3 Common Misconceptions about Healthcare.gov
1. Healthcare.gov broke because of overwhelming traffic
This is like building a bridge out of driftwood and nails, and then claiming that it collapsed because too many cars were driving over it at one time. Each page load on Healthcare.gov generates over 60 seperate HTTP requests to the same host, downloading completely unminified and uncompressed assets for a grand total of 2.5mb transferred. This is an additionally crappy excuse because tools to simulate server load are everywhere. (One even ships on OS X — open up a Terminal window and type ‘ab’.)
2. Healthcare.gov’s code was developed by Development Seed and open-sourced by HHS
While there is an open-source release of the healthcare.gov frontend, this is only one component of Healthcare.gov, and it’s not the one that’s currently failing. The CMSgov/healthcare.gov repository is the static Healthcare.gov frontend, built by Development Seed and raved about in the Atlantic amongst other publications.
3. Healthcare.gov was so hard to build because of strict regulations in the ACA
As a web developer and someone who has spent 6 months scrutinizing federal IT contracts, this claim is extremely hard for me to believe. Not once in my career have I thought to myself, “wow, if only my client hadn’t have given me such strict specifications.” Sure, a myriad of regulations can be difficult to navigate, but our job as programmers is to create well-architected abstractions of complex systems. You know what makes sites like Healthcare.gov hard to build? Layers upon layers of contractors, a high ratio of project managers to programmers, and a severe lack of technical ownership.