I think that a key point is missed here. No one questions that each device has a thorough and voluminous trail of testing. What they question are the standards that the tests are held to. As the fine article stated, most if not all devices are only tested against in house developed criteria. And while the device in question may have been found compliant to all internally generated requirements, it can still fail in life threatening ways that may not have been tested. The point is that the FDA largely defers judgement on just what to test in a device to the manufacturer itself, which amounts to self-regulation. And we all know that corporate psychology does not possess the capability to regulate any action other than pursuit of profit.
I contracted as a software requirements test writer once, and although the company was very thorough and rigorous, the requirements and tests for the versions prior to my time were not adequate to inspire my confidence as an long time programmer. In many cases, the requirements were inconsistent or missing. Missing requirement were not tested. Tests in many cases were written by trained writers, not programmers or testers. Many of these tests would pass as written because the test procedure was wrong. Of course the process is long, and most companies including this one eventually find and correct these types of issues. I would still have to say that leaving it up to each manufacturer to determine just what is to be tested is a recipe for device failure. At the end of the day, the product I worked on was considered compliant because all of the self-determined software requirements passed the V&V tests they wrote themselves.
I think that most people see the value of getting a second opinion in medical decisions, so it wouldn't be far fetched to think that people might want a third party to review medical devices too. If the FDA's ruling acts force them to defer testing to the manufacturers, then the public deserves the right to seek that second opinion. It seems that these days, the FDA is little more than a source of get out of jail free cards to the industry it is supposed to regulate. The point this article tried to make is do you want decisions about the quality of a device implanted in your body to be influenced by reputation or profit motives? I think we can all safely answer no to that question.
Licensing professionals is another method large corporate / old wealth interest use to limit competition. The original industrialists, like Ford and Carnegie do not want independent producers to compete with them. It happened to every tradesperson since the mid 1800s. Requirements for professional licensing cuts the supply of labor, raises the market prices for those still participating and increases the demand for those skills and products. With the market primed, a corporation with near unlimited capital and fully limited liability comes in with mass produced and cost minimized solutions that undercut any professionals left in the market. The result is few if any licensed professionals remain sole proprietors or partners for long. Eventually they take the work at the nearby factory for less money, paying more for "professional" services like doctors and accountants (death and taxes). Those company benefits come out of their wages and most of their annual raises will go towards maintaining their health insurance.
Take the early Mini / Personal Computer days for programmers. Very little off the shelf software existed. Business had little choice but to hire free-lance programmers to write custom software. Most needed to keep these programmers on retainer in order to make quick alterations of code to meet new market conditions and all were worried about long term support. The independent programmer could have made a career at one or two firms just as a lawyer or accountant might today if they have large clients.
But in order for off the shelf software to have a serious market in large enterprises, the independent programmer would have to go. So they get the "professional" treatment just like all other tradespeople have received. Certifications are issued and packaged software professionals are the replacement for highly skilled and familiar free-lance custom designers. Business adjusts to an inferior product and less competent labor in order to perpetuate the idea that bigger is better and a piece of paper is more important that practical know-how. Everyone is expected to get back in line, the wild frontier is now closed to the general public, come back in fifty years for the grand opening of the hacker museum were we celebrate how much better it now that nanny Micro-Mac-Bm provides all our "trusted" information needs with new titillating models coming next month.
The writer states in section 0 that proprietary software will stay indefinately. Had he looked at the industrial revolution an even paid much attention to the computer revolution he would have seen that published standards endure and proprietary standards are starved off from the market.
Linux needs open standards and specifications so that end users can retain their data and devices. Microsoft did that with standard device driver models, PlugandPlay protocols, and read access to as many competing data formats as they could buy. I think that until the models changed in Vista, most people could get their hardware to "just work" with Windows from version to version. In fact, I think DOS users had more open and complete device interface documentation than Linux enjoys today. But without agreed upon industry standards to build critical power applications and device drivers, Linux will struggle to provide for the needs of the average desktop user.