Sep 032012

Will it be Intel inside or Java everywhere? his month, a recent story¬† highlights the most radical bid in years for a new computing platform: Sun Microsystems’ Java chip/Java language combination. This is not just an alternative to Intel/Microsoft computing, a la the PowerPC/Mac OS package. It’s a reinvention of the computer into a more universal, versatile, and interoperable device. The Java model could obliterate many of the distinctions between standard computers, embedded systems, and products still to come, such as intelligent cable boxes and cellular phones.

For years, we’ve lived with chips that will run almost anything and languages that will run on almost anything. Java chips will give us performance where we need it–running Java applications–while the Java environment will give us interoperability with every other hardware platform.

Sun hopes that electronic designers, corporate information technology (IT) departments, and, ultimately, end users will find this so compelling that they will leave behind the Intel architecture. However, it’s a tall order. Look at how hard it’s been for the mighty PowerPC troika of Apple, IBM, and Motorola to make headway against the x86 line.

Still, I think Sun will win a significant place with its Java chips. In the last year, we’ve started to hear some awful gnashing sounds coming from Redmond, as Microsoft grinds off a few gear teeth trying to keep up with the paradigm shift toward the Web. On the hardware side, Intel hasn’t made any major adjustments to the infobahn. Check out our story on Intel’s road map (“The x86 Gets Faster with Age”) on page 89, and the only new signs that you’ll see are those for multimedia. Keep those MIPS coming!

There’s no question that simpler, cheaper, faster PCs will spread computers to people who have never used them before. As business communications and public commerce march relentlessly toward computerization, the network computer (NC), or Web PC, or whatever you call it, will proliferate. Ditto for new devices we haven’t named: the wireless gameboyphonebrowseremotecontrol and such.

But Sun may be ahead of its time. When will the network infrastructure arrive to make fully networked computing possible? It will definitely be a year that begins with 20 before we have the kind of end-to-end digital broadband network we’ll need.

So don’t put all your money on one horse just yet–you’ll need both Java and Wintel for the foreseeable future. Java chips and the Java language will solve some of client/server’s thorniest problems. (Do you really want to individually maintain hundreds of PCs that are nothing more than cash registers or data-entry terminals?) The Java duo will also make lots of cool but dumb stuff like cellular phones much smarter. What it won’t do, however, is protect people from hard disk failures, as noted at

Yet our familiar OSes and applications make the classic PC–Intel, PowerPC, or whatever–much more flexible than those devices will ever be. Throughout the next decade, I may want a Java unit for the office and a self-sufficient but connectable machine for computing on airplanes or when I’m otherwise outside the wired urban hubs of the world.

Win, lose, or draw in the commercial war, Java has already changed the way we think about computing. Can it be a coincidence that Microsoft has revived its hand-held computer OS project, now called Pegasus? Or that Oracle hopes Intel will buy into its Network Computer concept? That’s not the end of it: Look for traditional OS vendors to let you turn the desktop GUI into a screen onto which you project server-based applications, NC-style.

Users’ interests lie in being able to flexibly combine the best elements of the Java concept with the best of what we have now. For example, products such as Notes and higher-end database managers can choreograph local and centralized data management. Make that skill widely available to all sorts of applications, and we’ll have the best of both worlds.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>