Some things are relatively easy to predict when it comes to the IT sector. Each year, there are a number of well-publicised computing disasters, and there is no reason to believe it will be any different in 2005.
Even when disasters are avoided, business users of IT have often been left with a sense of disappointment as cherished projects fail to produce the expected benefits or financial return. As Crispin O’Brien, head of technology for KPMG in the UK, points out, over the years the software business has frequently failed to come up with the goods: “It hasn’t done what it said on the tin,” he argues, borrowing a phrase from a UK varnish commercial.
The IT industry would like to believe, too, that its commitment to customers is more than skin deep. Observers such as Mr O’Brien are optimistic that the software industry is displaying a new pragmatism and maturity, forged in the challenging business environment of the past three years.
So will there be more disappointments for enterprise users of IT and telecoms next year? Or will some of the emerging trends of the past few years turn into unstoppable bandwagons? In communications at least, there is plenty of activity. But elsewhere, next year could also be the point at which some technological chickens head wearily for home, as the adverse consequences of the enthusiastic adoption of advanced systems become apparent.
Information overload is one. Many companies, as James Hall, Accenture global managing partner for technology makes clear, are drowning in information. “Our traditional tools for organising and managing this flood of structured and unstructured information are running out of puff,” he says, pointing to the fact that it is often easier for an executive to find information on the worldwide web than to extract it from the company database.
So enterprise content management systems are high on the wishlist for many companies, as is the kind of software marketed by companies such as Business Objects or Cognos which enable executives to draw useful insights from the data. Furthermore, Mr Hall argues, the information deluge will keep growing: he foresees that within five years, objects rather than individuals will start the data ball rolling. He means, for example, that a car engine will take the initiative in reporting back to its manufacturer or distributor via wireless and the web.
Moore’s Law, the doubling of computer power every 18 months or so, has traditionally been the physical justification for many predictions of the technological future. So there was some shock this year when Intel, the world’s largest semiconductor company and dominant supplier of personal computer chips, announced two months ago that it would no longer try to squeeze more speed out of its Pentium range of microprocessors, the current workhorse of the PC industry.
There was no great surprise about the reasons: for years, chipmakers have struggled with a mismatch between high processor speeds and low data flow rates. Furthermore, chips running at several million hertz produce enough heat to grill a squadron of chickens.
Intel’s announcement, however, is the starting signal for the emergence over the next two years of designs for high performance PCs, games consoles and servers based on multiple microprocessors on a single chip, multiple chips on a single circuit board and multiple systems configured as a grid. Michael Gold of Stanford Research Institute argues that these designs will outperform Moore’s Law over the period: “Performance could increase by a factor of four to 16 for several classes of servers and desktop computers”, he predicts.
Meanwhile makers of graphical processing units (GPUs), the chips to which traditionally the task of handling images was offloaded from the central processing units or CPUs, are claiming these devices could form the new heart of the PC. David Kirk, chief scientist for Nvidia, a leading GPU maker, says the devices are cheaper than CPUs but capable of handling tasks from general purpose computing to scientific simulation. He predicts the emergence of servers powered by a combination of a multiple core CPU and a multiple core GPU. Such a device could rival the speed of the Japanese Earth Simulator supercomputer, housed in a three-storey building the area of four tennis courts.
In communications, voice over IP - telephone calls delivered over internet technology - is set for explosive growth, according to Niel Ransom, chief technology officer for the French telecommunication group Alcatel. Most of the world’s major telecoms operators are already building IP networks, replacing their 25 year old switches with computers capable of handling data.
Meanwhile the hottest show in town is Skype, a UK-based company founded by Niklas Zennstrom and Janus Friils, who created the peer-to-peer music download service KaZaA. Skype is a peer-to-peer international internet telephony service which is both free and said to overcome the usual VoIP problems such as poor call quality, low call completion and the inability to penetrate firewalls. Skype is backed by the legendary Silicon Valley investment house Draper, Fisher, Jurveston. Tim Draper believes the company is the “hottest viral marketing phenomenon since Hotmail”. How will it make money? By charging, it seems for premium services while retaining a free basic service. But what effect will a successful Skype have on the revenues of traditional international operators?
A raft of technologies is being deployed to render telephone users mobile and connected. BT, for example, is working with a seven-company consortium including Alcatel, Ericsson and Motorola to develop Europe’s first fully converged fixed-mobile communication service. It is expected to be launched in the first quarter of 2005. The service, “Project Bluephone” is a staging post to the situation where telephone subscribers have one telephone, one telephone number and one telephone bill. In the home or office a small base station will be linked to the internet through a broadband connection and to the handset via Bluetooth short range radio. Already some 15 per cent of European phone users have abandoned a fixed phone line in favour of a mobile phone. In the US, Cingular Wireless, the largest US wireless operator is similarly working on fixed-mobile convergence.
The main event, however, is likely to be the battle between third generation mobile phone services and other wireless technologies for broadband data communication on the move. “There is no doubt that this will be the make-or-break year for 3G” says Colin Jackson of the Brussels-based consultancy OTR “both in terms of whether it will prove to be useful and whether it works”.
Competition will include WiMax, a wireless technology much like the short range Wi-Fi through which you can surf the web from a coffee shop, but with a much larger range. WiMax, which is expected to begin full scale deployment in 2007 has a reach of between 1.5 and 15km compared to the 30 metres afforded by Wi-Fi. It may not provide much of an advantage in an urban setting, but it could be the answer to cheap and effective broadband deployment in rural areas.
Mr Jackson of OTR, however, worries about the amount of valuable data that leaves offices every day written into the laptop computers, personal digital assistants, mobile phones and portable, USB-based memory devices. He foresees a time when employers may require employees to deposit a complete copy of the data to which they have access in case questions have to be asked at a later stage. He rightly predicts that such a measure could be seen as unreasonably intrusive.
Security will be an important issue next year and for the foreseeable future. Linux and other kinds of operating software may benefit from the problems Microsoft is experiencing in securing Windows against the ill-intentioned. The biggest danger remains the possibility of a sustained attack on the internet. Originally designed to withstand the effects of a nuclear blast, the internet could prove vulnerable, indeed has already proved vulnerable, to sustained, software based denial-of-service attacks. The internet is too central to too many international and national activities to be allowed to fail.
Get alerts on Technology sector when a new story is published