Brook’s law and the Obamacare website

nine-women-one-month, the Obamacare website, has so far been an unmitigated disaster. Obviously there are serious programming problems that go far beyond a few “glitches” or bugs, beyond heavy traffic and server overload.

Anyone who knows anything about IT development knows a complex program needs to be tested extensively before it’s rolled out. And that didn’t happen. We know now that testing was inadequate because there was little time left for it. We know that a major change (requiring accounts be set up before options are presented instead of after) was requested just two weeks before the site opened.

Developers across the country have been rolling their eyes. They know what’s been going on. They know what was done and not done, and when, to get the website up and running. They know how much time was allotted for actual development (not nearly enough) and testing (almost none) before the October 1 launch date.

And they know when government officials promise the site will be fixed in one month because a “tech surge” has been implemented, that promise is probably worthless. (Of course, it depends on what the meaning of the word “fixed” is … )

Why? Brook’s law.

Brooks’s law says “Adding manpower to a late software project makes it later.” Wikipedia has a full explanation of the principle and how it works (which you may recognize even if you haven’t known its name). When more workers are added to a project, they must be trained and brought up to speed by the existing workers, temporarily decreasing the amount of work done by the existing workers. At the same time, new communications channels must be established to fully integrate the new workers, further complicating the effort to keep the entire team working in sync. Brooks’ succinct and memorable summation:

Nine women can’t deliver a baby in one month.

But it looks like the Obama administration is trying to do just that.

Bets, anyone?




8 thoughts on “Brook’s law and the Obamacare website

  1. Personally, I think the whole mess proves the fact that government rarely if ever does anything better than the private sector and usually worse, and usually more expensively, and usually more incompetently. That’s the problem when you have a group of people who spend other people’s money and have no skin in the game. They just believe that no matter what the mistakes, they can go back to the endless money source and keep patching it up and throwing more money at it forever.

    And honestly, what did anyone expect? When you have a law that is 3,000 pages long you have to expect that that degree of complexity is a recipe for disaster.

    They’ve literally had years to figure this out – but apparently that was not important to those responsible. If this had happened in a private company and had had this kind of impact on the public, people would be going to jail, being ravaged by the press, and testifying in front of endless Congressional hearings. Since it’s the gubbermint, we’re all just expected to give them a pass and theorize on what went wrong. Seriously?



    1. It’s a colossal mess on a scale that only the government could produce (but then, the legislation was seriously flawed to begin with). A private company with such an awful roll-out would simply go broke because no one would do business with them.

      I am, with great difficulty, refraining from a lengthy rant about all the Obamacare problems that have been exposed this month. Much of it I’ve covered before, and I intended to limit this post to the website’s problems. However, the thing that shocks me the most right now is the revelation that hundreds of thousands of people who already had coverage with private companies and were happy with it have suddenly and unexpectedly been kicked off their plans because of Obamacare regulations that were not publicized. In fact, we were told repeatedly — by the president himself — that people who liked their insurance could keep it. Dumping people off existing private policies because the government deemed those private policies inadequate is unconscionable.

  2. I can’t say it would have made things perfect, but I think writerchick’s “no skin in the game” comment points to a way this could have worked much better: Everyone involved in setting this thing up, from the president, to congress, and right down to the lowest ranking worker in the development chain, should have known up front that they themselves would have to benefit or suffer right along with the rest of us!

    1. Since my son is a developer, I’m hearing a lot from his point of view, and I’m not blaming the developers in this. The government took too much time writing the specs, cutting into the time left for the developers to do their job. They did the best they could, right up to the launch, with no time left at all to do the testing that should have been done. Testimony yesterday indicated that they had repeatedly warned the higher ups that there were significant problems, that the site would not be ready by October 1, that not enough testing had been done, etc. But management, as is so typical, was oblivious and perhaps didn’t even grasp the technical problems involved. They just kept to their schedule of an Oct. 1 launch, no matter what.

      So no, I don’t blame the contractors in this case and don’t think they should be stuck with the results of this debacle. I would, however, be more than happy to insist that every politician and political appointee in DC be required to switch to Obamacare and lead by example!!

  3. How long did they have to create this site? When was this passed? March of 2010? If so, then if developing the site was put into action immediately, there is more than enough time to get this site up. I haven’t seen any timeline as to it’s development and would like to know.

    1. I’ve forgotten the figures my son got. Something like two and a half years from start to finish — with the usual hangups — the specs weren’t done on time and the techs couldn’t get started without them. They weren’t allowed enough time to get their work done and ended up the in the last few weeks before launch trying to implement a major change rather than testing the final product. The Son thinks they ought to scrap what’s been done and start over with something short and sweet, just enough to start processing applications. I’m inclined to think they should stop telling people it will be ready by the end of November and just admit it’s going to take months. “Sorry folks, come back in 6 months and we’ll try again.”

        1. You couldn’t do any worse than they have!! The only managers I trust to know what they’re doing are those who’ve been in the trenches first, who’ve risen through the ranks, done the job themselves, and know exactly what’s going on.

... and that's my two cents