So that's alright then.
Er, what sort of disaster is it?
The Ministry of Defence has moved to quash claims that its Defence Information Infrastructure programme is an 'unmitigated disaster'. It has told GC News that thousands of new IT terminals are being installed each month, and that changes in the programme have been made to respond to the demands created by the wars in Iraq and …
What metrics is it being judged by? Everyone in IT has had projects which have not gone 100% to plan. This is normally due to the client moving the goalposts midway though. Everyone in IT has also most likely had a project which was slagged to high heaven by end users despite being on target and on time. End users have their own metrics for judging the performance of a roll out and in most cases the fact that there is a roll out of something new is enough to condemn it in the eyes of those who have to then relearn something.
I would hate to think that a project I was working on would be judged on the subjective opinion of "Maureen from Accounts" who does not like the fact that her green bar at the bottom of the screen is now black and so would judge the roll out to be an unmitigated disaster in her mind.
Channel 4, in true media style, has leapt on the leaked hearsay of a few emails which contain unqualified opinions rather than asking for the real data on performance metrics to be released.
Sometimes consultants are good, they can come in and really help you solve a problem. Other times consultants are bad, really bad - a good example is just about any large scale system. No matter how much you know about computers/software/etc... it won't help you create a better system if you don't know just as much about the world in which the system will be installed and used. You just end up with a system designed by outsiders and only meets the needs of outsiders. Big government should be banned from using consultants to lead, develop, and deploy large scale systems.
Govt's should develop their own internal competencies and handle these things themselves. I realize that would change the lives of quite a few consultants, but there really are too many of them out there anyway.
"this has a keyboard or screen attached so where's the idiot's guide"
"She claimed that when the DII was first planned the "defence environment was completely different. We weren't fighting in two theatres....."
So their planning was for their current environment? and based on an assumption that the current environment would persist?
It also seems that they have again fallen for the monolithic model - it is more important to have seamless interoperability between systems than it is to have systems that actually work in the first place. Then we can go on and design things to fit the system rather than to deliver a worthwhile outcome.
Senior management in Government need to understand
- that there are different solutions to different problems
- idiots guides are exactly what they say on the tin
- sometimes you actually need to employ people with skills who know what they are doing
- one-size doesn't fit all
Firstly IANAC.
Secondly, I suggest that the problem may be that certain traditional areas in the UK may not always provide a dedicated career path for IT professionals.
Consequently, they can lack breadth and depth at some levels in IT and are therefore forced to buy in expertise when there isn't sufficient available internally.
Sadly, on really large projects there is rarely enough time and funding given to requirements capture so projects usually don't really know what they need until part way through the implementation. The project timescales are also usually so long that both the initial requirements and the people responsible will have changed by the time that the contract has been awarded and work started.
In my cynical view sucess or failure of a project in financial terms is centred on effective change control, and the only way to succeed is to have a strong investment in change control within both customer and supplier. Of course, both sides must also keep in touch with both reality and changing user requirements.
So almost any large IT project will go through at least one phase of "It's not what I wanted! Well, it is what you asked for!" and will succeed only if both sides can agree to a revised schedule and revised requirements.
[I have noted that it is a slow news month - there is a lot of fluff at the moment, including http://www.theregister.co.uk/2007/11/15/stolen_wifi/ , based on little fact and doubtful statistics]
Summary - bad things have happened in the past, customer and supplier now report everything is fine.
Perhaps a mitigated disaster?
and I am becoming more and more of the opinion lately that rather than trying to redesign out the myriad flaws in existing systems that we prepare the next generation in the normal, small increments and slowly phase out the older kit by integrating it or simulating it within the new systems. With the kinds of timescales and budgets available for these projects I'm pretty sure most the hurdles can be overcome involved in say either porting a windows 3.1 app to a more modern flavour or if need be writing it into a virtual environment and programming in any of the other hardware requirements needed to simulate it properly that cant be done off the shelf.
"sometimes you actually need to employ people with skills who know what they are doing"
All the time would be Better and Beta2, Colin ....... being AI Virtual Governance Presented by Skilled IT Professionals with Scripts Provided for Real FeedBack which appears to be defaulted all too often to Reaction rather than any actual Thought about what should be happening next with your own Input.
The Problem with Government is that it just does not think Big enough to be Effective at Governance. However, IT Does so that is who deserves Public Money [which Governments tend to assume/purloin as if their own].
The very fact that they submit to such fiscal controls in the Provision of Public Service is proof enough of their unsuitability for the Task and they definitely don't have Sight of the Bigger Pictures. But then, if that role contents them, why would we deny them it, although it would be delusional of them though, with everything having Moved On to AI Binary Age of Digital Control.
After having worked close to Mod, and also having worked in various commercial organisations as well - such as the ill-fated NR (!!) for example, I can safely say that MoD requirements process is the best there is, and the technical advice MoD receive is second to none.
I can only imagine that EDS have MoD by the short and curly's yet again, and that their 'consultants' are about as experienced as a twenty-something year old graduate can be - i.e. - not.
Time and time again it seems that gargantuan government IT projects are ill-defined and unmanageable. Hence the need for an adaptive, responsive requirements process. Companies like EDS are just interested in the bottom line. They prefer an outdated waterfall aproach, or else charge for any changes.