Skip to comments.‘Tech Surge’ Planned to Fix Obamacare Exchanges
Posted on 10/20/2013 2:10:29 PM PDT by Oldeconomybuyer
The Department of Health and Human Services said Sunday it was bringing in outside help to resolve some of the technical woes that have beset the federally run insurance exchanges, which the agency acknowledged has not lived up to the expectations of the American people.
We are committed to doing better, agency officials said in a blog post that also said that our team is bringing in some of the best and brightest from both inside and outside government to scrub in with the team and help improve healthcare.gov.
Spokespeople for the agency didnt immediately respond to questions seeking more information about the development, which it is billing as a tech surge.
(Excerpt) Read more at blogs.wsj.com ...
Technology is NEVER a substitute for good management (decision making)
And who will be determining who "the best and the brightest" are? How will the administration know?
Will there be a new turkey as project manager, willing to serve as sacrificial lamb?
The administration appears already to have made a political decision not to do a redesign and rewrite - suppose that doesn't work? What will they do then?
Has anyone considered how expensive and time-consuming it will be to maintain this turkey, what with all the hastily applied patches (that likely will cause other "glitches")? OMG!!! Heaven forbid that anyone should do any thinking instead of running around like chickens with their heads off.
It would take nothing less than scrapping the entire existing system and starting with an ARCHITECTURALLY SOUND DESIGN.
And at a minimum, it would take 2 to 5 years.
No matter how many warm bodies you throw at it.
This thing is going to be (more of) the same CF.
I’ve got ya by 10 years, and I agree with your assessment.
Hey Laz, will this be like Obama’s “surge” in Afghanistan?
Headline: Obama to increase tech force by 10,000 in effort
to conquer glitches.
Headline: Obama to pull tech troops out after Christmas...
Headline: Glitches now control half of province...
You've clearly been in this industry for a long time. You've done a damned good analysis.
I cannot fault a single one of your points.
I would add, the only way the current system could EVER work is if they standardized all integration interfaces. Kinda a Henry Ford approach to development at the government level. All data objects would need to be standardized, all components would need to have the same interfaces.
The only industry attempts I have seen that have (relatively) succeeded were in EDI, and in another venue (communications protocol) TCP-IP (and all related protocols). Those got pretty close to standardization.
But the GOVERNMENT? Succeeding at a standarization like THIS? I suppose ADA did a reasonable job but that was driven by the defense INDUSTRY, not the government.
This thing may come to reliable fruition in 10 years with the current 'bandage' approach. They'd save 5 years if they scrapped and restarted. The only *true*, *bulletproof* solution is a universal government standardization, and that would take at least 3 years right off the rip.
The only solution is a nimble team led by amazing people, or in this case, 50 to 100 nimble teams.
One can only hope...
My experience in this sort of thing is nil, but I am
a student of history and attempts to standardize
humans never brings any good.
Obamacare makes Hillarycare look positively workable,
You’d have thought Kerry could have at least given
him some tips...
Having done a lot of project management I can tell you with a lot of certainly, that throwing bodies to fix a dysfunctional implementation plan will fail badly. Throwing bodies that actually do work may help if the plan is solid and you simply need resources, but if the project execution plan is broken, as this one is, it will only get much worse. Me I’m hoping for an epic failure.
Witness our first attempt at it: The standardization of firearms, followed by successes in standardizing railroads, telegraph signals, and then manufacturing in general (Henry Ford being the most prominent example). Standardizing data is the next logical step, and it escapes me why this has not been undertaken before now. We've made half-hearted stabs at the concept, but .. what we need is a Data Revolution, like the Industrial Revolution was.
I would have thought that as the internet matures
standardization would naturally develop.
Although this can conjure up images of the tower
of Babel, it would seem a logical step.
Glad to see Ted Cruz is the most important man in DC. Outside advice is probably; "Add more servers".
It won't be fixed because I'm not there. I've written train wrecks, so I know how to fix them.
No, it didn't.
Well, parts did, like HTML, XML, HTML5, and the like. But the underlying programming did not.
Then too, standardization of parts and manufacturing
is less than two hundred years old or so.
I wonder what will happen when it can be applied to
...I’ve written train wrecks, so I know how to fix them...
our team is bringing in some of the best and brightest from both inside and outside government to scrub in with the team and help improve healthcare.gov.
Scrub in with the team.
Don’t ya just love those medical metaphors.
I hope they all wash up first.
I do in house software, so my perspective isn’t the same as big team kind of software writing. However, as I go back through my old code, I always find a lack of documentation and places where I haven’t a clue what I was doing. I would suspect in something as massive as Ocare, the new team will have to take several steps back just to figure out what the hell is going on, all to understand a failed product. 50/50 they would want to rewrite parts from scratch.
Who pays for this rewrite, by the way? Medicaid enrollees???
Let TIm Cook and Zuckerberg take a look at it. There’s a reason they didn’t want this job. But now 0bozo is calling in his chips with an all hands on deck SOS. I wonder if they will honestly tell him that the design and architecture suck, and to start over..
“”Having been through one or two of these myself, the new team will uncover so many fundamental flaws in the architecture that they will require at least 2-3 years of redesign, plus another year (at least) of testing prior to roll-out. Add in bureaucratic delays and you have a recipe for continuing disaster.””
Hope you’re right!!!! We’ll have front row seats to the unraveling of obozo!!!! We might get to see him taken out of the WH in a straight jacket! A dream come true.
Fundamentally flawed. Solution, throw more people on the job. We had a saying where we worked. You can’t take 9 women and make a baby in a month. Pull the friggin plug.
For those of us who predicted this, it's right on schedule.
ObamaCare is exhibiting classic symptoms of the Death March:
In project management, a death march is a project where the members feel it is destined to fail and/or requires a stretch of unsustainable overwork. The general feel of the project reflects that of an actual death march because the members of the project are forced to continue the project by their superiors against their better judgment.
The fields whose project management practice first named these related phenomena are software development and software engineering. Other fields have since recognized the same occurrence in their own spheres and have adopted the name.
Death marches of the destined-to-fail type usually are a result of unrealistic or overly optimistic expectations in scheduling, feature scope, or both, and often include lack of appropriate documentation or relevant training and outside expertise that would be needed to do the task successfully. The knowledge of the doomed nature of the project weighs heavily on the psyche of its participants, as if they are helplessly watching themselves and their coworkers being forced to torture themselves and march toward death. Often, the death march will involve desperate attempts to right the course of the project by asking team members to work especially grueling hours (14-hour days, 7-day weeks, etc) or by attempting to "throw (enough) bodies at the problem", often causing burnout.
Often, the discomfort is heightened by the knowledge that "it didn't have to be this way," that is, that if the company wanted to achieve the goal of the project, it could have done so in a successful way if it had been managed competently (such as by devoting the obviously required resources, including bringing all relevant expertise, technology, or applied science to the task rather than just whatever incomplete knowledge a few employees happened to know already). Patent underresourcing is especially offensive at a large corporation with sufficiently deep pockets; at least at small companies, a gap between resources and needs is understandable, but at large, profitable, cash-rich companies, underresourcing is not a necessity and thus feels to most workers like stupidity. Business culture pressures, such as the long-noted phenomenon of corporations pursuing short-term maximization of profits via cost cutting or avoidance that is damaging to long-term best interest, may play a role in addition to mere incompetence.
Among the most infamous death march projects are the Denver Airport baggage handling system and WARSIM, a U.S. Army wargame. The latter project was originally called WARSIM 2000 at its inception in the early 1990s. A decade after its original scheduled delivery date, WARSIM has yet to support a single Army training exercise, but is still being funded, largely to vindicate those who conceived of the system and defended it over the lifetime of its development. The WARSIM schedule slipped many times. Moreover, WARSIM has a clumsy architecture that requires enough servers to fill a small room, while earlier "legacy" wargames run efficiently on a single standard desktop workstation.
The term "death march" in this context was discussed at length in Edward Yourdon's book Death March: The Complete Software Developer's Guide to Surviving 'Mission Impossible' Projects (ISBN 0130146595), which has a second edition simply titled Death March (ISBN 013143635X). Yourdon's definition: "Quite simply, a death march project is one whose 'project parameters' exceed the norm by at least 50 percent." 
Exactly what I was thinking. Now to see how long it takes before they make the classic mistake of starting over from scratch instead of fixing the existing code.
So, he’s bringing in the “techies” and the “best and brightest.”
You didn’t have “techies?” You didn’t have the “best and the brightest?” So why the H not?
That’s what happens when you hire your friends instead of the pros.
Drudge has “Emergency surgery” headline.
No. Bring in the DEATH PANEL and put this dog out of its misery.
The term "Flat Line" will soon be used.
Regarding throwing more resources at a software problem:
“You can’t get nine women pregnant and have a baby in a month.”
Don't mean I sure wouldn't want to give it a shot. ;)
———all without accountability——
either you make it work or you die........ that’s the deal
it’s called the law of diminishing returns
for every person added after the overload point, there are diminishing returns on the labor investment.
I gotta remember that one. Brilliant.
In some cases, I might agree with you.
Not in this one.
Start fresh and you have a small shot. Keep THAT code and yer dead.
(I hope they keep the code) lol
I could start with MySql, Apache Wicket with Twitter Bootstrap and crank out that site in two weeks, bug free.
I've been on Enterprise teams, and right now I'm on a quasi-enterprise deal. It's a single guy doing it -- me -- but I must serve as BA, DBA, PM, coder, tester, documenter -- and I'm pulling it all off only by the dint of my experience.
When I heard the particulars of this enterprise, I gulped and felt sorry for the Obamacare team. As much as I hated the concept, I felt sorry for the poor fools that got sucked into it.
Actually, I also doubt that anyone could crank out this size of a site -- with this sort of user load, and with the number of systems you would need to integrate to -- even with a solid team of A+++++ players -- in anything less than 4 years.
Even Facebook evolved over a half-dozen years. Amazon, a dozen.
I am curious about the number of lines of code. There must be a ton. Anyone new to the project will take some time to get up to speed. With so little testing, the original programmers don’t even know if they captured the required functionality in the original code.
Best thing would be to pull the web site and after the ‘new’ experts get a handle on things issue a new schedule of releases that result in the full desired functionality being delivered.
I predict that will not occur during Odumbo’s second term.
Well to clarity, I wouldn’t build the whole thing at once, just put out the parts you need to get started, then add in the rest over time.
They didn’t need the entire system to be 100 % complete to get it up and going, they could phase different pieces in over time.
Waterfall-component approach. Only viable if a solid plan for ALL the components are in play. You could Agile each component for best speed, but the design itself would need to take about 9 months for a good solid peer-reviewed “Give it your best shot, shoot down my idea” brainstorm design series of sessions.
You’d also still have the issue of integration to over 90 seperate applications. If you could leverage the agencies owning the apps to deliver a certain interface, that could help your schedule.
‘sites problems extend beyond well-publicized front-end obstacles’
Nice way of saying that the backend sucks as well.
It MIGHT be feasible by the end of Obama's fourth term.
6 months to a year? HA HA. Too late. Not possible. You’d have to start from scratch to get it really fixed.
That’s going to cost them. Uh, I meant us.
I think I’m going to go with post 97. :-)
The basic functioning of this web site should be so simple, that any decent programmers should be able to make it from scratch in 6 months to a year. But part of the problem seems to be that Sebelius and the HHS were extremely slow in getting the specific requirements to the programmers. The other problem is to what extent this web site has to interface with pre-existing government databases like the IRS. That part could be an enormous hassle for the programmers and might not be possible to do in the “real time” way this web site was expected to perform. But stuff like selecting the security questions and password for your account...it’s inconceivable how that could be causing errors, and that indicates the original programmers didn’t know what they were doing.