Aardvark DailyNew Zealand's longest-running online daily news and commentary publication, now in its 19th year. The opinion pieces presented here are not purported to be fact but reasonable effort is made to ensure accuracy.
Content copyright © 1995 - 2014 to Bruce Simpson (aka Aardvark), the logo was kindly created for Aardvark Daily by the folks at aardvark.co.uk
Please visit the sponsor!
History is littered with the bones of badly conceived, spec'd and implemented large-scale IT systems.
Here in NZ the acronym "INCIS" still strikes fear into the hearts of those who were involved with it even decades after it blew up in their faces.
Let's not forget more recent fiascoes such as the ill-fated NovaPay system that captured the headlines for many months and left some teachers scratching to pay the mortgage or put food on the table.
And now we have the Auckland City Council's "NewCore" project that seems to have gone way over budget and over-time.
I thought information technology was a science, not a game of chance.
Why haven't we learnt from all those previous failures?
Why is it that highly-paid IT professionals and project managers can't seem to add one and one to make two?
Is the analysis, specification, coding and deployment of any non-trivial software project simply too hard for the feeble human mind?
Hell, we can land robotic probes on Mars and comets hundreds of millions of miles away -- but we can't get even close to estimating the cost of a local IT system?
As someone who has spent a fair percentage of their working life in the IT industry, I know full-well that there are many, many pitfalls and "unknowns" that have to be identified and quantified during all phases of projects but honestly, it's not rocket science.
One of the biggest dangers with such projects is specification creepage -- the situation where those commissioning the project demand changes before the agreed system is completed. How do you avoid this doing very bad things to your timelines and budget?
Simple... tell those requesting the alterations to spec that they should come back once you've finished building what was already agreed. Signing off on a specification is not something to be taken lightly -- it becomes binding on both sides. Any subsequent changes to that spec must be just that -- subsequent to the completion of the project and subject to additional cost. I've seen far too many relatively trivial projects go wildly over-budget and time, simply because the customer wanted "just a small change" half-way through the development.
I've no idea what's gone wrong with the NewCore project but ultimately it is down to poor professional standards on the part of those involved.
However, if you want something to really worry about, read these news stories and see how the IRD's proposed massive IT project is already blowing out:
Remember, the IRD already has a pretty poor track record when it comes to implementing even relatively small IT projects: IRD abandons Oracle student loans system after spending $21m.
I'd love to hear readers predictions as to the chances that, what will be one of the largest-ever IT projects in this country, will come in on-time and on-budget.
Please visit the sponsor!
Have your say in the Aardvark Forums.
Remember, this is purely a gift, you'll get nothing other than a warm fuzzy feeling in return.