Here's some excellent insight about the signup disaster from someone who's been there and done that: http://www.cnn.com/2013/10/14/opinion/bellovin-obamacare-glitches/index.html?hpt=hp_t4 This is really worth reading. The author is a computer science professor, but you don't need to be a geek to understand him. His take, bottom line, is that this was a really big project and those tend to fail in the private sector as well, and most of the problems can be traced to a moving target. Politicians keep changing the functional specification (that is, what the program is supposed to do), keep changing the budget, keep changing the personnel involved, and keep the go-live date the same. Guaranteed catastrophe. Here's a good sample:
Funny, but that's exactly what those of us on the right were saying about Obamacare in whole. Ironic how healthcare.gov is a microcosm of the entire socialized medicine that Obamacare is, huh? Guaranteed catastrophe, indeed.
I did read the article, and it was pretty interesting. And it does nothing but confirm what I've been saying for about a week now: healthcare.gov is Obamacare in a microcosm. Getting the website up and running is the easy part, and yet that was epic fail. What part about my post confuses you? Is it the word "microcosm"? Too many syllables?
As the article informed you, getting a software project of that size and scope up on time and within the budget is damn near impossible. You should stop repeating talking points and read the article. Nobody is confused. You knew long ago that the ACA was terrible, and no matter what happens, that's "proof" that you were right. A software project is not a microcosm of anything, it's a software project. I once worked on a large software project under a contract with a major player in the microcomputer market. Turns out we underestimated the time and money that would be required (this is EASY to do, because you can't predict the unpredictable problems you'll have. MOST software budgets are based on no such problems cropping up). So come delivery date, we're weren't close to finished. But we HAD to be finished according to the contract! So we shipped what we had (which was no better than healthcare.gov at the time), changed the sign on the programmers' door from "development" to "maintenance" and kept on developing. What else could we do? So was this occasion a "microcosm" of the company I worked for? Nope, this was a one-time event, my company was well run, and quite successful.
Right. And I've been saying the exact same thing about Obamacare since day 1. That is why I laugh at the morons who give cost projections and say we're going to save money. Not only are those projections worthless, but anyone who believes them is just a f'ing retard. This website was projected to cost about $90 million and it ended up costing over $600 million. It is Obamacare in a microcosm. (And just how pathetic is it that a $500 million cost overrun is a microcosm of the whole system??)
Me thinks you are wasting your time. No really, these people have already decided the Affordable Health Act will fail, and *******mit they wont be happy until it does. Anyone who has ever been through a software launch in private industry knows everything the professor said was spot on. And I have been through an SAP launch, so I do know what I am talking about
What does it matter why the train wrecked? We have seen the train and it is wrecked on a gentle curve because the progressives put several different wheel sizes on the engine and most of the cars are way too high and way too long to go around any curve at all. There are one way doors and no windows so if you get on board the BH Obama-care train you can't get out and you will have no idea where you are going to end up. there are a lot of train conductors collecting taxes to pay the dems' cronies to drive the train, keep putting it back on the tracks and telling you to just remain calm and don't worry about the bumps in the tracks because all $600M programs have trouble on start up after three years of preparation.. If you need a heart transplant you will have to bring your own replacement heart and may have to help install it because the good heart surgeons will be retired or working over seas where they can may better pay.
As an architect, owners do do that... all the time. The programmatic design requirements for the structure change, the building systems and techniques are adjusted, budgets are revised... If a building collapsed, the architect would still be on the hook. I have had many times when I have told owners that they are going to have to either stop making changes, or bump the deadline. THAT is part of what clients are paying me for, to look out for their best interest. If the website wasn't ready, they should have said "we aren't launching it". One thing I have noticed from the computer fields is that they have excuses for a multitude of failures and embarrassments in that industry. Excuses are abundant. Government or private doesn't matter.
Call of Duty , a video game , was ready for the millions who were going to log In at 12:10 am . Seems like the government could have taken notes .
Lol , but we aren't talking about EA , which didnt ever have the server load of COD games . Millions played as soon as they got home .
Now imagine that for political reasons well beyond your influence, the design requirements are changed weekly and often drastically, the budget cannot be increased, the deadline cannot be changed. As an architect, you are intensely aware that under these circumstances, you are going to be WAY over budget and over schedule. So NOW imagine that, for political reasons, you are prohibited from saying so at risk of your job and your career. Yes, of course. Except of course the go-live date was set long in advance, and heavily advertised, and has become politically untouchable, so no help from that quarter. And "the owners" making these changes are, you know, the President and Congress and Supreme Court of the United States, so you're not about to tell them to go pound sand, right? You can bet all you own that those in charge of the software development have been telling the politicians daily all these things - that they are WAY behind schedule, that every engineering change order puts them further behind, and that the damn thing flat doesn't work. But those are mere technical details, and are so totally trumped by political considerations that they're not even worth listening to. I agree, this is a state-of-the-art issue. Those space probe programs go through testing like you would not believe - weeks of testing for every line of code, to the point where by the time the damn thing is launched, nobody has manufactured any of the components for YEARS. And sure enough, some obscure thing that wasn't fully tested had a hyphen missing, with fatal results. This despite code reviews out the wazoo, everything known to the state of the art. Computer code is unfortunately brittle, very small errors can easily bring down the whole thing. Even today, for Windows Vista (two releases old by now), I get literally thousands of bug fixes a month. Maybe someday we're pass through the singularity, when all computer code is written by computers. And even then, no possible computer code can anticipate every possible hardware glitch, much less have a graceful workaround for it.
I like you and stuff... but this is specifically what I did for many years, and on occasion still contract for... enterprise level portals. While I certainly agree that everyone involved at HHS should be fired for not getting their (*)(*)(*)(*) together... but they didn't even start coding this till this spring. The problem is an architectural one, which will have to be completely rewritten. It was not an issue of rewriting due to scope creep or changing targets. They had about 6 months to develop and implement a solution, and failed. The software was completely unready... and the administration chose to launch it broken rather than delay it and give the republicans firepower. And of course... then there was this terrific act of stupidity... But who cares about all that? Not the administration: http://www.nytimes.com/2013/10/13/u...ouble-at-health-portal.html?pagewanted=1&_r=0 So... mismanagement on every level... FULL AWARENESS as to the technical problems and failures... the WH pushed on... and parked a steaming pile at healthcare.gov... and pretended it was "normal" and to "expect a few hiccups"... which is essentially what you are doing. Rationalizing what amounts to an epic failure in leadership and project management. Oh... and the code is (*)(*)(*)(*). It is built in html and floats data around with js... Nobody does that. Nobody. Not weblogic, websphere, sybase, sun one, oracle... nobody. While js is used extensively to move data around and complete tasks... it is within a secure framework generating html etc on the fly. This is... a website. I mean... go wget it. And the administration has been lying about it. This is an interesting perspective from Forbes. I don't know that I buy it... but worth a read. If it is so good... why can't we just use a simple calculator for our state and see the costs without giving the government people do not exactly trust at the moment all sorts of information? NPR has a calculator... http://www.npr.org/blogs/health/201...uch-will-obamacare-cost-me-try-our-calculator Feel free to give it a whirl
Since you know what you're talking about with this stuff, I'm glad you could drop by and help out. I think starting late was a problem, and clearly a moving spec was a problem, and certainly massive mismanagement was a problem. When we get down to it, these turn out to be political problems far more than technical problems. When it turned out that the workoad the system needed to handle was far greater than anticipated (due to the High Court deciding that half the states could dump it all on the feds), the architecture (borderline at best) became hopelessly inappropriate. And at THAT point, I'm sure there were plenty of people inside the skunk works who were well aware that they were building a broken bridge to nowhere, BUT political pressure simply bulldozed them under. The Obama administration simply could not afford to admit being unready in the face of tea party trumpeting. I wonder if anyone today has the sense to be developing a workable architecture in the background while "bug fixes" are performed in the foreground. I do not trust this particular slant. While it might be politically expedient to disguise costs if they're going to be unexpectedly high, certainly this is not the way to do it. I think I understand this part. These things are complicated. I've seen the NPR and Kaiser calculators, and these are approximation tools at best, since rates depend on many factors - ages, family sizes, incomes, state/county of residence, smoking habits, on and on. And THEN you get into the meaningful details - just exactly what counts as family? What if they reside in different states? What if they haven't smoked for a week now? What exactly is "income" - should it be based on 2012 tax return 2013 return, 2014 return, personal estimate for next year? What if your estimate just happens to maximize your subsidy, but turns out to be, uh, inaccurate? How many choices should you have for plans (four?), for providers (depends on state and county?), etc, etc. etc. Now, these personal details turn out to be very important. I looked at my case, and depending on exactly how these details play out I'm looking at anywhere from $0/bronze $25/silver (monthly premiums) up to $7000 a year! Sure, I can game the calculators all I want - but when it comes down to actually signing up for a policy, the calculater estimate is free advice and worth no more than that. Right now, after using these calculators, I have NO CLUE what I might be looking at - certainly not enough to do better than flip a coin on whether I should sign up or pay the penalty. Now, should insurance be that complicated? When I head on down to my local insurance agent, IS it that complicated? Well, uh, turns out, yes it is. It takes hours of one-on-one personal time to go through all of the various add-ons, options, discounts, packages, deductibles, maximums, etc. for PCP, specialist, hospitalization (in and out patient), vision, dental, in and out of network, and on and on and on. And depending on how I mix and match, the premiums might vary by a factor of three! So OK, is it a terrible architectural mistake to collect every bit of this highly relevant information, before presenting ACTUAL COST NUMBERS for people to consider? Should people make buying decisions based on estimates that might be off by a factor of three? Imagine the publicity if people get to shop first, they pick a plan based on prices, THEN they go through the detailed personal information, and by golly the price triples! What I learned from using the calculators is that the most important factor for me is income. Not age, not any existing condition, not smoking, not even bronze or platinum plans. And the calculators told me that the subsidies are VERY sensitive to income. For a silver plan using Kaiser's calculator, estimating my income to be a mere $1000 difference, made a difference in the silver plan premiums from $25 to $75 a month! There is NO WAY I can estimate my next year's (2014) income to within $1000. It's entirely possible our income may place us below the poverty line - in a state that does not expand medicaid. And suddenly I have no insurance at all. Great. I would not enjoy architecting a system to handle all this. Without the details shopping is moot - pure guesswork. Without shopping first, why tell your computer all those details and hope those bozos don't sell it? And the system needs data from you, from the IRS, from the insurance companies, and who knows who all. And it all has to be written in six months, by 55 different contractors, to hit a moving spec, and all has to fit together. YOU do it.
ppthh. You bring up a video game and then dismiss a counter argument? EA had way more server load then COD because EAs Sim City required you to be logged into their servers to play the game. At least learn the game architecture before spouting off. Heck, if you want a better example, RockStars GTA5 knew what to expect, tried to compensate, and it still wasn't enough and are still working on bugs and issues in the system.
lol, Sims? COD set record sales and sold over 6 million copies in 24 hours. Its an online game. Do the math. At least learn sales figures and logic before trying to step into it.
Having password problems is basic stuff. This "professor" is obviously another left wingnut making excuses for Obama's failures. The website was poorly designed and built by people who don't know what they are doing.
Perhaps you should learn the difference between Server-side Mplayer and Client-side. COD Is generally smooth because most of the work is done on the Client side. The games are hosted by one of the players rather the on the COD central Servers. Sim City was Server-side in that the players had to connect to EA's servers which hosted the rooms and players and controlled most of the mplayer elements, which is why when you have 2 million try and connect at the same time you have issues. COD doesn't have those same issues. GTA5 is a much better example. Same development time (maybe even longer depending on when the US Govt started development of the system) as they started in 08. Same launch date (Oct 1) and Hey guess what, still having major connection issues. Oh yeah, and they had 11 MILLION first day sales, and are probably on track to 15-16 million by now.
I don't care about this nonsense . The government could only serve under 100k , a video game served multiple millions . Perhaps you can spot on about why they did , but who cares ? In the end the government couldn't do what a video game could !
Hilarious. I like how when I show your argument to be invalid then suddenly 'You don't care about this nonsense'. Anyway, while yes there are issues with the site and it's construction, to say that a video game could do it is a fallacy, as video games with better programmers, more time, and basically an entire system of business designed only for creating said video game, still has issues 2 weeks after implementation.
Oracle side is bulletproof. I can do it in six months with an autistic DBA, a Chinese engineer and an Indian code monkey... and I will do it for half the cost. I slammed out product. My job was solely securing large new contracts with live demos and endangered projects (weren't meeting milestones for rollout). I put together fully implemented betas for companies like Sprint, Toshiba, Fanny Mae... that tied every aspect of their businesses software needs by role into customizable portals with heavy business intelligence tools. Every piece of software every job used available through a single secure interface. I am fully aware of the challenges, however I had the benefit of working over a sound architecture... but complicated I do. I do serial acquirers. I tie thousands of unlinked businesses, software, and datasources together. I am the borg. I win awards and chicks dig me. I can also work 20 hours a day for months at a time. I am a code machine... but it is probably my humility that truly makes me perfect. I don't much trust the slant of the Forbes article either... however there is a lot of good info in there if you ignore the conclusion they arrive at. I don't think they ask that many questions outside of "do you have xxxx prexisting conditions, are you a smoker, how old are you etc". The hours of one on one aren't happening on the website.
But we are talking about a fundamental and systematic issue with the computer industry dropping the ball.
