Birmingham's $125M 'Oracle Disaster' Blamed on Poor IT Project Management (computerweekly.com) 117
It was "a catastrophic IT failure," writes Computer Weekly. It was nearly two years ago that Birmingham City Council, the largest local authority in Europe, "declared itself in financial distress" — effectively declaring bankruptcy — after the costs on an Oracle project costs ballooned from $25 million to around $125.5 million.
But Computer Weekly's investigation finds signs that the program board and its manager wanted to go live in April of 2022 "regardless of the state of the build, the level of testing undertaken and challenges faced by those working on the programme." One manager's notes "reveal concerns that the program manager and steering committee could not be swayed, which meant the system went live despite having known flaws." Computer Weekly has seen notes from a manager at BCC highlighting a number of discrepancies in the Birmingham City Council report to cabinet published in June 2023, 14 months after the Oracle system went into production. The report stated that some critical elements of the Oracle system were not functioning adequately, impacting day-to-day operations. The manager's comments reveal that this flaw in the implementation of the Oracle software was known before the system went live in April 2022... An insider at Birmingham City Council who has been closely involved in the project told Computer Weekly it went live "despite all the warnings telling them it wouldn't work"....
Since going live, the Oracle system effectively scrambled financial data, which meant the council had no clear picture of its overall finances. The insider said that by January 2023, Birmingham City Council could not produce an accurate account of its spending and budget for the next financial year: "There's no way that we could do our year-end accounts because the system didn't work."
A June 2023 report to cabinet "stated that due to issues with the council's bank reconciliation system, a significant number of transactions had to be manually allocated to accounts rather than automatically via the Oracle system," according to the article. But Computer Weekly has seen a 2019 presentation slide deck showing the council was already aware that Oracle's out-of-the-box bank reconciliation system "did not handle mixed debtor/non-debtor bank files. The workaround suggested was either a lot of manual intervention or a platform as a service (PaaS) offering from Evosys, the Oracle implementation partner contracted by BCC to build the new IT system."
The article ultimately concludes that "project management failures over a number of years contributed to the IT failure."
But Computer Weekly's investigation finds signs that the program board and its manager wanted to go live in April of 2022 "regardless of the state of the build, the level of testing undertaken and challenges faced by those working on the programme." One manager's notes "reveal concerns that the program manager and steering committee could not be swayed, which meant the system went live despite having known flaws." Computer Weekly has seen notes from a manager at BCC highlighting a number of discrepancies in the Birmingham City Council report to cabinet published in June 2023, 14 months after the Oracle system went into production. The report stated that some critical elements of the Oracle system were not functioning adequately, impacting day-to-day operations. The manager's comments reveal that this flaw in the implementation of the Oracle software was known before the system went live in April 2022... An insider at Birmingham City Council who has been closely involved in the project told Computer Weekly it went live "despite all the warnings telling them it wouldn't work"....
Since going live, the Oracle system effectively scrambled financial data, which meant the council had no clear picture of its overall finances. The insider said that by January 2023, Birmingham City Council could not produce an accurate account of its spending and budget for the next financial year: "There's no way that we could do our year-end accounts because the system didn't work."
A June 2023 report to cabinet "stated that due to issues with the council's bank reconciliation system, a significant number of transactions had to be manually allocated to accounts rather than automatically via the Oracle system," according to the article. But Computer Weekly has seen a 2019 presentation slide deck showing the council was already aware that Oracle's out-of-the-box bank reconciliation system "did not handle mixed debtor/non-debtor bank files. The workaround suggested was either a lot of manual intervention or a platform as a service (PaaS) offering from Evosys, the Oracle implementation partner contracted by BCC to build the new IT system."
The article ultimately concludes that "project management failures over a number of years contributed to the IT failure."
How? (Score:5, Interesting)
the council was already aware that Oracle's out-of-the-box bank reconciliation system "did not handle mixed debtor/non-debtor bank files.
How is this possible? That seems like one of those basic functionality items systems in the 21st century should have by default. Is Oracle truly that pathetic?
Re: (Score:3)
Would not surprise me one bit.
Re:How? (Score:5, Insightful)
With any project of this scale, where the system will be the new financial transaction system and source for all audits, you would want to run the old system and the new system in parallel for six months to a year before transitioning fully to the new system. Even testing the previous year's transactions into the new system would have shown unanticipated categories or transaction types that it appears weren't available in the new system.
If the previous system required five years of tuning before it could adequately record and account for each type of financial transaction, to think the new system would just magically have exactly the same features on day one was a serious oversight.
Anyone who has tried to convert a Lotus 123 or Quattro Pro spreadsheet into Excel knows the formulas and references are the most difficult thing to translate into a new spreadsheet. To just assume every financial transaction type, relationship, graph, budget, report, error checking, and audit document was correct without running real data through the system was irresponsible. The project manager on this project really messed up.
Re:How? (Score:4, Insightful)
Sure. And that is why you want a first-rate engineer with on-target experience to be the final decision maker. Not some no-clue politicians that do not understand technology (or reality) one bit.
Re: (Score:2)
Spreadsheets seem designed to particularly make it incredibly difficult to adapt. All based on the original Visicalc design of column+row designation of variables. Later spreadsheets did add more general purpose variables, but it's a feature not used as much and not standardized. Also the lack of comments on what that mysterious number in $X73 means, or what other cells use it, etc.
Re: (Score:2)
Also, systems of this scale NEED customizations.
It's unclear if a trial was conducted using at least a month's worth of transactions to evaluate whether the financial reports meet expectations.
However, a trial period of six months to a year would be more appropriate.
I doubt Oracle would allow free access to their system for that long.
Re:How? (Score:5, Informative)
I have worked with Oracle financial modules for the past 20 years, the HR/Payroll modules for an additional five years and the development and reporting tools since the mid-90's
Oracle ERP "out of the box" is pretty generic, and Oracle sales reps are more interested in closing the deal, rather than delivering a completely functional system
Oracle Consulting will bill you hundreds of dollars per consultant hour to provide the finished system, and take years to do so
The key to ANY functional Oracle ERP system is obtaining a competent 3rd party implementer that understands the business environment, can communicate with the people who will be using the system AND can deliver the configurations, customizations and reporting that meeting the clients needs
Beyond that, it is important to identify the key business people who can define the requirements, write the test cases and perform them as part of a comprehensive acceptance process
When these things are present, you get a functional system that will require constant support and modifications on an annual basis as new releases are pushed out and key security patches installed
Ultimately, most organizations realize the need to hire their own Oracle support, development and reporting staff to maintain the system, but they will likely always require Oracle Cloud support at some level to remain functional
What happened in this story is really a worst case of incomplete scope definition, poor implementation and a total lack of acceptance testing
IMO the Project Manager (or the whole Project Management Office) of this debacle should be kicked to the curb for letting it go to production
Re: How? (Score:3, Insightful)
Mod parent up.
This smacks of experience-derived wisdom.
Re: (Score:3)
Also, "project managers" are usually the bottom of the management totem pole. Many aren't too experienced either. So it's not too unusual to see some junior project manager saddled with a task completely out of their depth and you can see the incipient stress induced madness in their eyes.
(of course there are also competent project managers, even great ones, just not enough of them to cover all the projects)
Re:How? (Score:5, Interesting)
I think both of them got what they deserved, but I'm more inclined to think that financial culpibility really rests with the board. Sure, you can fire the project manager for being a git, but if you're looking for someone to sue, the project manager's probably not the best target.
Every failed project I've ever seen (not just IT) has had an uninvolved and overdemanding project sponsor. Good project managers resign from those situations before it gets too bad. Stupid project managers hang in there and go down with the ship.
Re: How? (Score:2)
The project manager should at least partly see themselves as the last line of defence between the limited resources of the dev team and the unlimited wants of the project sponsors. The PM should provide realistic expectation management to the board and also push back against unachievable demands or ideas that are just plain dumb.
The board for their part ought to realise that they have (presumably) hired trustworthy people to work under them, and trust their advice.
When this ecosystem is broken at any level,
Re:How? (Score:5, Insightful)
This is how corporations work. The project manager exists to be fired if the project goes poorly, in order to protect the executives and/or the board. They are ablative.
Re: (Score:3)
Thank you. I will do my level best to remember that.
Re: (Score:2)
Board Chair: What a disaster that was!
Board Vice-Chair: Right, right. What a nincompoop we put in charge.
Board Chair: We'll not do that again.
Board General Member: Who's nincompoop was that? I didn't hire him.
Board Emeritus: Me neither.
Board Vice-Chair: Well someone had to have hired him?
Board Chair: He wasn't a rogue element was he?
Board General Member: Oh probably he was. That explains it. We'll take care not to do that again in the future!
Board Vice-Chair: So what's next on the agenda?
Board Chair: W
Re: (Score:2, Informative)
I've been involved in one way or another with a couple dozen database projects, only three of them were Oracle systems. All three were multimillion dollar multi-year fiascos. One was later replaced with SQL Server and the other with Informix, the third was only partially functional and quietly abandoned.
Re:How? (Score:5, Insightful)
I have worked with Oracle financial modules for the past 20 years, the HR/Payroll modules for an additional five years and the development and reporting tools since the mid-90's
Oracle ERP "out of the box" is pretty generic, and Oracle sales reps are more interested in closing the deal, rather than delivering a completely functional system
Oracle Consulting will bill you hundreds of dollars per consultant hour to provide the finished system, and take years to do so
Completely believable, but why is this necessary? Birmingham City Council isn't unique. It's doing the same job, under the same legislative regime, as every other council in England. Why is a bespoke solution even being developed? Surely it would make more sense to have one centralised, heavily-tested solution to roll out across the country?
Re: (Score:2)
Completely believable, but why is this necessary? Birmingham City Council isn't unique. It's doing the same job, under the same legislative regime, as every other council in England. Why is a bespoke solution even being developed?
Ultimately, it is about money. You could write something to replace what Oracle does... so why don't you? (Do you begin to understand?)
Re: (Score:3)
Because Oracle does bespoke systems. So do many other of those types of companies. They make a LOT of money being bespoke, and selling you their team of professional services professionals to fashion your fine fitting suit of data. Seriously, it's a big profit engine when you don't have a one-size-fits-all solution.
I used to work on business software once: worst job, worst pay, most bruises on the soul from being crushed, etc. Biggest profit came in from a mix of professional services and extended suppo
Re: (Score:2)
Re: (Score:2)
I have worked with Oracle financial modules for the past 20 years, the HR/Payroll modules for an additional five years and the development and reporting tools since the mid-90's
Oracle ERP "out of the box" is pretty generic, and Oracle sales reps are more interested in closing the deal, rather than delivering a completely functional system
Oracle Consulting will bill you hundreds of dollars per consultant hour to provide the finished system, and take years to do so
Completely believable, but why is this necessary? Birmingham City Council isn't unique. It's doing the same job, under the same legislative regime, as every other council in England. Why is a bespoke solution even being developed? Surely it would make more sense to have one centralised, heavily-tested solution to roll out across the country?
What's even worse is that the sales reps will deliberately lie, offer a lower price than competitors and then claim extra work is needed once the implementation begins on government contracts under the belief that the government will pay until completion (which, to be fair has paid dividends up until recently). Governments used to be reluctant to sue bad suppliers so a government contract was considered license to bill, the biggest problem was lying longer and harder than your competitions (IBM, HP, Cogniza
Re: (Score:2)
Former executive in charge of a large (25,000+ user) Oracle ERP implementation for a public-sector entity. Inherited the system four years after go-live with no reconciliation between the ledger and the bank. I've done the signing paper checks to schools while looking really sorry on the news thing. A large part of the 'bespoke' is self-inflicted. Yes, everyone uses GAAP, or some such. But, when you have a ton of semi-autonomous agencies in the cabinet each defining custom processes, their own componen
Re:How? (Score:4, Funny)
Re: (Score:2)
Re:How? (Score:4, Interesting)
Managers may have the wrong incentives. Ie, if they get a bonus for getting it out on time, but no reduction in bonus for having a product that doesn't work right or that goes overbudget, then the manager will push very strongly for making the deadline at any cost.
Also, if you think about it, almost all problems are due to management in some way. Even a brilliant manager with a terrible team would be the fault of even higher management who insist that the team continue with a a project doomed to fail. Even natural disasters could be the fault of management not taking appropriate risk analysis, having redundancy, allowing for extra unanticipated time in the schedule, etc.
This particular case is an obvious management failure. However, if the managers got paid a lot of money and were able to continue on in other jobs with high pay later, those managers may consider this SNAFU one of their great successes.
Re:How? (Score:5, Insightful)
Anyways, the real issue here is the city council went live despite being told it wouldn't work. It's a classic "fuck around and find out" moment. These are not failures unique to any large software vendor, these are failures of project management as the article stated.
Re:How? (Score:5, Insightful)
you bought the business application from SAP, the database from Oracle, and let each focus on their core competencies.
There's a line in that 90s movie The Firm where Gene Hackman asks Tom Cruise "What is the difference between tax avoidance and tax evasion? Answer: whatever the IRS says it is." Oracle's licenses work much the same way.
Databases are not Oracle's core competency. Impenetrable & extortionate licensing regimes are their core competency.
Re:How? (Score:4, Interesting)
Occasionally I get email spam from Oracle, and they tend to be pretty aggressive. Every time I get one I'm so tempted to say "No thanks, I don't want to get sued." We don't have any current business contracts with them so they can't force us into a software audit, which I'd wager is something their salespeople aren't used to dealing with. The irony that it's highly illegal for them to send us marketing emails at all is not lost on me. Though they once had a salesperson contact us because they saw traffic related to one of their products coming from a domain that they assumed was ours, and she wanted to get on a call with us to "explain" how its licensing works. And yeah, it came from their sales team, not their legal team. Think about that. Turns out this domain wasn't even ours, but their "buy now or get sued" subtext was obvious regardless without even talking to them, which we didn't even do -- we just handed it to our own legal counsel and never heard any more of it.
Does make you wonder just what kind of a sleazeball would agree to join Oracle's sales team. They're even worse than the worst used car salesmen. Threatening to sue is, it turns out, directly baked into their sales model, which they call “Audit, Bargain, Close”.
Re: (Score:2)
Re: (Score:3)
Databases are not Oracle's core competency. Impenetrable & extortionate licensing regimes are their core competency.
I don't disagree, but it is worth noting that of all the software that Oracle has and peddles, their RDBMS is far and away the best of it. IME it's almost as good as they say it is, or at least it used to be. It's been quite some time since I touched it, but last I did, it had good performance and reliability and didn't fall on its face when stuffed with large data.
Perhaps that's why technical people can get suckered into working with them. They don't do everything wrong.
Re: (Score:2)
Yeah, agreed, sarcasm aside the Oracle database *is* top tier.
Re: (Score:2)
I don't expect this to still be the case. I wouldn't be surprised if nothing has changed!
Re: How? (Score:2)
Re: (Score:3)
And were they wrong? In this sort of situation it's usually a choice of which vendor you want to fleece you.
Re: (Score:3)
It is the same with any situation of picking a vendor; if the price looks too good, buyer beware. Incidentally I used to work for a company that made very specific customizations to this type of Oracle system. We did not full implementations but worked with other companies that did that. On one project the implementers wanted to do our part of the contract but the customer did not award them that contract. As with many of these jobs, the implementers sent their least experienced people to keep the bid low;
Re: (Score:2)
Re: (Score:2, Insightful)
mixed debtor/non-debtor bank files.
I have no idea what that even means.
But I do know the word "Oracle".
It means the project was extortionately expensive, was bound to cost more than quotes, and designed to lock the city in to decades of dependence.
Re: (Score:2)
People who owe you money. People you owe money to.
The system (apparently) could not handle an account that was *both*.
Or (possibly) transactions involving 2+ parties where the other side was a mix of debtor/non-debtor accounts.
Not every relationship allows immediate offsetting. And you can't use a flag, but day-timers (how many days before offsetting is allowed, by them, by us: go by the later date)
Re:How? (Score:5, Informative)
Is Oracle truly that pathetic?
Yes, Oracle is terrible.
Oracle is the worst company I've ever had to deal with. I'd trust Microsoft before I'd trust Oracle.
Horror stories abound. I am flabbergasted that people still sign contracts with them.
I worked on one project where we provided 20 servers to Oracle. They were supposed to configure them and set up custom software. Months went by with no communication. It turned out they had wiped the servers and assigned them to other tasks. They basically just stole our equipment.
During the ACA rollout, several states contracted with Oracle to build their websites. They were the most expensive projects, and all of them failed on day one. The worst was Oregon. They spent 100 times as much as Kentucky, yet Kentucky's site, built by state employees, worked flawlessly, while Oregon's site, built by Oracle, was a debacle.
I have never, not once, heard anyone say they were happy with Oracle.
Feel free to ask me for more horror stories. I love bad-mouthing Oracle. It's my way of getting even for all the times they screwed me.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
the council was already aware that Oracle's out-of-the-box bank reconciliation system "did not handle mixed debtor/non-debtor bank files.
How is this possible? That seems like one of those basic functionality items systems in the 21st century should have by default. Is Oracle truly that pathetic?
Our place had a 2 month Oracle implementation that took over 5 years. Yah, Oracle ain't all that.
Re: (Score:2)
Is Oracle truly that pathetic?
Oracle is 100% as pathetic as your statement of requirements. Of course their system can handle these things. But it won't be delivered to you to do so if you don't write it in the specification.
Re: (Score:3)
Practically all Oracle based systems are custom. They aren't off-the-shelf software. You're paying for a large team of professional services to write the custom software for you. Any large customer has unique and complicated requirements. Granted, many solutions can be adapted to different partners without too much fuss, but given the scope of these projects even tiny changes become giant hurdles. And not just Oracle, but things like SAP/R3, or anything database oriented.
Now you've selected your lowest
And that is why we need software liability (Score:3)
In this case for the morons that decided to switch this unfinished system on.
Re: (Score:3)
Agreed, I for one, am gobsmacked by the apparent lack of acceptance testing
Re: (Score:2)
It happens when you give non-engineers the power to make engineering decisions. These people just do not have what it takes and do not understand that technological reality is absolutely mercyless. And that is why they should never be allowed to make such decisions.
lack of understanding (Score:4, Insightful)
Certainly Oracle is a platform that comes with high costs but these projects that fail like that are not about Oracle. You can blame Oracle all you like, it could have been SAP or IBM or Microsoft or RedHat or anything, if your planning is shit, if your scope is undefined and not fixed, if there is no ownership of the project, if nobody is personally responsible for the costs (and surprise, governments are never personally responsible for the costs, a private enterprise would have stopped paying much earlier than any government) then it doesn't matter, your project will be a failure. Failure to plan is a plan for failure. Access to unrestricted public funds is ripe for corruption. There is no ownership, no accountability, no personal responsibility, nothing.
If you want a project of this size to be successful you have to first define what and who will set limits, restrictions on spending, on scope, where the boundaries are, how to control expenses, that's before doing anything else. You cannot have multiple owners, no personal responsibility. At least if there was someone, whose pay and potential bonuses were linked to the *success* of the project and any failure would reflect on them personally, losing their shirt and maybe freedom, then you could start a project and hope that there will be some success.
Oracle is like walking drunk in E St Louis at 2AM (Score:5, Informative)
Certainly Oracle is a platform that comes with high costs but these projects that fail like that are not about Oracle. You can blame Oracle all you like, it could have been SAP or IBM or Microsoft or RedHat or anything, if your planning is shit, if your scope is undefined and not fixed, if there is no ownership of the project, if nobody is personally responsible for the costs (and surprise, governments are never personally responsible for the costs, a private enterprise would have stopped paying much earlier than any government) then it doesn't matter, your project will be a failure. Failure to plan is a plan for failure. Access to unrestricted public funds is ripe for corruption. There is no ownership, no accountability, no personal responsibility, nothing.
By your logic, one can get mugged anywhere at any time...which is technically true, but I'd still strongly advise against walking alone and drunk around East St Louis at 2AM or many parts of the South Side of Chicago or bad neighborhoods in any country/state/region. Similarly, any company can produce a quagmire, but Oracle is not well regarded. My company uses it both their RDBMS, which well...for those who don't use it...it really really really really really sucks in many very very very stupid ways....and Oracle knows this...and doesn't care....and neither does your boss...because some director at your company has a great relationship with their local Oracle rep. But OK, that's a product from the 80s that has evolved little since the 80s and has all sorts of STUPID STUPID STUPID 80s era restrictions and few innovations, despite huge success and a shocking price tag....but with RDBMS, you do have to be fair and consider that it is a legacy product with many customers from the 80s, so some innovations and optimizations, they cannot do too easily. And it's still the most popular DB out there and most of us professionals have found ways of dealing with and making it less painful...sadly, by moving logic to middleware tiers, quite inappropriately.
But the Oracle financial software....HOLY SHIT IT'S BAD...Oracle Financial software has some of the worst UIs I've ever seen and is constantly crashing and just is BAAAADDDDD....like early 2000s web developers fucking around with Adobe Flex, but doing a terrible job, bad. It's not even that old, so there's no excuse. It's a horror show. I know ERP software is not glamorous...but this is shockingly bad. OK, it's ugly as sin...the people that use it are not the ones making decisions, so it's not too shocking, but the unreliability and the number of e-mails I get from our IT dept about it going down are insane.
Oracle was in a position to be the dominant server-side software vendor. RDBMS is a money-printing machine and all Larry Ellison had to do was make good software to leverage it...but he failed and he failed hard...he is a corrupt piece of human garbage. If he made MS-grade software...not GREAT software...just stuff about as good as SharePoint...he'd be ruling the software world right now...but he'd rather collect checks and do the bare minimum than do anything for his customers. Enterprise software doesn't HAVE to suck...it just usually does, especially with Oracle. It's like if Tony Soprano's least competent friends ran a software company....that's Oracle...predatory, ineffective, lazy, low-effort, just shitty.
So if a city chooses Oracle and gets fleeced for a system that was delivered poorly...for those of us in the industry....yeah, it's about like hearing some guy got mugged because he was stumbling around in the worst neighborhood in the city drunk off his ass and alone. It SHOULDN'T happen, but we all know it does and we just tell those we care about "don't let it happen to you."
Re: (Score:2)
I'm not going to try and debate everything you say, but the days of server side table-based triggers and stored procedures are being taken over by mid-tier BPEL processing in the current FUSION/Cloud versions [oracle.com], and much of the clunky interfaces that you see are demanded by the client base who suffered through prior 9,10 and 11.x versions of the software
If you hate stored procs, you're ignorant of DBs (Score:2)
I'm not going to try and debate everything you say, but the days of server side table-based triggers and stored procedures are being taken over by mid-tier BPEL processing in the current FUSION/Cloud versions [oracle.com], and much of the clunky interfaces that you see are demanded by the client base who suffered through prior 9,10 and 11.x versions of the software
Anyone who complains about stored procedures doesn't understand them. They're not the problem. Arbitrary stupid limits are the most painful for me in RDBMS. PL/SQL is a HORRIBLE language, but so were many languages back then. Once you learn it, stored procedures are very powerful. I have made a good part of my career porting shitty Hibernate code that takes minutes to run into stored procedures that take less than a second to run. Stored procedures are not FUN, but for a 1000x improvement in performan
Re: (Score:2)
In the land of Oracle, all stored procedures are written in PL/SQL a descendant of ADA and cousin to Pascal, which includes object oriented features and native SQL connectivity that is revealed in Oracle development's constant use of SQL based cursors
In case you can't tell, I am a fan of Oracle database and PL/SQL and make my bread from Oracle applications, so as they say... It is best not to shit where you eat
Beyond that Oracle Licensing is a bane to anybody supporting any product sold by Oracle, as they
Re: (Score:2)
Re: (Score:2)
If Oracle innovate so much, why are so many of their products from acquisitions?
Oracle's RDBMS success: fear of change (Score:3)
Your opinion does not reflect reality. The Oracle RDBMS is not perfect by a long way but there are good reasons why it's one of the most successful databases and it can't every time be attributed to relationships leading to deals. If you compare to every other RDBMS product, Oracle has remained relevant by doing far more innovation than their competitors. Oracle continues to innovate even now. However, whether they can compete in a market that is moving more toward data processing than data storage, we will have to see.
Hmm, those innovation days must predate my career, which is now 25 years of working with Oracle RDBMS, 4 certifications, and it's one of my primary career focuses. Oracle's success is because people are afraid to change, often for legit reasons. It's usually the platform they've already installed. Half of the NoSQL movement wasn't because NoSQL was so great, but because Oracle was so awful and no one wanted to give SQL Server a chance (which last time I ran it...which was really LONG ago...was FAR FAR su
Re: (Score:3)
Oracle forces you to do all of the low level disk partitioning, why? There may be a few situati
Re: (Score:2)
My company uses it both their RDBMS, which well...for those who don't use it...it really really really really really sucks in many very very very stupid ways....
I still think this is best exemplified by one simple statement: "" = NULL
Re: (Score:3)
If he made MS-grade software...not GREAT software...just stuff about as good as SharePoint...
But you said that Oracle's financial software was crashing all the time and has trash UI... that is about as good as Sharepoint.
Re: (Score:2)
Oracle is known for being basically a scam so I'd have to blame B'ham's project management/IT team for this too.
If I was in charge of the authorities transport pool, and they needed a fleet of busses to transport school pupils, and I put in an order for 200 Chinese Tuk Tuk's, at an exorbinant cost when they are known to be cheap and garbage, is that really the manufacturers fault? Any competent IT manager should have known better.
Plus the way these corporates can offer to provide a service for a set price
Re: (Score:2)
This happens in part because people in management overestimate the cost of hiring staff and rolling your own, while underestimating the cost of buying because they don't understand the costs are ongoing and you still have to hire people and roll your own variant of the thing you purchased.
Re: (Score:2)
Governments can and do fail. You are completely free to move to somewhere that has happened: the local state won't stop you.
An yet you keep shitposting about governments from a country that has a strong one to keep you and your precious profits safe while you shitpost. You're a hypocrite.
Poorly negotiated deal (Score:2)
They should have required a flat rate cost structure and a max cap for any out of scope changes.
Oh well, it was only tax payer money, who cares?
Re: (Score:2)
Re:Poorly negotiated deal (Score:4, Insightful)
There is always more money for tea and biscuits.
Raise taxes.
Re: (Score:2)
Re:Poorly negotiated deal (Score:4, Funny)
Tea and biscuits are cheap enough to be covered by a small 10% tax increase. Maybe they'll have to sell a few buildings and will turn into a desolate wasteland. But, by God and Queen, there shall be tea and biscuits to the very end!
Re: (Score:2)
My Charles III 1£ coin is asking who this "Queen" person is.
Re: (Score:2)
"My Charles III 1£ coin is asking who this "Queen" person is."
That would be Camilla, wouldn't it?
Re: (Score:3)
Tea and biscuits are cheap enough to be covered by a small 10% tax increase. Maybe they'll have to sell a few buildings and will turn into a desolate wasteland. But, by God and Queen, there shall be tea and biscuits to the very end!
Turn into?
This is Birmingham, it already is a desolate wasteland.
Also it's king now (A.K.A. Sausage Fingers).
Re: (Score:2)
If it's already a wasteland then there's only one thing left to do.
Bring out the tea and biscuits!
Re:Poorly negotiated deal (Score:4, Interesting)
In the UK there is a limit on how much local councils can raise local taxes, without triggering a referendum. The threshold is currently 3%.
The thing that really screwed them is historic cases of unequal pay. Women were not paid as much as men for similar work, so they owed over £700 million in back pay and an on-going uplift of around £14/month for increased pensions and the like.
Re: (Score:1)
They can just change or ignore that law.
We must have our tea and biscuits!
Re: (Score:2)
How would a Labour run local council convince a Tory government to change the law just for them?
In fact, why didn't they just "change or ignore that law" instead of going bankrupt in the first place?
Re: (Score:2)
I don't ever comment on the internal party politics of other countries. Not my place.
Maybe they will change it. It's obviously an ongoing situation and needs to be resolved one way or another.
But whatever happens we know tea time shall never go tealess!
Re: (Score:2)
So the actual solution they are implementing is to sell off a load of public assets to cover the costs, because they can't raise taxes or get a central government bail out. That may change after the election.
Re: (Score:2)
"They can just change or ignore that law."
They can't change the law; the law is enacted by Parliament, not a city council. And the legal system gets rather sniffy when you decide to just ignore the law.
Re: (Score:2)
Oh, I see. You guys still actually follow your laws. Interesting.
Maybe they'll just stiff Oracle then.
Either way, tea shall be served at the regular hour.
Can't blame this one on Agile (Score:2)
This being a fixed-price contract, they would have had to have the specifications all laid out at the beginning of the project.
Some would like to say that agile processes are bunk. I say that bad project management can tank an agile--or waterfall--project.
Re: (Score:1)
The original contract was that they would switch to Oracle from SAP and change their business process to match the way that the Oracle software works.
Contract was signed.
They then decided they did not want to do that and modified the contract so that Oracle customize the software so that it worked the way they were currently doing things.
As for the bankruptcy it was more caused by the 1 billion lawsuit that they had to pay for the council not following the law on paying people.
They could have gone with SAP (Score:1)
like everybody with half a brain in Europe, but ya know... they regained their sovereignty, so they had to exercize it.
Re: (Score:3, Interesting)
This ONE Oracle fail made the news.
We've stopped paing attention to SAP/4HANA implementations failing because... that number is just silly and boring by now.
They're both over-priced, non-functional, garbage but, in many ways, SAP is worse.
Re: (Score:2)
Part of the overrun was they had to keep running SAP in parallel to verify results.
Which of the project manager was to blame? (Score:2)
Or all 65 of them?
Re: (Score:2)
Not likely (Score:2)
The issues described here are a mix of different kinds of incompetence, not just project management. The admins didn't fix the software's warnings/errors. The designers didn't account for all the use cases. And the accounting for the cost of the software was off by a factor of 5. Basically everybody on the BCC side was incompetent. I hate Oracle, but their stuff does work if you use it right.
Of course and they're also now asking for AI (Score:1)
no surprise (Score:2)
perverse incentives (Score:2)
Not the main cause of bankruptcy (Score:2)
The extract makes it look like the council declared bankruptcy because of this cost overrun. It was only one factor; by far the largest was the cost of compensating female staff for years of underpayment compared to males in similar roles.
Clueless government managers... (Score:3)
From an IT perspective, most of this is on Oracle. They are a big company, they really ought to know how to manage a software project. OTOH, from a PHB business perspective, Oracle did everything exactly right: they sucked five times the original contract price out of the customer.
The actual blame goes to the managers in the city government. They should have written a fixed-price contract, with fixed delivery deadlines and penalty clauses for not meeting them. Likely, they didn't, because Oracle wined-and-dined the individuals responsible, making them feel extra super special so that they wouldn't bother worrying about the contractual details. Once things started to go south, they threw good money after bad - the classic "sunk cost" fallacy.
Don't use the EU Flag Icon for this. (Score:3)
The UK isn't in the EU anymore. As a continental European I have to nitpick this and rub that one in for the Brits. ... That aside I'm pretty sure this sort of IT disaster is universal and can be found in EU mainland as much as in the US of A and other places.
Re: (Score:2)
I wouldn't be surprised to learn the Slashdot editors do these things on purpose, as a joke. Like they sometimes use the DEC "Digital Equipment Corp" logo for anything "digital life online" .
Oracle screwed Oregon too (Score:2)
I live in Oregon. Oracle charged my state about $400m for ACA related software that did not work. When the state sued, they offered to compensate by giving some additional software licenses for free. I thought that was more of a liability than any kind of benefit.
I am surprised that anyone does any business with Oracle at all. They seem likely a very scammy company.
Re: (Score:2)
Color me *NOT* surprised (Score:1)
Always blame middle management (Score:2)
Comment title says it all.
Invest in open source egov platforms.. (Score:1)
In India, the eGov foundation's [egov.org.in] ) DIGIT [digit.org] ) platform has contributed to effective digitization of government services with an open source platform. It is used extensively in many states of India for revenue collection, reporting, etc..
This was made possible by passionate technology professionals who wanted to make a difference "Keeping citizens at the heart of public service delivery"
I liked that BCC publishes its meetings for public viewing and scrutiny now.. at https://birmingham.public-i.tv/core/por
Re: (Score:2)
Wrong wrong and Wrong (Score:3)
American sees American thing (Score:2)
Re: (Score:2)
It may be a narrow definition, but your examples make your post almost self-contradictory. If Bavaria (not a "local authority" by any means: it's a regional authority) and Moscow (divided into 146 municipalities with their own councils) were to count as local authorities then so would London, which as you point out earlier is a conglomerate of various local authorities.