Forgot your password?
typodupeerror
The Courts Government Medicine Oracle

Oracle Deflects Blame For Troubled Oregon Health Care Site 163

Posted by samzenpus
from the who's-to-blame dept.
itwbennett (1594911) writes "Oracle is gearing up for a fight with officials in Oregon over its role developing an expensive health insurance exchange website that still isn't fully operational. In a letter obtained by the Oregonian newspaper this week, Oracle co-president Safra Catz said that Oregon officials have provided the public with a 'false narrative' concerning who is to blame for Cover Oregon's woes. In the letter, Catz pointed out that Oregon's decision to act as their own systems integrator on the project, using Oracle consultants on a time-and-materials basis, was 'criticized frequently by many'. And as far as Oracle is concerned, 'Cover Oregon lacked the skills, knowledge or ability to be successful as the systems integrator on an undertaking of this scope and complexity,' she added."
This discussion has been archived. No new comments can be posted.

Oracle Deflects Blame For Troubled Oregon Health Care Site

Comments Filter:
  • by Anonymous Coward

    Oracle doing its usual crapware!

    PS:I'm a sysadmin of Oracle applications and they REALLY is shit!

    • by gtall (79522)

      My only experience with Oracle is with Oracle forms. I've done some extensive work with Java in the past, it is really breathtaking how badly Oracle can design interfaces using it. It is as though they absorbed all of Sun's expertise in interfaces and puked it back out with their own special distaste for useability.

    • that the software was not capable of being understood, or the documentation was skanty and faulty, that the Oracle consultants were dunderheads and got in the way, on and on.

      nice opportunity for a court of law and special masters to sort out. make it happen. King Ellison I is not always right.

  • by Greyfox (87712) on Thursday April 17, 2014 @10:17PM (#46784897) Homepage Journal
    It's that people think they can drop Oracle on top of a crappy design and that will somehow magically fix it. By the time people get done trying to use brute force, ignorance and massive amounts of IT resources, you may as well have Dbase III on your back end. Oracle might let you get away with a shitty design if your application didn't really need a database, but it's not going to help you that much if what you're trying to do is complicated enough to need one.
    • There's also nothing about running a health exchange that requires the use of an Oracle DB. The volume of data and transaction rate wouldn't stress any database. The secret sauce is in the application code, not the DB. They should have found a developer with a proven track record of delivering LAMP-style systems rather than pissing money away on a gold plated turd from a company with no grand history in developing web sites.

      • by Anonymous Coward

        It's not the web site that's the hard part, it's the back end integration work. Throw in HIPAA and HITECH and you've got a stew of poo and screw you.

      • by ShanghaiBill (739463) on Thursday April 17, 2014 @11:54PM (#46785289)

        Oregon's health website is a monstrosity. They budgeted way too much money, were way over ambitious, and involved way too many people. The opposite end of the spectrum is Kentucky, which budgeted the least amount of money, and was thus forced to implement a streamlined site with a small lean team. Kentucky's website was ready on Oct 1st, and has run since without a hitch.

        • by greenbird (859670)

          The opposite end of the spectrum is Kentucky, which budgeted the least amount of money, and was thus forced to implement a streamlined site with a small lean team.

          Amen to that. The larger the budget the greater the probability of failure. Software development has to be done by evolution not revolution. You start small and slowly add features in the form of modules with well defined interfaces. Any other approach most likely results in failure or if in any part successful, a monstrosity that is close to unsupportable.

          • by gtall (79522)

            Yes and no. If you aren't careful, you'll wind up with a dirty snowball which wasn't so much designed as it was congealed. Modularity will not save you. If your site won't scale, then you are going to need a new design.

            I agree with starting small though, as long as the tradeoffs are understood and management doesn't cough up a hairball when a redesign is necessary.

            • by greenbird (859670)

              Modularity will not save you. If your site won't scale, then you are going to need a new design.

              That's more a matter of a good approach and design when developing your modules. If you write crap modules you're right. If you're modules are designed and developed to be scalable it won't be an issue. In large projects modularity is almost always a better approach than monolithic. I have yet to come across a project where that wasn't the case. It's especially the case where scalability is an issue. It's much easier to develop and rag out scalability issues within each module then it is after you've develo

    • by Tablizer (95088)

      It's that people think they can drop Oracle on top of a crappy design and that will somehow magically fix it.

      That's probably what the sales-person told them.

      "Just drop your leaky little app into our Oratron Fix-O-Matic 5000 and out will pop perfect shiny reports and data! You have my trustworthy word!"

      However, seems they forgot to get it in writing.

    • Well, Oracle WAS dumb enough to take on the project in the first place!

  • by roc97007 (608802) on Thursday April 17, 2014 @10:17PM (#46784899) Journal

    ...in this case, they might have a point.

    • I'd assume that their legal team would be running around the company quietly busting skulls if they didn't.

      "Incur significant legal exposure during the course of fucking up a high-profile project for a government client" isn't one of those good strategies.

      Doing one or the other can actually be surprisingly lucrative; but both, less so.
      • by Kjella (173770)

        Time and material contracts basically means renting consultants by the hour, short of outright criminal behavior there's no promised time frames, deliverables or guarantees of functionality or quality. The upside is the lack of formalism, I've developed many reports on a T&M basis and basically if you want a filter here and a total there and to add one more column and add a traffic light here and a drill down there just say it and I'll keep working on it until you're happy. Heck, I've taken "requirement

        • Exactly, if Oracle were contracted on T&M then they were simply acting as an (expensive) body shop, ie they supply the bodies to the client, the client tells the body what to do. Basically Oracle takes $3 from the client, pays the body $1, and pocket's the difference. T&M on a project such as this is a cash cow for the vendor, it can only work in the client's favour if the client knows what they are doing.

          Disclaimer: Having been a body for other multinationals on similar projects, the $3:1 ratio
    • by plover (150551)

      Oracle consultants were in the midst of the mess, they saw the failings, they repeatedly reported to the state that the project was going off the rails, and yet they still managed to cash their paychecks.

      Had the consultants actually threatened them with "either you hire a professional to do the systems integration or we're off the job," and had they then removed themselves from the failing project, they'd be 100% blameless. But they didn't walk away, they just wrote some CYA memos and collected their money

      • by Zontar The Mindless (9002) <plasticfish@info.gmail@com> on Friday April 18, 2014 @06:46AM (#46786293)

        Oracle consultants were in the midst of the mess, they saw the failings, they repeatedly reported to the state that the project was going off the rails, and yet they still managed to cash their paychecks.

        Then the consultants were doing their jobs.

        Had the consultants actually threatened them with "either you hire a professional to do the systems integration or we're off the job," and had they then removed themselves from the failing project, they'd be 100% blameless. But they didn't walk away, they just wrote some CYA memos and collected their money.

        But it was not the consultants' job to do this. In fact, if they'd walked off the job as you advocate, they'd very likely be opening themselves up to a lawsuit for breach of contract.

        • But it was not the consultants' job to do this. In fact, if they'd walked off the job as you advocate, they'd very likely be opening themselves up to a lawsuit for breach of contract.

          Actually, I posted this a bit hastily. What I should have said was something more like, "If they'd walked off the job as you advocate, they'd very likely be opening up Oracle to a lawsuit for breach of contract, and themselves to getting fired by Oracle for doing so."

        • by TapeCutter (624760) on Friday April 18, 2014 @10:50AM (#46787397) Journal
          What the GP is suggesting is that Oracle the company (as opposed to the individual consultants) should have walked away from a taxpayer funded money pit but chose to continue "taking candy from a baby". Other's have walked away from similar disasters in the past in very public fashion, IBM walked away from a $800M project in NZ in the late 90's and Fujitsu walked away from a $1B project in the UK a few years ago, both claimed to be happy with the profit levels but were unwilling to continue because the government were unwilling/unable to follow their project management advice, making it impossible for them to deliver. Multinationals do not want to be seen as being unable to deliver a government contract, government work is their bread and butter and in politics reputations matter. Oracle didn't take the "high road" when their own consultants were predicting disaster, now they are getting public blowback from the client, which is why their PR department has fired up on this issue.

          OTOH Oracle (as their PR points out) were not managing the project they were on a time and materials contract, which most people in the industry would understand as meaning "we will give you what you ask for, but don't blame us if it is not what you want". The client obviously wasn't listening to the "don't blame us" part when they signed the contract.
    • ...in this case, they might have a point.

      The point being that if you give the ENTIRE project to Oracle instead of just part of it that it would all have been wonderful?

      • The point being that if they'd given the entire project to Oracle, then the entire project would in fact be Oracle's responsibility. But they didn't, and so it isn't.

  • by tlambert (566799) on Thursday April 17, 2014 @10:17PM (#46784901)

    Never bring politics... to an electronic documentation of timeline fight with a database company.

    • Oracle does this on nearly every project they work on. Same with IBM, TCS, Wipro, etc. All of those H1B visa abusing companies will never deliver on what they say and instead will fight the meaning of every word in the contract.
      • by ADRA (37398)

        I was an after-the-fact consultant to fix a botched IBM project, and I guess the penalties were so high that IBM bankrolled the re-implementation with on-shore workers to get the project finished. Whenever you go into bed with an SI, make sure you have an iron clad contract and many penalty clauses.

        PS: WiPro is almost entirely indian, so I don't know what H1B has to do with anything unless there's some contract requirement to have staff on-shore, which makes little sense when the purpose is to cut dev. cost

        • indeed, WiPro originally was a vegetable oil manufacturer, and stands for Western India Products Limited. I've dealt with them many times via supporting clients while working support at HP. Once you can get past the Indian's odd "your a stupid American" attitude (which usually I can't blame them!) they are mostly competent. In my situation, most of their "issues" and "failures" actually result from too-restricted access to various systems their trying to work on, crappy equipment their forced to use by t
          • Once you can get past the Indian's odd "your a stupid American" attitude

            Such as using your instead of you're or their instead they're, right?
  • So, we're seeing the fingers of the horn-hairs pointing at each other for a failure that their brown-nosed underlings caused on both sides. Don't blame the geeks, blame the suits.

  • by linuxguy (98493) on Thursday April 17, 2014 @11:34PM (#46785213) Homepage

    We were forced to sign this contract. On gun point actually. And then they said that we should take their money or they'll break our knee caps.

    You see, we are the victims here.

    Larry

    • by Anonymous Coward
      Ah yes, much lke the current healthcare law. Sign up or else.
  • by geekmux (1040042) on Thursday April 17, 2014 @11:46PM (#46785255)

    "...'Cover Oregon lacked the skills, knowledge or ability to be successful as the systems integrator on an undertaking of this scope and complexity,'

    Gee, that's funny. And here I thought I was in the majority in thinking that it is in fact Oracle who lacks the skills, knowledge, or ability to fix that piece-of-shit Frankenstein they want to label a working product.

    I suppose if you thought you were buying a perpetual bug and patch service, sure. They're fucking awesome at that. I might even be so bold as to say #1 in the industry.

    • by Kjella (173770) on Friday April 18, 2014 @01:04AM (#46785537) Homepage

      "...'Cover Oregon lacked the skills, knowledge or ability to be successful as the systems integrator on an undertaking of this scope and complexity,'

      Gee, that's funny. And here I thought I was in the majority in thinking that it is in fact Oracle who lacks the skills, knowledge, or ability to fix that piece-of-shit Frankenstein they want to label a working product.

      False dichotomy, it's not one or the other.

  • by Karmashock (2415832) on Friday April 18, 2014 @12:12AM (#46785329)

    All of them.

    In the case of all of them failing you have to look at the common denominator because ALL of them failed.

    Newsflash... Oracle was not involved in all of the exchanges.

    The central problem was that the rollout was rushed for political reasons.

    If it were slowed down then the republicans might have had more success killing it before implementation. Even now it might well die. So the democrats rushed the rollout.

    And this is the result.

    That is not Oracle's fault. We all have experience with projects that are rushed through planning to the point where they are unworkable.

    That's all this is... nothing more or less.

    • by Anonymous Coward

      All of them.

      In the case of all of them failing you have to look at the common denominator because ALL of them failed.

      Newsflash... Oracle was not involved in all of the exchanges.

      The central problem was that the rollout was rushed for political reasons.

      If it were slowed down then the republicans might have had more success killing it before implementation. Even now it might well die. So the democrats rushed the rollout.

      And this is the result.

      That is not Oracle's fault. We all have experience with projects that are rushed through planning to the point where they are unworkable.

      That's all this is... nothing more or less.

      Yeah, you're right. I don't know why so many of us jumped on Oracle here. I mean c'mon, of all the products I've ever worked with in my data center, Oracle always installed just beautifully and ran flawlessly. And cheap too! Starting at only a few million regardless of company size. And even when we didn't even budget for it, Oracle was nice enough to hand us more invoices and bills as time went on, because you know, the occasional patch was needed, and we did run across some tiny bugs and annoyances f

      • I have lots of experience with Oracle and you had better believe that the state of Oregon did as well.

        The state signed the contracts with Oracle and fucked up.

        JUST LIKE EVERY SINGLE OTHER EXCHANGE IN THE ENTIRE COUNTRY.

        At this point you're dealing with very simple logic.

        Solve for X and then agree with me.

    • by Tablizer (95088)

      was that the rollout was rushed for political reasons. If it were slowed down then the republicans might have had more success killing it before implementation.

      That might be "political" reasons, but it's also practical reasons. If you want to get certain things done, you have to race against competition trying to kill it.

      For about 100 years various presidents and lawmaking groups have tried to enact a medical insurance program of some sort, only to see it smashed down. With that kind of record you know you

      • Its a concern that is not Oracle's fault or responsibility.

        You tell me you want a house built in two days... Fine... but if you expect there to not be serious problems and unforeseen issues as a result of rushing through the planning phase and then constantly changing the mission premeters throughout development... then you're being unreasonable.

        Projects that are effective tend to have two things going for them.

        1. A lengthy planning phase where everything is spelled out in detail.

        2. A no nonsense implementa

        • by whoever57 (658626)

          You tell me you want a house built in two days... Fine... but if you expect there to not be serious problems and unforeseen issues as a result of rushing through the planning phase and then constantly changing the mission premeters throughout development... then you're being unreasonable.

          Newsflash .. customers are often unreasonable.

          In your scenario, if the contractor signs a contract to build the house in two days, then the contractor is liable if the house is not ready in two days. As the contractor,

          • I think you'll find that in any contract, there is a notion of specificity.

            If you don't specify what "done" looks like in the contract then the contractor can argue that "his version of done" satisfies the terms of the contract.

            Furthermore, they didn't hire Oracle on to do it. they rather did the integration in house using Oracle contractors to answer questions and do parts of the project.

            It could be argued by Oracle that the the failure was in the integration and not in anything they provided.

            As to the cos

    • by Cyberax (705495) on Friday April 18, 2014 @01:05AM (#46785541)
      Actually, no. KyNect worked without downtime on its frontend, however its backend was not very stable - as it had to interact with the federal exchange.
    • You're close ... it's the "business" itself that is unworkable in this case . That's the common denominator.

      "Do a bad idea ... but on a computer!" does not stop it from being a bad idea. It just digitizes the implementation of the damage.

      • if by business you mean the ACA, then I think most would have to admit it was very poorly done.

        If you mean insurance itself... that worked for many years prior and continues to work for home insurance, car insurance, etc without much of a problem.

        The issue comes largely from politicians writing rules for a very complicated issue that they don't understand and lack the proper respect to handle slowly and rationally.

        Anything that takes more then 10 seconds to understand is either not addressed at all or half

  • by Vlijmen Fileer (120268) on Friday April 18, 2014 @12:18AM (#46785353)

    It's the customers' fault. EVERYBODY in the IT business already knows that Oracle invariably gives you:
    - Bizarely high price
    - Incomplete project result
    - Project delays
    - Low quality
    - Extreme vendor lock in
    E.v.e.r.y s.i.n.g.l.e p.r.o.j.e.c.t they do.
    I'm not sure whther to cry or laugh at this. Just don't go with Oracle, every sane IT professional knows that.

  • Oracle should accept some losses and quietly make an amiable-as-possible exit. Why air dirty laundry about clients? Even if the State is partly to blame, being a loud asshole makes you less likely to get future gov't contracts.

    • Oracle should accept some losses and quietly make an amiable-as-possible exit. Why air dirty laundry about clients? Even if the State is partly to blame, being a loud asshole makes you less likely to get future gov't contracts.

      You have no idea the level of arrogance that exists at Oracle...

    • by munch117 (214551)

      You're assuming they have a good reputation to spoil. They don't. They get contracts because there's a superstition that the way to build a database-backed system that scales is to buy the most expensive database you can find.

  • The project was a huge success. It separated Oregon from their money, right?
  • It does sound as though the primary blame has to be put at the Oregon's officials since Oregon was the lead on the project. The lead is always 100% responsible for the project, after the project failed they are trying to say "ohh, not my fault"... If the project was off the rails early on, they should have seen it -- regardless of communications and adjusted (and if Oracle was not doing it's job - fired them). Obviously Oregon wants to have a scapegoat, but apparently forgot to pay them for that service.
  • by drolli (522659) on Friday April 18, 2014 @07:06AM (#46786355) Journal

    I am working as a consultant.

    My good advice to every customer is: dont buy consultant work as time and material. Buying as time and material puts the wrong incentives to everybody:

    -Your own people will feel that they still can just use them as normal workers and keep all decisions (and thus responsibility) to themself

    -The consultants dont care, since just doing what your own people tell them without thinking is what gets their monthly timesheets signed. If something goes wrong they can even sell more hours, not less

    -The consulting company does not care (and rigthly so since that was not what you asked for) and will send you inexperiences junior consultants wherever possible.

    -Coding quality has to be reviewd by your own people (or just accepted as it is)

    -Your own people are usually vastly inferior at project management in comparison to the average senior consultant - in a non T&M contract the usual situation is that you get the things done in time or you will loose money.

  • by kbdd (823155)
    Let's summarize: Oracle took money to perform a job. They disagreed with the way it was handled. Nevertheless they stayed in and kept collecting money. Now they say they had noting to do with it?
  • I have been working on a large data project for another state - this state has outsourced everything to 3 or 4 large companies. That itself is not so bad, but the state doesn't have anyone left to make decisions. Instead it is all left up to the vendors. It is difficult for vendors, even when trying to do the right thing, to know what the business (state) needs or wants for some things.

    Trying to implement proper security controls and create separation of duties when everything is outsourced is hard to do. E

  • by Rambo Tribble (1273454) on Friday April 18, 2014 @11:44AM (#46787855)
    ... that the quality assurance contractor for the project, Maximus, had this to say, "Oracle's performance is clearly lacking. Their inability to adhere to industry standards and professional software and project management tenets warrants further review."
    • by Tharkkun (2605613)
      I bet they also tried to install it on under powered hardware. Just like the companies try to install enterprise class software on a 3 TB USB drive because they don't want to pay the cost of a NAS/SAN.
  • I've worked on several fiascoes that the State of Oregon has tried to build over the last 30 years and they always end up as fiascoes. Motor vehicles, child support system, a consolidated database of Oregon state welfare recipients. They all failed for one reason. Oregon State workers are a bunch of lazy, incompetent, featherbedding incompetents.

    They take off on sick leave for days at a a time without even notifying their bosses. Are promoted based on minority or gender status instead of competence, si

  • ...but don't dismiss Oracle's arguments without examination. It may be that Oracle is trying to shift the blame, or their part of it. But, that is also the default position for governmental organizations when shit hits the fan. Is it a coincidence that almost all of the exchanges that failed or are in deep trouble chose government as the SI? Regardless how you view government, it must be conceded that there are few incentives for efficiency or success. Sadly, this also extends to contracts let by said organ

"How do I love thee? My accumulator overflows."

Working...