This year, desktop Linux finally emerged to assume a tangible enterprise presence, albeit one that resides more in the marketing collateral of IT vendors such as Sun Microsystems, Novell and Red Hat than on the average mainstream or even limited-purpose enterprise desktop.
Still, the Linux desktop products of 2004, including Suns Java Desktop System, are effective, well-made Windows alternatives, and I expect to see these solutions begin chipping away at Microsofts desktop monopoly next year.
One of the big advantages that desktop Linux will enjoy moving forward is the competitive climate of rival distributions, both from major vendors and from the constellation of small projects and companies that produce Linux distributions.
The big Linux story of next year may be competition from UserLinux, an enterprise-oriented Linux distribution based on the very popular and well-established Debian Project. UserLinux, which is now available in beta form, will lack the per-seat costs that accompany Red Hat Enterprise Linux and will enable companies to get support from numerous sources.
However, companies out to jump the Windows ship in the year to come will find that, after having been tied so tightly to a single vendor for so many years, no attempt at defenestration will come without appreciable migration pain. Open-source applications have grown more mature, as my colleague Anne Chen points out, but in many cases, an alternative to a Windows-only application doesnt exist or isnt ready for prime time.
For one thing, the typical Web-based application, which includes everything from network appliance configuration interfaces to hosted CRM applications, remains stubbornly dependent on Internet Explorer—which is to say, on Windows.
If vendors cant bother to support multiple operating systems on the Web, which is inherently cross-platform, the prospects for widespread porting of proprietary applications to Linux seem somewhat bleak.
However, with no end in sight to the worm and spyware hassles that IE and Windows are heir to, many companies may find that a switch to a less vulnerable desktop platform such as Linux or Mac OS X is well worth the trouble, as is demanding full cross-platform support in the products they buy.
While were on the topic of alternative platforms, 2004 also saw handheld computers grow predictably more muscular in terms of processor horsepower and on-board memory, storage and wireless radios. However, long-unresolved issues related to mobile Internet connectivity, input mechanisms, and display and battery life limitations—along with a too-stagnant handheld software market—have left these devices all beefed up with nowhere very interesting to go.
Weve used Ethernet-based KVM units from Avocent in the Labs, and Id love to see handheld hardware and software makers pursue this sort of technology as a way to allow users to take advantage of their increasingly capable mobile devices while docked at a standard-size keyboard and display.
Of course, until handheld computers are built with longer-lived power sources—perhaps in the form of fuel cells—were not likely to see these devices escape their roles as strict PC subordinates.
On the mobile Internet front, I was encouraged by the performance I recently observed in testing UMTS (Universal Mobile Telecommunications System), EDGE (Enhanced Data for Global Evolution) and 1xRTT wireless data services. However, carriers must cut service costs to expand their subscriber bases and make the mobile computing market more attractive to application developers and service providers.
Senior Analyst Jason Brooks has been a member of the Labs staff since 1999 and was previously research and technology coordinator at a French economic development agency. Jason covers the operating system, productivity application and wireless space and is a leading expert in the area of open source.