Uni staff fall back on Excel to work around mis-coded transactions in Oracle system The fallout from Edinburgh University's ill-fated Oracle HR and finance implementation continues with one department recording thousands of mis-coded transactions relating to more than £300,000 in spending. More than two years after the Fusion HR, payroll, and finance system went live, the Scottish university's Biology …
> Outside of being a functional human being, I maintain that the single best skill you can develop for work in an office is getting good with MS Excel. It's never going away.
Silly unknown, we only moved out of caves due to an Excel-powered time machine. Here are the last 100 or so characters of the key formula: "))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))),0)"
The almighty messes that both Birmingham City Council and the University of Edinburgh have got into with Oracle's products are an indication that they ought to be avoided at all costs.
The ancient Oracle at Delphi issued prophetic wisdoms and guidances whereas the Larry Oracle consistently issues massive and horrendous cost overruns.
I know which Oracle I would believe and trust.
Read his autobiography, the guy is the usual slimey, 100% snake-oil flim-flam salesman. The only reason Oracle doesn't get sued more often is not that they can defend their atrocious products, it's that they have a bigger legal team than anyone else.
Oracle RDBMS was good from around v7.3 to v11 and after that, it's just a fricking mess of bolted on, half baked crap. As for their cloud, I'd be more inclined to call it a puff of steam than a cloud. They're now having to put their kit inside MS datacentres in order to get any sales, offering huge discounts. Oracle will like be one of the next behemoths to fall within 15 years.
When the output of the Delphic Oracle could be understood, it was often banal, politically self-serving, ambiguous, and/or padded out with instructions to do more worship, but Delphi's reputation management was superb and people really wanted something they could trust. So they did.
I was wondering that too. Does any company out there have a story of "We migrated to Oracle Fusion, everything went smoothly, the project came in on time and on budget, plus the system works well and satisfies all our financial and operational needs"?
No? Thought not.
OFM is the bane of the integration between the many disparate databases in my employer. Transaction too large? CRASH. Communications timeout? CRASH.
Now, I could give benefit of the doubt and say that it might be inexperienced personnel trying to drive the thing as the cause but, in every org I've ever been anywhere near OFM has been a liability. Don't understand the need for it, and don't understand the popularity of it.
It does nothing that a good old cron and a few shell scripts residing alongside your respective databases that need to do periodic tasks cannot do vastly better.
Does any Oracle product work anywhere? 10ish years ago I was working somewhere with an Oracle provided program that still didn't have 64 bit support and the interface for which required - although was not displayed through - IE6. As in, if you updated to IE7 it would crash when you opened certain parts of the UI. We had to maintain a windows 2003 server terminal services machine to run it because none of the staff desktops would. This isn't because it was old either, it was the latest version of the product available.
The real question with these kind of disasters is:"Who got paid by whom to push this through?".
It may not be payment in direct currency, but the "old boy's club's you scrub my back, I'll scrub yours" would certainly count for payment too. We can be sure that the "stakeholders on the floor" (AKA the low-life employees who have to use this crap) were not heard.
This doesn't sound like a problem with the tool - it's the processes and the people. If you want to book something to a job number or account that I'm responsible then I should get to approve it before it goes on the system, precisely to stop this sort of thing. That's how it's worked when I've owned a budget; on some systems it was electronic approval and on others I got paper timesheets and signed them before an admin entered them. Either way, nothing came out of my projects unless I understood and approved it. It sounds as though people were (are?) able to book to any old job number with no approval for the transaction before it went on the system. It also sounds like crap project management: hmm - I've had 3 FTEs on this job for 6 months and so far the spend is £500 against a forecast spend of £50k. Wow, what a top manager I am - on-track for a massive underspend and big bonus.
I think you have it back to front in this case. In the research side of universities it's usually the budget holders who are doing the spending, so they don't need to approve it as well. At Edinburgh, though, they have no easily available record of what they have spent and how much is left. So it's not that the wrong people are spending money from the budgets; it's that the right people are spending teh money butthe budgets don't exist.
I know - very well indeed - a senior person at Edinburgh who assures me that the system is still utterly chaotic across the whole university.
(Different AC)
They could be on a very sticky wicket - the money will be coming in from grant funders who generally have the right to audit spending on a project. If you can't show to their satisfaction that you spent it within the rules, on what you asked for, then the University may have to give some or all of it back. So I can see why people are keeping their own spreadsheets etc.
(Speaking as someone who has had to send in photocopies of paper log sheets for expenditures of less than a tenner, four years previously...)
Apparently, the Vice-Chancellor and Principal of the University received a salary of £418,000 in 2023 and benefits from additional expenses. I think the real issue is that when your senior staff are receiving that kind of money it becomes difficult to question their performance without bringing the whole appointment structure into disrepute and so it becomes a question of hanging together or hanging separately. That would seem to apply as a general principle in both business and academia.
It was “discovered“ that the Bath Uni VC was on a similar salary and had to leave out of embarrassment.
I know everyone uses MS Office and Windows and admins generally hate it because it’s so fiddly and as a result has something of a mind of its own, but this Oracle stuff seems to be in a whole different league. Surely Sale of Goods Act could apply if it’s not fit for purpose and promised functionality never appeared, but then the lawyers would hide behind it not being true off the shelf and being individually customised. However I’d love to see what they write up as successful reference implementations.
> It was “discovered“ that the Bath Uni VC was on a similar salary and had to leave out of embarrassment.
At the time, the Bath Uni VC was the highest paid VC in the UK, and some considered that to be unacceptable. It wasn't much of a discovery, given that it was a matter of public record.
Top VC salaries have since (in 2022) exceeded £700K; see:
https://www.studentbeans.com/blog/uk/heres-how-much-your-vice-chancellor-is-earning-while-youre-struggling-to-pay-your-rent/
All together now, chant the MBA mantra with me, "Profit at all cost! Profit at all cost!".
The reason no one under 30 can afford a house or kids aren't saving money and know they have no future, slimey, scummy MBAs churned out into the world, desperately squeezing every last penny in profit from "blood stones". Squeezing workers from the lower-end classes and now they've all been bled dry they're coming after the educated middle-classes for their money.
"Profit at all cost! Profit at all cost!".
"Senior managers don't get any kind of bonus pay regardless of their performance."
10 second search.
From University of Nottingham's website
"As part of the Reward Strategy, managers are able to reward exceptional individual or team performance via a one off payment through a Special Contribution Recognition Bonus, or an Exceptional Performance Bonus. "
Still think there is nothing for Senior managers / Directors?
Universities never listen to their own experts: they prefer to hire in consultants. It probably doesn't make much difference in the case of business schools, whose staff are as useless as their products, but every single IT disaster I saw in several decades as a lecturer was predicted by people who know about things and whose warnings were ignored.
>Universities never listen to their own experts
The problem is very few entities of any kind now listen to their own experts (presuming they have not already eliminated all their experts). Banks, universities, local authorities, infrastructure companies. None of them listen to anything except the siren call of maximum profit extraction. The society we have built is a dead end of meaningless accumulation; unfortunately, it is a dead end which is likely to end our existence as species.
Many of the worst mistakes I've seen in universities are when the "experts" (academics) are allowed to make any kind of decision outside of teaching or research.
Often there is a "halo effect" whereby it is assumed they will be smart at running large organisations because they sure know a lot about 16th century dutch textiles. In reality they tend to lack vital skills including planning and communication.
My experience at university was that anything computer-related was at least ten years behind what anybody in the real world was actually using. It took three years to get to something that I'd already been doing five years before I started there. I spent a lot of my first year wondering where the actual computers were, and when we'd be allowed to use them, only to discover...... there weren't any.
I survived a similar overhaul at a previous job and it seems to me that these all follow a basic script:
1. Vendor promises everything will work brilliantly and flawlessly without the users needing anything more than basic training and some new software. The higher-ups think "Wow, $vendor is a world-class giant corporation; surely they know what they're doing." and then selects the cheapest package option which is woefully under-spec'd, likely intentionally and knowingly by the vendor so they can sell up-charge add-ons to fix the problems they already know will occur.
2. End users are promised both transparency and engagement during the conversion process but end up being ignored or dismissed instead in order for the idiot who chose the vendor to save face and/or their bonus.
3. The old, but well regarded and bespoke finance system is shut down and for the next 5 years business processes in the company are completely reworked to fit the vendor's idea of how and where things should go; institutional knowledge is jettisoned and significant amounts of money are spent on vendor fixes and add-ons because in Step 1 the higher-ups low-balled it.
For all the money they spent, they could have hired a few dozen programmers who could have worked directly with the finance staff people to build exactly the sort of system that they want while still making it standards compliant on the back-end.
Speaking as an ex-Inoapps person, I can tell you their people were very good, at least on the HCM side that I dealt with. The Uni's people were pretty sharp too. Where it mostly goes wrong on Higher Education and other big projects is:
- Academic head of departments who refuse to be the actual managerial dept head. Approve spending requests? Waste of their precious time. Conduct performance reviews? Someone else's problem. The head of dept. role is a distracting burden they graciously accept for a couple of years, before the next prof gets a go. and they go back to their research with the right box ticked on their CV
- Shifting requirements. They sign up to "Adopt, not adapt" and it gets scoped and priced on that basis, then almost immediately start adding changes or saying legacy processes must be retained, even if they're patently cr*p. If you accept the lowest-price bid, it will have the least leeway for indecision, delay and general client bullsh*t
- Testing gets cut to the bone. Running late? Just do less testing and go live anyway. The SI tells you it's a bad idea, but you've promised the Vice-Chancellor / CEO / Minister and already delayed three times
- Reporting is always thought of too late. Can't build the report until you know what the data structure looks like. If you have to customize the data structure, you can't go with the standard report
- Data migration is always a dog's breakfast. Cr*p in means cr*p out. Always late, understaffed and incomplete, especially when the client insists on staffing it themselves, or just loads it onto the existing teams trying to keep the old systems working until the shiny replacement is finished
Oracle is rarely the issue. Workday, SAP, etc. all have their own good and bad points. It's the custom processes, indecisive client SMEs, and shoddy programme management that usually let them down.
(AC as I still need the work!)
You would like to think any University Department would have a business manager for admin stuff like this.
Even a GP Surgery has a Business/Practice Manager to take operations/admin/finance/payroll away from clinical staff who - like academics - who really have far better things to do with their time: Though providing oversight.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
