It’s not just 5 million lines of code, folks.
It’s not just having to call Verizon to the rescue.
It’s not just that it doesn’t work now.
The Healthcare.debacle website is a harbinger of doom because it reflects the fundamental inability of government to run such a sweeping portion of the economy as is represented by the health care sector.
The problem is inherent in the government process, as this Washington Post article unwittingly exposes (bold in original, italics added):
Timothy B. Lee: In your work as a computer science professor, you’ve taught a lot of smart programmers, many of whom went on to work at top technology companies. Why does government have trouble recruiting this type of worker?
Ed Felten: I think some of it has to do with the way government personnel works. Top programmers are in high demand, and they can get excellent salaries or the possibility of major upside from stock options. That’s true even in entry-level positions right out of college. It’s difficult for the government to compete with that on a civil service pay scale.
Another part of it has to do with the slowness and inflexibility of procurement processes, which people with a software development mindset find frustrating. It takes a long time to do things, and there’s a requirement of formal documentation and requirement statements and so on.
That makes it difficult to operate with the kind of agility that startup people are used to, this idea of putting out a simple product and then iterating rapidly, learning from your customers and scaling up gradually. That’s the usual model that startups follow. That’s difficult to do in government, and it’s not what happened with HealthCare.gov, which was meant to start operating immediately at full scale and with full features.
That health care dog won’t hunt. It can’t. First it has to fill out a mountain of paperwork and go through a hunt-procurement process that will take months.