Birmingham set to miss deadline to make Oracle disaster 'safe and compliant' Birmingham City Council — Europe's largest local authority — stands accused of being a "ship adrift in the ocean" after it failed to confirm it could make its troubled Oracle implementation "safe and compliant." According to auditors, the replacement for the council's SAP-based financial system has gone so badly BCC has not …
https://www.personneltoday.com/hr/birmingham-city-council-union-agreement/
Personally, my view is that if you need to second 30 people from their normal jobs for a limited period to calculate accurate accounts, that is what you do. Then we(*) know what the situation is, and people in BCC will know what processes need to change to conform to the assumptions made in the software (can't believe I typed that last bit).
I realise that The Register focuses on the ongoing Oracle disaster as it is the IT angle, but the agreement on work grading means that the more major historical liabilities can now be quantified.
Sheffield and Glasgow have smaller equality liabilities because they are (surprise) smaller authorities. Would be interesting to calculate a liability per job figure. But that of course would require financial software that works.
(*) Council tax payers
"I realise that The Register focuses on the ongoing Oracle disaster as it is the IT angle, but the agreement on work grading means that the more major historical liabilities can now be quantified."
The actual quantum of outstanding equal-pay claims is reasonable well estimated, the detail is merely to enable a settlement offer per case, and you're right that can be done essentially by hand. Unfortunately for the burghers of Birmingham there's the question of other liabilities which also need to be added up and netted off against the council's unknown asset position, such as £2.5bn deficit from the Brum share of the local government pension fund, and an assortment of other promises and commitments. I'd guess the city council's unfunded liabilities are about £3bn in total, there's something of the order of 480k homes in Birmingham, if we assume that net of housing benefit there's 430k properties contributing council tax, then that's an average of £7k per household, obviously weighted by net council tax contribution - unless of course government bail out the Bungling Brummies.
"a "vanilla" Oracle implementation cannot perform all the functions its heavily customized SAP system executed successfully"
In my experience that's true for any large and established ERP, whether public or private, and why ERP/CRM changes go horribly wrong and cost the earth. So that begs the question which individual or committee decided that an out of the box system with a few teaks could possibly replace a well established ERP that they'd spent at least 15 years running.
It's a shame the linked webinar here has been removed, because otherwise we might have the answers: https://www.linkedin.com/posts/chris-boyce-5184b921_webinar-why-birmingham-city-council-chose-activity-6877571266019151872-tL7x
However, Stocks said that, though significant, cost overruns were not his "core concern."
Please validate Brooke's law once more, add more people and throw more money at the project.
In the meantime, Birmingham will gradually transform into Oracleham. I'd say they are halfway through, but that isn't a "core concern" either.
Our councils is *description of organic waste*
Question: how to make it look that we want to improve things, without actually doing anything?
Answer 1: we could send out some surveys and review our targets, so we could prioritise things that align more with the needs of residents
Answer 2: our system is slow, that's why we can't deliver anything. Maybe let's change the system?
Answer 3: good point! we will be able to slack off whilst the system is being upgraded!
Answer 4: and if it turns out to be a car crash we will be able to slack even more!
Answer 5: Everyone, let's sing hokey cokey!
Surely there was a period of dual-running? When migrating massive systems like this, you don't simply turn off the SAP services on a Monday morning and rely entirely on the as-yet-untested Oracle: you populate the new system with test data and check it does what you need it to do. If it doesn't work, you halt or delay migrating users and live data until the problems are ironed out. Sure, you won't catch them all - but being able to do something as routine as producing financial statements must surely have been a baseline requirement.
Even then, there's usually a period of parallel running where data is entered into both the known system and the 'new' system. Again, if the outputs don't match - or if you discover a process takes 30 people rather than 3 - then you should delay the disarming of the old, stable, system.
Sounds to me a bit like BCC need to swallow a bitter pill and renew their SAP licenses. Yes, it'll cost a fortune because SAP know they have them with their pants down and will aim for full penetration - but it wouldn't be the first time a major project organised by a public body was badly managed (I give you NHS IT, HS2, any number of power stations and, if the Eye is to be believed, the northern Freeports). Umm...unless they didn't keep the SAP system current - abandoning that system this early in the changeover would be almost criminal!
If I'm missing something, forgive me - and do please help further my education.
The time overrun might be the problem here. There'd be a cost to just keeping the SAP system going. Worse still, if the new system was going to have to handle some new requirements those might have had to have been added to the SAP if it were to be kept going for an extended parallel run. It's not difficult to see how, once they'd run over time, they could get to the position where they might not be able to keep SAP running.
"It sounds more like they gave up on SAP very much too early, to save money one could guess."
Probably, and probably at the same time as slashing the Oracle implementation budget. This is likely because of what a Section 114 notice means. It isn't actually a bankruptcy, in the way a company goes bust, leaving its creditors high and dry. In UK law a council has to balance its budget during the year. if it become apparent that during the year it is going to be unable to do that, the senior finance offical has to issue an S114 notice, and that freezes all spending, meaning no new commitments and a requirement to start cutting existing spending and raise revenues (eg parking, entertainment, council rents etc). If things are so bad that central government have no faith in the council, they'll takeover.
By issuing an S114, the CFO is usually accepting their career is over; For councillor's it's often much the same, as the shame and stigma rightly stick. But that means that as things get bad, lots of possibly unwise tricks are pulled in the doomed hope of staving off S114. So during the months leading up to Brum council throwing in the towel there will have been desparate measures to try and make things look better by cutting spending on things that the public can't see.
"It sounds more like they gave up on SAP very much too early, to save money one could guess."
That was my point. They may well have been constrained to give up SAP - budget, end of contract or whatever. If, at that stage they were still unable to reconcile the results of running Oracle in parallel they were well stuffed.
If I'm missing something, forgive me
I think this one has such a complex history that we're all largely in the dark about the various machinations, though it's difficult to avoid the conclusion it was doomed from the start.
The SAP implementation goes back to an outsourcing partnership with Capita which clearly did not stay its intended course. BCC declined to answer a number of FoI requests about this on the basis that it was a commercial venture and therefore commercially confidential. The details of its dissolution are probably equally out of the public domain so it's not possible to know whether BCC actively put themselves in a position where all their eggs were in one basket, though that certainly appears to be how things have turned out.
Quote: "a "vanilla" Oracle implementation cannot perform all the functions its heavily customized SAP system executed successfully"
And yet, the contract was originally signed for a vanilla installation.
Is there really no penalty for being utterly incapable of doing your job in Local Government?
"You are forced to leave with obscene amounts of money."
How so? Councillors don't get paid off. Curiously enough the last Labour goverment briefly tried to make it lawful for councillors to get redundancy payments, fortunately they lost office before they could put that into practice. And under S114 a council will only be able to pay what was the contractual requirement for making a council offical redundant.
As for rounds of public speaking, what public speaking? "Ladies & Gentlemen, a big round of applause for today's after dinner speaker at this year's glitter and star spangled Prudential Regulaton Authority annual awards! Bracewell Fatgirth will talk about his experience at financially mismanaged Scroteshire County Council, and the events leading to him issuing a S114 notice!"
If I were those councilors, I'd just resign, en masse, and leave the problem to be dealt with by real experts: the kind who could readily be drafted from the commentariat, in a sortition-based system. Unpaid, of course. Yes indeed, city governments used to run like that. It worked for Renaissance Florence for three hundred years.
I'm sure the Oracle Sales team is working tirelessly to reassure Birmingham that they made the right choice.
I'm sure that the implementor is working beyond the limits of human capacity to string together customizations and PaaS solutions to give Oracle the outward appearance of being 'safe and complaint' (by the standards described in Indian technical colleges).
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
