The Ministry of Justice has hauled in EDS for what promises to be a lively discussion about the future of the prisoner record programme. The National Offender Management Information Service (NOMIS) was meant to provide a central database for offender records to link prisons and the probation service. Offenders would have one …
How hard can it be?
A database (120,000 ish active records at any one time)
Secure access for the prison service, probation service and possibly the police
I keep trying to think of how this could cost 500 Million or more and I'm sorry, I can't do it.
I am sure there must be reasons why this is such a difficult proposition which makes the costs rise so much. I just wish someone could explain why, not just in this case but almost every Goevernment IT project that at first glance seems a simple proposition the contractors end up blowing the project out of all proportion.
A lot of money for exactly what?
"The project was originally slated to cost £234m but this has gone the way of all government IT projects - by August 2007 it had doubled to £690m. Unions put that figure closer to £1bn."
Assuming this has no issue w.r.t. National Security, I see no reason why the Business Case and Project Initiation Document can't be put in the public domain. What exactly are the tax payers getting for close to a Billion pounds... a national database for how many offenders and how many users... are they giving every user a new laptop and everyone on the database an electronic tag ? Is training for the users taking place in Barbados ?
Yet again EDS is unable....
....to develope a database. Come on, how difficult can it be to build a database and the front ends?
Yes, I know it's more then an Access DB. But you only need a DB to track 100k of prisoners - what about all these delivery firms DB's they have to track million of items in transit in many countries - few ever fail or cost as badly etc.
EDS have a habit of only doing the bare minimum to comply with the contract; with the least cost by the poorest paid developers who couldn't give a rats arse coz they may be getting made redundant soon!
I know, I worked for EDS years ago!
I could knock up something fairly similar for a few grand. This is the UK taxpayer being held over a (pork)barrel.
And they couldn't even keep it secure.
Nightmare on Marsham Street
NOMS is internally known by the above title -- the street address of its HQ.
Quantum, the all-singing all-dancing IT system, should have been in years ago, Technically, it was. But only if you redefine the scope of the project to be Microsoft Office, not any prisoner/prison database.
Almost all prisons still plod along with LIDS (the Local Inmate Database), which runs in a DOSbox inside Windows.
And, by some strange unfixable quirk, a Quantum userid is good for only one jail. True, it's a national system: you can log in in any prison. But you only see the intranet for your one "home" jail.
That works fine for officers and others who only work in one jail. But think of probation staff, education staff, area staff, and others who circulate around multiple prisons, They either have to have multiple Quantum ids (which is against regulations, and a little hard to arrange these days) or shoulder surf at any prison other than their "home" one.
Plus, if a user has neglected to log out, the only way to get past the "this terminal is locked" message is to find them, or to recycle the power. The second option will cost you up to five minutes of watching Windows lumber into action.
And remember, almost all the money spend so far has been on a rollout of Windows Office: LIDS' replacement (Called C-NOMIS) remains as vaporware.
The whole system is a huge joke....Or at least that could explain all the laughter heard from the EDS offices
Yet another EDS failure ...
Do they actually have any _happy_ customers with projects that came in on time and on budget?
* former Bull/Integris employee who worked under EDS in the CSA and EOI fiascos
How does one go about getting one of these contracts? Whenever I quote for a job, assuming the requirements don't change, the price doesn't either.
If I tried to double the price I'd be had up for breach of contract!
Can you imagine the fuss if a plumber or electrician did this kind of thing halfway through redoing the house?
@ AC 15:41 GMT
"I keep trying to think of how [a database of 120,000 ish active records at any one time] could cost 500 Million or more and I'm sorry, I can't do it."
Here's a likely cause: on the gubmint side, there's a committee, and they keep thinking of new bells and whistles to add to the project and never take into account the cost of unnecessary system complexity. And the chair of the committee is some NuLab hack who has neither the balls to say "Enough!" nor any useful insight into system design issues.
Such committees fall in love with the concept "oh, someone may have [this or that unlikely situation] and we have to accommodate them." The database designs arising from this kind of loose thinking tend to be extremely rigid, lacking (for example) lots of free-form text fields for entering miscellaneous commentary.
How do they keep getting contracts?
How does EDS keep getting these contracts? Who keeps hiring them? Have EDS ever delivered on a contract on time, working and on budget?
EDS is a great way to make a lot of money.
For several years I followed EDS around cleaning up the cesspool that they made of various companies network, DNS, and mail gateways.
It's good work if you can get it. Maybe I should get some friends together and bid on this project.
Having worked on similar projects in the dim and distant past
I'm fairly certain the project has failed because the civil service want it to fail.
In the private sector IT replaces people because companies need to compete or go out of business. In the public sector if IT replaces people then budgets are trimmed, civil servants laid off and precedents set. Great for the taxpayer. Bad for Sir Humphrey.
It's an irony lost on government planners that the civil service who are the key stakeholder with a powerful vested interest in the failure of the project are the ones who get to set (and reset, and reset, and reset) the project guidelines.
Cue planned failure.
That also neatly explains the question of why EDS keep getting rehired. They know that if they actually point the finger at the civil service they'll never win another project again. If, on the other hand, they keep quiet and take the blame (as they did for the Child Support Agency and numerous other public sector projects) then they win the next project (and the cycle begins again).
It's only mystifying if you fail to understand that the civil service is competing directly with the planned IT systems designed to streamline their workings.
Surprise, surprise - another government contract goes massively over budget (although not nearly as bad as "BLoad" seems to think - what planet is he/she on?). Rather than point the finger at the contractor (cheap shot imho) why not look at the 'customer'?
I've been (un)lucky enough to be on a couple of government contracts and they've all been the same - you get a statement of requirements that then seems to undergo a phase of weekly (or more frequent) changes. This period of change usually lasts from about two weeks after you start until about a month after you've delivered it!
And the problem with doing this is, (apart from the total frustration it causes in the poor schmuck trying to do the work), that every time you get changes it adds costs and really doesn't help quality.
Plus this isn't just an IT specific - as someone was remarking to me the other day, that eyesore that the Scottish Parliament sits in cost about four times what was quoted - and no EDS, IBM, CapGemini, etc involved there I thought.
Last thought - is it just me, or doesn't it sound increasingly like the NAO should be exercising oversight on these large value projects earlier in the lifecycle? After all, if they can stop these slipups then surely that'll save the UK tax payer some money?!
The problem with any Government contract is that the original contract is negotiated by Senior Civil Servants and Ministers who don't actually know what is required of the final product, they usually keep the users out of the loop until the final stages when the product is discovered to be unusable. At this point there are massive changes to the original product which results in the supplier being able to name their own price for the alterations. If users were consulted from the outset there would be a lot less wastage on Government projects. The trouble with the Civil Service is that it is so gradist it is unbelievable and if you're not a high flyer then you don't have an opinion.
Posted anonymously as I don't want a police visit for revealing Government deceptions.
Have EDS delivered any Gov projects on time and to budget with good quality etc?
Yes, the ESA project, Employment Support Allowance. Now seen as a flagship and example of how things can be done right.
Of course this is because most of the staff were ex civs that knew the system backwards.
Many of whom face redundancy now. I guess the ESA releases after we get laid off may be more like EDS's preferred outcome, a disasterous moneysink.
AC because, well, it's obvious.
Mines the one with the CV in the pocket.
Standard Government contract...
Generally speaking, Government projects that go this wrong are very poorly specified or very poorly managed, sometimes both.
Add in the possibility that EDS hasn't put the right resources in at the right time - a frequent problem when I was there was that you'd end up with people working in completely unsuitable positions just so that they were 'chargeable' - with no training or intent of training by the company.
And don't forget that EDS would have won the business with a rock-bottom price and a contract that penalised change with lots of extra charges. Standard practice for consultancies bidding for government projects these days. I've worked with others (Capita, CapGemini to name but two) and they are all cut from the same cloth.
Things will not improve unless and until the government changes they way they specify their requirements (involve the end users from the beginning) and manage change. Then you'll be able to see when the problem is down to supplier incompetence (that is, PM incompetence).
And to all those people who say 'I could do that for 100k' - you couldn't. Even when the software is relatively simple, the infrastructure, support, disaster recovery and security requirements are never simple for a government contract, and everything has to be documented in detail and reviewed and approved by other consultancies retained by the government for this purpose. Of course, they have to prove their worth by finding issues even when there may be none (you know who you are).
Don't get me wrong, I'm not defending crap management or flawed technical implementation but I've seen enough government projects to know that most of the time it is down to bad management and/or customer interference.
"How does one go about getting one of these contracts? Whenever I quote for a job, assuming the requirements don't change, the price doesn't either."
the clue is in the question, THE REQUIREMENTS ALWAYS CHANGE.
Sometimes the requirements change too dramatically.
Fujitsu said no:
- Xmas Round-up Ghosts of Christmas Past: Ten tech treats from yesteryear
- Special Report How Britain could have invented the iPhone: And how the Quangocracy cocked it up
- Analysis Microsoft's licence riddles give Linux and pals a free ride to virtual domination
- Massive! Yahoo! Mail! outage! going! on! FOURTH! straight! day!
- Bring it on, stream biz Aereo tells TV barons – see you in Supreme Court