Oregon Withholding $25.6M From Oracle Over Health Website Woes
itwbennett writes "Oregon is holding back $25.6 million in payments from Oracle (out of some $69.5 million Oracle claims it is owed) over work the vendor did on the state's troubled health care exchange website. The site was supposed to go live on Oct. 1 but its launch has been marred by a slew of bugs and it is not yet fully functional. This week, Cover Oregon said it had reached an agreement with Oracle laying out 'an orderly transition of technology development services, and protects current and future Cover Oregon enrollees,' according to a statement. Oregon officials reached the deal with Oracle after the company reportedly threatened to pull all of its workers off the project and essentially walk away."
Too many companies deliver sub-standard software without any risk at all, especially in big projects.
Mistakes do happen, but underbidding is too common.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
...at least had great propaga...I mean, advertisements.
Must be nice to be able to fail at a project such that they owed you $69 million, but you don't actually have to make it work.
Perhaps states should make a rule stating that large projects must be broken up into deliverables of $1 million increments.
Engineering and the Ultimate
Just one more thing though.
Oracle should pull all of its workers off the project and walk away after giving back all money already paid.
If you don't deliver what you've been told to deliver, you shouldn't get paid.
I don't have any personal experience with Oracle the company. But I've spoken to a half-dozen or so of their clients, and not one of them has ever had a successful completion of a project, and they've all gone over budget. Purely anecdotal evidence, I know.
I'd be interested to hear if someone has had a good experience working with Oracle...? But if the overwhelming consensus is negative, how do they continue to gain new clients?
Even without the $25m owed in the contract, Oracle is probably still profitable on the deal.
I bet they maintain 60-70% margins... and that's on the services side...
It's an honest question. I am a programmer of embedded systems and microcontrollers, my expertise is at the other end of the computing spectrum.
As much as I like to blame Oracle, the state may have added serious requirements at the last minute that complicated everything. These articles doesn't say anything about it. Same seems to go for all the troubled exchanges - so what's the problem?
Is there anyone on here with some insight?
As a customer of Oracles, and having these very same products including Sieble... all I can say is "You should have asked me first"
This is exactly what we're going through. They sold us a suit of "integrated software products" that were in no-way integrated or even related. They charged us to configure the software, then when the software didn't work, told us it was configured wrong. Then when it was time for a new contract tried to exempt themselves from liability for "Configuration changes" and threatened to not renew and not fix the issue unless we did sign. (we didn't and almost ended up in court)
Then when they were making changes their support teams would log into their software through various back doors and make changes without notifying us, leaving a trail in the audit log with "NULL" in the place where the user account that made the change was supposed to be logged. They remotely modify white lists into the application suite without permission despite specific contractual agreements that they would not. We've got Oracle Employees whitelisting their home DSL accounts and logging in at random at all times of the day.
Oracle is the worst Vendor I've ever worked with. They are incompetent, malicious, vindictive and will outright lie, con and steal from their customers. They literally deprecated our ODBC connection to a SASS once because we weren't going to renew our contract and they wanted to charge us to move the data off their systems. Luckily we had planned for such a thing and already had a replication database in-house. God I hate Oracle.
This web site is a front end that is supposed to federate all the suppliers of health care insurances. Since there is no clear and complete standard interface, most of the work goes into making "glue code" to get all the insurers hooked up to the system. The visitor has to be able to experience all of this real time. Try interfacing with over 50 slightly different but very similar complex computer systems that each have their unique protocol. Writing good requirements documents is an absolute nightmare, let alone unit tests, full flow tests, regression tests, security tests and whatnot. Once you have that, you might get to writing code and discover that your performance will suck horribly due to all the special cases you have to account for....
I was promised a flying car. Where is my flying car?
The site "worked"...it was built to specifications. It's **the purpose of the site** as directed by non-tech health industry people that was the failure.
If you looked at the original site, it was essentially a guide to signing up for ****private insurance**** like Kaiser Nazi-mente, which run the Oregon Health Authority
Those private companies wrote the requirements for what the site would do!
Look at this interview with the original IT manager: http://www.oregonlive.com/heal...
from the above link:
Thank you Dave Raggett
Here's an interview where the original IT Manager of the project (who raised red flags & got fired)...they even told her to attribute her firing to a death in the family to cover their tracks!
http://www.oregonlive.com/heal...
The site wasn't meant to list & rate available policies...any kind of Yelp clone could do that, probably with alot of off-the-shelf libraries. The site was meant to be a revenue channel for the private insurance industry.
Thank you Dave Raggett