Monday, November 18, 2002
IT ... Mission critical? Not yet.
A funny piece by Computing's Mole haranguing a big UK bank's system problems. I'm the last person to gloat over IT problems - we've had a few in government of course. But as we roll out increasingly complex systems with more functionality, cleverer widgets and increasing criticality for the customer, do we really have the ability to integrate, support and manage the IT we've got today? I was thinking about this especially because I'd just read a piece on John Gotze's weblog (who has been writing increasingly prolifically recently - I'm having a hard time staying up with him. He and I seem to be exchanging more and more links - we're not just in it for the search engine placings, honest!) . The note was about the future of web services and quoted an AtosKPMG consultant's theories. I've stripped out the ones that aren't relevant to my point (but they don't contradict it): 1. The standards underpinning Web services will hold up. 2. The financial services, travel, energy and public sectors will be among the first to embrace Web services. 3. Web services will dramatically change the software market. 4. Web services will become the basic building block of technology and business infrastructures. Point (1) is hard to disagree with - those standards are going to evolve but they work today and no reason to see why they should change, it's where and how they're used that will matter. Point (3) and point (4) are the ones that really scare me. Web services are only likely to be the building blocks of services inside the firewall (how will I figure out whether I can trust anything outside my own firewall as a corporate?), and the only way that they are going to dramatically change the software market is by giving us all something new to buy, something new to integrate into and something new to worry about schema standards, authentication calls and polling submissions. And then point (2) says it all ... two of the four industry groups (banks and government) have the worst records at implementing new technology - so unless someone else is going to do it for them (us), how will this work? And this comes from someone who led the team that built what you could arguably call the first public sector web service (maybe even the first?), the UK's government gateway. So, it's not that I'm against web services - I just don't want to see us fall victim to another oversold technology that will force us all into another sweeping upgrade path which most of us will fall off (and it's a long way down into the abyss from where we are now). I'm increasingly bothered that the new new thing (to quote Michael Lewis) becomes the be all and end all. But what matters still is "the perfect service". Please don't tell me that I need new new things to deliver services that match those criteria? Surely if Amazon can sell books with what it has today, I can deliver e-government services without a whole load of new new things? We've not even scratched the surface of delivering content managed websites with personally tailored information, let alone giving updates on services via mobile phone text messages? And now I have to build/buy/borrow/steal some web services? Agh.
Posted by Alan at Monday, November 18, 2002