Does anyone remember the cult-sci-fi movie Dark Star? This first film from director John Carpenter has a team of bored astronauts in a beaten-up spaceship, controlled by an unhelpful computer that holds the crew it serves in total contempt, on a mission to seek out and destroy unstable planets.
The best moments of a decidedly smoke-filled plot come when one of the astronauts attempts to persuade a malfunctioning, intelligent but suicidal nuclear weapon not to explode while it is still attached to the hull of the spacecraft.
Dark Star is, increasingly, a metaphor for computing in the 21st century, as systems become more "intelligent", with more than half of the computers in existence today operating in an 'unmanaged' environment.
Artificial intelligence isn't quite up to the standards of Dark Star yet, but the complexity of the systems and the demands that we place upon them is driving autonomous capability at roughly the same speed as Moore's Law, whereby the amount of information storable on a given amount of silicon doubles every 18 months.
Take Windows XP as a modest example of more control and intelligence being built in to the operating system. Contrary to my initial reservations about the product, I really quite like it because it doesn't demand my intervention when things go wrong.
I'm suffering from a series of system crashes - from which XP is, theoretically, immune - but which "in true life", as my eight-year-old daughter would say, still happens.
XP recovers the system and the corrupt registry each time and even obligingly passes the details to Microsoft over the Internet, in contrast with my Windows 98 machine beside it, which crashes Outlook several times a day and leaves me pulling my hair out. This is why I rarely turn it on anymore.
Of course, this is hardly the stuff of HAL in the film 2001: A Space Odyssey, but I predict another decade will see the end-user increasingly removed from the management of the "ghost in the machine".
Computing is still, arguably, in its "pre-industrial" stage of evolution. In the relatively short space of time that I have built my own career in this industry, the information landscape has become almost unrecognisable, driven by two great leaps, the local area network and the arrival of the Internet.
The next great jump into grid and peer-to-peer (P2P) computing will introduce new levels of complexity which, in turn, will drive the demand for more intelligent software to manage the intricate web of relationships and services between different systems, in much the same way as a Boeing 777 doesn't really need "a man with a hat" anymore. He's only there to keep the passengers happy.
Given the pain the industry is experiencing at the moment, I would predict that the status quo, in terms of systems management software, will remain very much the same, in terms of incrementally adding more complexity and offering "better integration and efficiency" for the next five years.
However, after this and before 2010, which will give time for the next generation of computers to replace the millions around us today, I would expect a conceptual leap forward in the way in which a planet, full of interconnected systems, is managed.
Whether this will leave tomorrow's IT director arguing with a sulking and unco-operative systems management programme is anyone's guess, but I predict that the management workload will steadily increase until we break through the administrative glass ceiling imposed on the computing environment by today's technology.
What's your view?
How near are we to hands-off system management? Tell us in an e-mail >> CW360.com reserves the right to edit and publish answers on the Web site. Please state if your answer is not for publication.
Zentelligence Setting the world to rights with the collected thoughts and opinions of the futurist writer, broadcaster and Computer Weekly columnist Simon Moores.