Healthcare.gov’s 500,000,000 Lines Of Code?

This popped up in my Facebook feed today:

Click to enlarge

You can enlarge it to see the punchline, but it’s this: Healthcare.gov supposedly requires 500,000,000 lines of code.

I’m going to call BS on that.

The idea is that the site was programmed with such an absurd degree of incompetence that it bloated to an unprecedented size. But when I saw the graphic, it seemed like someone was just saying “One billion gagillion fafillion shabadabalo shabadamillion shabaling shabalomillion.”

I guess if you count the front end, the database, every line of off-the-shelf code, and every state system, it’s maybe slightly a little bit conceivable that you can get 500,000,000 lines of code

Also, “lines of code” is not really all that useful in determining how big or inefficient a program is. I guess compiler reports counting all the libraries and everything else in all the various pieces of the system might yield half a billion “lines” of “code.” For example, more than 80 libraries called Boost have 19,458,640 lines handling all kinds of tasks so programmers don’t need to reinvent them every time. No one working on Healthcare.gov wrote a line of it.

Still, that’s still a long way from 500,000,000. It just sounds made up. The cost and non-functionality of the system is bad enough without exaggerating.

About Thomas L. McDonald

Thomas L. McDonald writes about technology, theology, history, games, and shiny things. Details of his rather uneventful life as a professional writer and magazine editor can be found in the About tab.

  • Clare Krishan

    if there’s a fault (I don’t doubt it just because of its monstrosity like you do) its not the graphic designer’s. The number comes from NY Times report referencing experts* on code to be REWRITTEN (implying there’s more code, and that some of that doesn’t need to be rewritten)
    “One specialist said that as many as five million lines of software code
    may need to be rewritten before the Web site runs properly. ”
    http://www.nytimes.com/2013/10/21/us/insurance-site-seen-needing-weeks-to-fix.html
    that got quoted elsewhere for example:
    http://thehill.com/blogs/healthwatch/health-reform-implementation/329583-report-obamacare-website-could-need-five-million-lines-of-code-rewritten

    ie professional pundits from “Expertopia” who knew Obamacare was social-engineering-in-secret-with-private-partner-collusion
    http://www.bloomberg.com/news/2013-10-30/what-everyone-knew-about-obamacare-and-wouldn-t-say.html
    “Why were companies bothering to massage their earnings if “everyone knew” that you had to back out the surplus charges? And why were equity research analysts issuing hold ratings when they actually meant sell? Presumably because not everyone knew that all these things were meaningless. Some people were buying stocks based on these insider tricks. But we never talked about them. We never even thought about them. By “we,” I don’t mean people writing the reports and financial statements; I mean those of us who were busily selectively revising them to get a truer result. No one in my financial-statement analysis class ever raised a hand and asked why companies engaged in all those gymnastics; we simply took it as a given. Some of my classmates presumably went on to write financial statements and equity research reports. But even those of us who didn’t just sort of … forgot about the folks all that “useless” writing was supposed to impress. We took it for granted that “everyone” knew which bits to pay attention to, and which bits were high-test horse pucky. That’s right: We’d moved to Expertopia. And the folks who didn’t live there were so far away that we couldn’t see or hear or even think of them.”

    Same for software programmers, no one questioned if they could earn a living writing “meaningless” code, so long as the political wind is in their sails they type away oblivious to client deadlines and testing outcomes. Same for Wall Street investors, no one questions if the fundamentals of the price discovery mechanism were “meaningless” now that the FED was adding $85 billion in new money to chase after old ‘encumbered’ assets (ie burdened with debt) – prices go up and up and up… until they don’t because there’s none left to pay (ie no young blood to cannibalize in spooky-Halloween-fashion).
    From The “Duh” File – Detroit Pensions
    http://market-ticker.org/akcs-www?post=225550
    I think it sucks that you’re not going to get your allegedly-earned “benefits” but the money to pay those benefits never existed and was never going to. This is exactly identical to knowing you make $27,000 a year, spending $37,000 a year, having $170,000 on the credit card and then whining that it’s “unfair” when the card company won’t let you run up another 10 large. This, incidentally, is our Federal Government right here, right now, today — just add the zeros to the above and you get our tax revenue, our spending, and our federal debt.”

    It really is that monstrous – here’s Philly and Allentown’s taxable real-estate (in relation to Detroit’s much-maligned non-performing wasteland of urban decay):
    http://www.zerohedge.com/sites/default/files/images/user5/imageroot/2013/10/Taxable%20real%20estate_0.jpg taken from
    http://www.zerohedge.com/news/2013-10-29/ten-us-cities-less-ten-days-cash-hand

  • Linebyline

    And then there’s this issue:

    if (foo == bar){
    echo ‘lorem’;
    } else {
    echo ‘ipsum’;
    }

    …can be rewritten as:

    echo (foo == bar)? ‘lorem’ : ‘ipsum’;

    Or for that matter, there are simple aesthetic concerns like whitespace:

    if (foo == bar)
    {

    echo ‘lorem’;

    }
    else

    {

    echo ‘ipsum’;

    }

    And that’s not even getting into whether comments count as “lines of code”: It’s no unheard of for a Javadoc block to be two to three times as long as the method it describes, at least for simple methods.

    And what *does* count as a line of code, according to whoever gave us this number? Whitespace? Comments? Are two short statements on a single line one line or two? What about a method call that’s split over multiple lines so each argument gets its own line? (I do that all the time for readability.) Does it matter if the argument is just a single value or variable as opposed to a more complex expression? What about the obligatory newline at the end of the file that Git complains about if you forget it?

    What I’m saying is, no one who knows anything about programming would treat “lines of code” as a seriously useful metric. Heck, IBM paying its programmers by the KLOC (thousand lines of code) is one of the stories older programmers tell us youngsters about the bad old days. The important things are: Is it readable, is it well-documented, and most importantly, does it work?


CLOSE | X

HIDE | X