
For the past decade, our adversaries have invested heavily in rapidly improving their militaries to counter our unique advantages. At the same time, the speed of globalization and commercialization means that advanced disruptive technologies are now—and increasingly will be—available to less sophisticated militaries, terrorist groups, and other non-state actors.
Maintaining our military technological advantage is about much more than a larger defense budget or a better fighter or submarine. These things are important, but to give our military the capabilities it needs to defend the nation, the Department of Defense must be able to access innovation in areas such as cyber, robotics, data analytics, miniaturization, and autonomy, innovation that is much more likely to come from Silicon Valley, Austin, or Mesa than Washington.
[...] There are those who say that even with changes like these, our nation's innovators simply aren't interested in doing business with the Pentagon. And after spending much of my career in Washington scrutinizing Pentagon business practices, I am not exactly surprised to hear such sentiments. But in the final analysis, I believe the brightest minds will always be driven to solve the world's toughest problems. These are the problems our military confronts every day. And these are the problems we can solve if we create an acquisition system that enables the Department of Defense to take advantage of the best minds, firms, and technologies that America and the world have to offer.
Original Submission
(Score: 2) by MichaelDavidCrawford on Friday June 12 2015, @06:49AM
I once scored an embedded consulting gig because the primary contractor selected a part then selected me because I had expertise with that part.
What they should have done was to have someone with domain expertise select the part. Just because the vendor's data sheet says "Taste's great, less filling" doesn't mean it's "Bear, Ready to Drink".
While my part in the contract was completely unclassified I am dead certain I know who "The Client" was.
After a while I called to say that I had some problem I was having to debug... then that I was unable to debug... then that I suspected a mask error in the chip and was hoping for a workaround, maybe the vendor's tech support could help.
The primary contractor blew a lot of money to have prototype boards manufactured in which this steaming pile of silicon was a crucial, central component. They distributed a bunch of the prototype cards to The Client's various sites then called to tell me that The Client was real happy with our work.
"So how's that firmware coming, Mike?"
"YOU SELECTED THE WRONG PART!" click.
It would have been quite easy to determine that the candidate part was the wrong part by looking over the errata for each of its many revisions.
The basic problem was that the chip's manufacturer had a specific, quite popular mass application for that chip but designed it in a largely conceptual way so that it would work for lots of other applications as well. I can see how that would be a good idea but it was always clear to me that the vendor had not tested that the chip actually worked the way it was documented to work.
Yes I Have No Bananas. [gofundme.com]