Oh, and, you know, Congress.
My thinking has kept me at the meta- end of this debate. I have tried, even in talking in places like FOX Business, to maintain two very simple premises:
- Healthcare.gov was a catastrophe because it was designed by committee, perpetrated by God-awful procurement processes, managed by an army of sub-standard sub-contractors and ignored the only thing that mattered: what the users wanted; and
- Open data, open standards and open software are the only things that can make this and other projects like it truly deliver what the users need to effect better outcomes while reducing the cost and size of government.
It depends on what the meaning of the word ‘Fixed’ is
I am highly concerned that Clay is confident that healthcare.gov will be “fixed” in “a few days”. This is terrifying. Healthcare.gov, when we view it through the lens of the users, was an abject and objective failure across the board. From a technical standpoint, it was a diabolically complex failure.
By “fixed,” I can only assume that people are referring to the ability of the website to deliver fundamental information of the sort people were saying they wanted – ‘How much, for cryin’ out loud, is this Obamacare thing going to cost me, or should I stick with what I got now?’
If that is “fixed,” then the only thing that is left is, you know, architecture, data handling and security.
Wait a minute. That is, you know, a whole lot of stuff left.
When I listen to the finger-pointing testimony of Cheryl Campbell, the weasel, future-former manager and SVP at CGI Federal before the House Committee on Energy and Commerce, I hear things that concern me. For example, hundreds of data sources, hodge-podged together in a really inefficient way; hundreds of connections to disparate and heterogeneous networks; terrible – in fact, crippling – enterprise identity management problems and poisonous contractor-pool politics; IRS system involvement… and everything points, in my professional opinion, to at a minimum a highly tempting landscape for criminal cyber attackers.
Because of the procurement and management and project management and coding issues, in my professional opinion (based on testimony and other statements by those involved), I believe little or no security testing or user testing was built into the software development lifecycle. If that is true then this means, flatly, that the code is insecure. Mind you, I’m not talking about the front end website – that thing has been fished out pretty good, and it’s my understanding from a number of places that the front end has been solid. As I mentioned on FOX, the fact that it pops up and all the static bits work well indicate that it’s OK.
Which is highly analogous to having a 2014 LTZ body and a YS 115FZ-WS 4-Stroke RC airplane engine under the hood that’s been built by my son and installed by me. This leads me to my first point:
It turns out that defining what “Fixed” means is as highly politically charged as was defining what “Broken” means. While the administration made a big deal about owning up to problems, and the President claimed that “The problems were inexcusable,” what he actually stated was that the problem was that the healthcare.gov site, “hasn’t worked as smoothly as it was supposed to work.”
Oh really? Anything after the first two words of that quote are walkback.
And Washington insiders referred to “Glitches.” As I said yesterday, A “glitch” is when I am driving through a tunnel and my satellite radio craps out for a second. Healthcare.gov is a complete and unmitigated catastrophe.
If we cannot define broken, lots of luck with defining “fixed.” I would say that, “fixed” would be a state in which people can:
- Get information; and
- Get an estimated cost and plan quote; and
- Register and apply for a specific plan; and
- Follow up and receive further information about the acceptance and delivery process; and
- Ask questions; and
- Be secure to reasonable e-commerce standards, i.e., that of Amazon.com, PayPal or BestBuy.com
If anyone believes that anything but the first two and possibly the third will work in days, and the rest in months, they are delusional. It’s not about getting a site up – my kid can do that on WordPress.com. It’s about getting a service that will be the most visible repository of personal information on the American people up and reliable.
Get it Open
The way to get this going faster is to open the process: open the code, and most important, open the data. Don’t tell me that the data can’t be open – a large subset that covers millions of Americans is already open and available. This proves that
- It can be obtained
- It can be Open
- It can be made available as an API-accessed service
Is the fact that three guys could open JimAndTonysObamacareCostEstimator.com in two days for the cost of Twinkies and Red Bull so abhorrent to the administration that they won’t open what is clearly government data? There are more than 98,000 datasets on data.gov – surely this one won’t cause more problems than the MultiBeam Bathymetric Data Base (MBBDB) when it comes to Americans’ ability to work with the data and make great solutions and applications from it.
Open the code. Open the data. Let’s let the best people in the world make the best apps in the world to get people the information they need to decide on a health care plan.
Healthcare in America is tough enough without having to fight with a website to get it. Opening the process and letting the users drive the technology and the best, most competent technologists build it is the only way this will ever be “fixed”.