I’m immediately reminded of the healthcare.gov project, starting out with a budget of ~$90 million (which is already suspiciously high) and subsequently costing ~$2 billion, before launching with so many issues that it was initially unusable.
Yes, and what came out of that? Some of the people who rescued that project with modern software development practices went on to found Nava, a public benefit corporation, to help the government do better work:
I also have a friend who left Google to join numerous other high performers at 18F and know for a fact that they have done good work benefiting taxpayers:
I believe the U.S. Digital Service is better than it once was, and while it's not all rainbows and unicorns and super efficient everywhere (no doubt there are tons of huge problems with money being wasted), I do think there is improvement, and hopefully some people reading this will go help out.
I'd much rather ATTEMPT to have an efficient free-file system than not try at all.
Running over budget? Sure. But I’ve worked on much more complicated outsourced corporate projects for companies whose total revenue is only a fraction of $2 billion. Public sector inefficiency is in a league of its own.
To be clear I’m not saying it’s a bad idea either, or not worth the money. But I would be surprised if it didn’t cost billions to implement.
It didn't cost billions and it was a key incident that led to a huge revolution in how federal websites and projects get built, e.g. 18f and the US Digital Service.
That was a much more technically difficult project than an IRS tax filing site will be.
It involved systems from multiple agencies and jurisdictions that had been developed separately and not been designed to exchange data with systems outside their own agency.
The IRS systems already talk to everything they need to talk to.
