Digging Out from Under

By Peter Coffee  |  Posted 2003-10-08 Email Print this article Print

Hardware and platforms enable more helpful systems, if we can only displace the incumbent models.

In a month that includes both the Microprocessor Forum and the Microsoft Professional Developers Conference, its convenient to live in California—despite the embarrassing spectacle of this weeks ill-conceived recall election. Im naive, Im sure, in hoping that readers will let me critique the recall process without regard to politics. Trust me, please: Im not talking about who gets elected, but rather about how it happens. And the only reason that I mention it here is that, while thinking about that elections peculiar mathematics, it occurred to me that perhaps its not the most bizarre decision-making process ever seen. Its exactly like the way that we produce IT standards: it takes a majority to get rid of what you have, but only a plurality to decide what will replace it. Like an incumbent politician, the IT installed base benefits from inertia that often overcomes low levels of user satisfaction. In fairness, Windows XP has improved peoples happiness with the dominant personal platform, but thats damning with faint praise: only about a third of users who are running XP on machines no more than two years old describe them as never crashing. Thats not much of a vote of confidence, even if it is considerably better than the 7 percent of users who say the same about Win98 or ME.
Both Linux and MacOS X outscored XP in that same survey—and I wonder if both absolute and relative satisfaction with XP may have plummeted since those data were accumulated, given that our colleagues at PC Magazine published their survey in July before the summer from hell of Windows security breaches. Ive never gotten so many e-mails from eWEEK readers begging for instructions on how to get worms off their machines.
But lets look at the big picture: system crashes and security cracks are merely lightning bolts of annoyance—noisy and destructive, but relatively sparse and infrequent—in a climate of PC use where its constantly raining time-wasting awkwardness and complexity. Im looking to both the Microprocessor Forum and the Microsoft PDC to usher in significant improvements in the areas where people most hate their machines. MP Forum will be highlighting processor innovations that both boost speed and cut power consumption: Id be happy to see servers running faster, and full-strength portable machines running longer, especially if their 64-bit CPUs can run 64-bit environments that keep track of everything I do in terms of every association that might matter. Theres only so much that applications can do to make sense of our lives: Microsofts OneNote, for example, will try to put it all in searchable piles. But the underlying model of storage has to provide a lot more help: at PDC, well be talking about the forthcoming Longhorn, which might finally be the Windows replacement that actually delivers a far superior model of storage to displace the cumbersome and user-hostile model of discrete files in a catacombs of physical locations.
Its been said that God was only able to create the world in seven days because he didnt have to worry about backward compatibility. And as UK columnist Jack Schofield observed, in the context of that same joke, the difference between a good idea and a silly idea is often more in context than in the substance of the idea itself: For example, I loved the idea of Smalltalk-80 when I first saw it in 1985, but the Tektronix workstation that ran it cost $15,000. Hey, if you wanted 4MB of memory and 80MB of hard disk space, you had to be willing to pay. Ten years later, Java did pretty much the same things, but (i) it used a syntax that looked enough like C that developers werent scared away, and (ii) it could do those things on hardware that cost about one-fourth as much. All of a sudden, a virtual machine and mobile intermediate bytecode made sense. In the same way, the plummeting costs of bandwidth, storage capacity and processing power make it much more cost-effective today to ask our machines to find things, rather than expecting us to figure out in advance where well want to look for them. As I contemplate Californias ballot of 135 replacement candidates, listed in deliberately random order, that seems like a particularly helpful improvement in our lives.Discuss this in the eWEEK forum. Tell me what youd like to find, if legacy tech werent blocking the view.
Peter Coffee is Director of Platform Research at salesforce.com, where he serves as a liaison with the developer community to define the opportunity and clarify developers' technical requirements on the company's evolving Apex Platform. Peter previously spent 18 years with eWEEK (formerly PC Week), the national news magazine of enterprise technology practice, where he reviewed software development tools and methods and wrote regular columns on emerging technologies and professional community issues.Before he began writing full-time in 1989, Peter spent eleven years in technical and management positions at Exxon and The Aerospace Corporation, including management of the latter company's first desktop computing planning team and applied research in applications of artificial intelligence techniques. He holds an engineering degree from MIT and an MBA from Pepperdine University, he has held teaching appointments in computer science, business analytics and information systems management at Pepperdine, UCLA, and Chapman College.

Submit a Comment

Loading Comments...

Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel