The recent RBS systems meltdown and the rumoured reasons for it are a salutary reminder to all as to how much we are all reliant on the continued availability of core IT systems; these systems are pretty much essential to modern life. Yet arguably the corporations that run these systems have become incredibly cavalier and negligent about these systems; their maintenance and long-term sustainability even in supposedly heavily regulated sectors such as Banking is woeful.
There is a ‘It Aint Broke, So Don’t Fix It’ mentality that has led to systems that are unbelievably complex and tightly coupled; this is especially true of those early adopters of IT technologies such as the Banking sectors.
I spent my early IT years working for a retail bank in the UK and even twenty years ago, this mentality was prevalent and dangerous; code that no-one understood sat at the core of systems, wrappers written to try to hide the ancient code meant that you needed to be half-coder, half-historian to stand a chance of working out exactly what it did.
If we add another twenty years to this, twenty years of rapid change where we have seen the rise of the Internet, 24 hour access to information and services, mobile computing and a financial collapse; you have almost a perfect storm. Rapidly changing technology coupled with intense pressure on costs has led to under-investment on core infrastructure whilst Business chases the new. Experience has oft been replaced with expedience.
There is simply no easy Business Case that flies that justifies the re-writing and redevelopment of your core legacy applications, even if you still understand them; well, there wasn’t until last week. If you don’t do this and if you don’t start to understand your core infrastucture and applications; you might well find yourself in the same position that the guys in RBS have.
Systems that have become too complex and are hacked together to do things that they were never supposed to do; systems which if I’m being generous were developed in the 80s but more likely the 70s trying to cope with the demands of the 24 hour generation; systems which are carrying out more processing in realtime and yet are at their heart, batch systems.
If we continue with this route, there will be more failures and yet more questions to be answered. Dealing with legacy should no longer be ‘It Aint Broke, So Don’t Fix It’ but ‘It Probably Is Broke, You Don’t know It…yet!’ Look at your Business, if it has changed out of all recognition, if your processes and products no longer resemble those of twenty years ago, it is unlikely that IT systems designed twenty years are fit for purpose. And if you’ve stuck twenty years worth of sticking plaster on them to try and make them fit for purpose; it’s going to hurt when you try to remove the sticking plaster.
This is not a religious argument about Cloud, Distributed Systems, Mainframe but one about understanding the importance of IT to your Business and investing in it appropriately.
IT may not be your Business but IT makes your Business…you probably wouldn’t leave your offices to fall into disrepair, patching over the cracks until it falls down…don’t do the same your IT.