As I’ve said, I don’t think that the healthcare.gov clusterfuck means that Obamacare is doomed. I assume it’ll be fixed eventually, and ultimately will be the least of the program’s problems. But, you know, I knew people in HHS who assured me (awhile ago) that Sebelius was on top of implementation, that competent management was her key skill. I’m not throwing all the blame on her – what do I know – but the idea of competent management is something we’re supposed to be about. We hated Bush for his policies, but also for his incompetent implementation for those polices. We hated the reverse Midas Touch, that he turned everything into shit.
Hopefully in the long run it won’t matter, but there’s a reason some of us expect our liberalism to be competent, especially when it’s really just technocratic centrism advertised as liberalism.
I’m definitely not optimistic, especially in the Radiohead sense (“If you try the best you can/ The best you can is good enough”). That said, if Sebelius was ever “on top of” the healthcare.gov implementation, she was in the same position as someone “on top of” a tiger or a stampeding water buffalo. If healthcare.gov does anything other than somehow, someday, after an avalanche of money is poured into it, sign up Obamacare participants, I hope it gives liberals some more leverage to attack the contracting culture in DC. When you combine complexities that Richard has documented with an outsourcing system that seems designed mainly to line the pockets of the private firms that participate in it, you’re going to get a clusterfuck. And that clusterfuck will happen despite the best efforts of smart people who are good managers.
Engineering complex software is one of the hardest things human beings do. The fact that every decade or so we see a new fad that’s supposed to revolutionize software development is pretty clear evidence of that fact. If we knew how to engineer software the way we engineer bridges or automobiles, nobody would be “reinventing” software engineering every decade. I watch these trends with a mix of amusement and disgust. The current one is “Agile Development”, which has a whole strange vocabulary related to Rugby. I know Agile is about to die because I’m hearing corporate types talk about “scrums” and “scrum masters”, two Agile buzzwords. My experience has been that as soon as the buzzwords become job titles, that trend is over, because the stink of corporate failure will overwhelm the magic bullet allure of the latest trend.
Good software development requires organization by smart people who are deeply committed to the task at hand, are directly responsible for the success or failure of the project, and have a big say in the resources needed and how they are deployed. Outsourcing firms throw (relatively) cheap bodies at tasks, they deploy resources stupidly and haphazardly according to their profit motive, and they live in a world where running up the bill, not delivering a working system, is the end goal. I wish Sebelius had changed that, but I’m not surprised she couldn’t.