Which arguments about Healthcare.gov are wrong

Posted by tobymuresianu on Oct 25, 2013 in Politics, Tech, Thoughts |

For the record, as an engineer with (an unfortunate amount of) experience with delayed software projects:

Using a “tech surge” to fix the problems with HealthCare.gov is a terrible idea, and every time I read about them “shipping some of the best and brightest minds” to Washington to fix these problems I cringe.

The more significant reason it’s a terrible idea is that you can’t solve software problems by adding more people or money. This is most famously illustrated by tech industry gospel The Mythical Man Month – a book describing the phenomenon of how adding people to a delayed project makes it more delayed.

Basically, what ends up happening is that you pull programmers who are building the product off and have them “bring up to speed” people that have just been brought in. They spend time training them, and then fixing the problems that the new people create, rather than working on the actual delayed product. The more people you add, the more overhead, and the less efficiently the project goes.

That said – I’m sure the experts brought in to fix the problem are aware of this phenomenon – and hopefully will be able to work around it, or maybe it’s all a bunch of hoopla to buy the developers time. At the end of the day – this product is just a website. It should not take that long to build. Obviously there’s important back-end stuff happening, but apparently that’s been functioning okay as the state exchanges have been using it.

You’ll hear excuses like “tech projects are often delayed” and “many websites have glitches” or “the traffic was more than what was estimated.” These are all ridiculous. Yes, tech projects are often delayed and time costs poorly estimated. However, this project is cost $170 million. $170 million. For a website. A website. $170 million. And Sebelius told the Wall Street Journal that ideally they should have had five years to work on the website. If every website took five years and $170 million to develop, there would be like three of them. No, designing an enterprise-class website is not easy, but the truth his that thousands of enterprises do more complicated things every day and manage to pull it off. As did the 14 states that managed to build their exchanges correctly.

People are saying Sebilius should be fired. I don’t know how responsible she is personally or not. It’s so easy to point fingers and I don’t think you should fire people just for vengeance.

I can’t say what the particular causes are – people are citing things like late requirement changes, which are the bane of many a software engineer, but in a well managed project should be prevented from happening.

But at the end of the day people are responsible for this. Anyone who didn’t raise alarm bells at a $170 million price-tag for developing a website should be fired, twice if possible. Obama and Sebilius both should have been told, if they weren’t, about the current status and risk associated with the project before launch and given the chance to delay it; and if they weren’t, whoever was responsible for not telling them should be held responsible. People who are bad at their jobs should not be given the chance to do new ones. I want to say that CGI Federal should not get more government contracts, but regardless after this debacle I’m not sure government agencies will be lining up to get some of that sweet HealthCare.gov magic.

Anyway, just wanted to share some thoughts I had – I had a little time to kill while I was trying to buy health insurance.

Copyright © 2018 Toby Muresianu – Comedian All rights reserved. Theme by Laptop Geek.