What's Not Hot (Alas!) for 2003

What technological breakthroughs won't happen in 2003? The same as in most years: all the ones you really want. When the hardware industry isn't churning out boring boxes same as they ever were (only faster), it's indulging in replays of market failures like pricey computers that mimic cheap TV sets and legal pads. When the software industry isn't churning out bugs and security glitches, it's stuffing its products with dubious new features that lead to more bugs and security glitches.

The more inventive consumer-electronics industry increasingly develops slick products such as camcorders that link to PCs, but the devices are becoming as complex as the machines they connect to. And with both computers and consumer gear, progress seems to have slowed on fronts that really matter, at least to me. Things like:

Speech recognition: The era when PCs will be great at speech recognition always seems to be just a couple of years away. Though dictation programs like Dragon Naturally Speaking do a decent job of capturing what you say, they're dependent on training a single speaker, and they're still clunky when it comes to fixing what they get wrong. Speaker-independent systems like phone menus that let you use your voice instead of a keypad work well, but only with limited vocabularies.

I love my Olympus DM-1 digital voice recorder, but I want to use it to record a meeting and then have software transcribe it accurately no matter who's talking. That kind of system doesn't even seem to be on anyone's radar screen.

Phone/data combos: Actually, this idea is getting there--see "The New (Improved?) Wireless Web"--but nobody has come up with the perfect combination of phone, data, and network. The best so far are T-Mobile's Sidekick and Handspring's Treo devices. Both have acceptable thumb keyboards, but neither has a removable battery or a great screen. The weakest link is a costly one to fix: U.S. cell networks that are so spotty you're never sure you'll see your e-mail when it arrives.

Long-lasting portables: Chemistry--as in batteries--is the most frustrating aspect of portable devices. Low-power chips help; for example, a new version of Wi-Fi should aid the staying power of wireless PDAs. But just lighting a screen sucks up lots of juice, and low-power displays have been slow to arrive in real products.

High-definition TV programming: Once you see a few good shows in HDTV, looking at a standard screen becomes downright painful. Plenty of fine, if costly, high-def TV receivers are available now, but the vast majority of shows are still distributed only in standard definition. This will gradually change as program providers swap out aging equipment for the good stuff, but bandwidth and fiscal constraints will slow adoption. At least HD DVDs are on the way. But not this year.

Comprehensible software: My Windows 2000 machine recently slowed to the speed of tectonic plates--except when it came to crashing. Eventually I discovered that Windows itself was arrogating 99 percent of my PC's processing power even when it was just idling. The latest service pack seems to have solved the problem, but I have no idea why. What happened to "self-repairing software"? What happened to useful documentation?

Alas, inexplicable software and inadequate manuals are standard features of computing today. And I wouldn't bet on their being fixed anytime soon.

Click here to see past Full Disclosure columns by Contributing Editor Stephen Manes. He has been writing about PCs for nearly two decades.