I continue to read reports about slumping PC sales and about all the chip makers and PC manufacturers have to offer is that they expect the cycle to pickup as the economy improves. I believe the source of the slump is the result of the following, lack of imagination and disappointment over speed.

In my 2002 Buying Guide, I mentioned that many Pentium 4, 2GHz or higher systems didn't seem much faster than my Pentium III, 733MHz. If you are moving from a classic Pentium to a Pentium 4, you will see a noticeable increase in performance. However, as far as I can tell, moving from Pentium III to Pentium 4 yields a modest increase at best. It would seem Moore's Law, the doubling of processing power every 18 months, does not apply to performance.

As a result, only the most processor-intensive work, such as graphics, require constant upgrades. In the film industry, this can be as much as every six months. Gamers claim an ever increasing need for speed, but hardcore gamers occupy a very small segment of a very large market.

Nonetheless, the average user and most businesses aren't "plagued" by such demands. Even the advent of digital cameras hasn't produced much of an increased demand. Perhaps as digital video reaches a mass audience, this will change, as users require more processing power to edit their own videos. But when it comes to running most applications, word processors, databases, even spreadsheets, if you have a fairly high powered Pentium 3, you'll do fine.

There's something wrong with an architecture when a doubling of processing power and RAM only produces a modest increase in speed. Why don't my applications open faster, why don't the menus open faster, why are there delays as I navigate around my system (and that's with 256MB of PC133 SDRAM, an ATA 100, 7200 RPM hard drive and a 64MB DDR Graphics card)? With all due respect to Bill Gates and his new focus on security, I think we need a lean, mean operating system-- a "Windows Lite."

I know there seems to be the feeling that there must always be new options, new bells and whistles, but I think a lot of users would respond to a smaller, faster, slicker operating system on their PC.

The other aspect of this equation seems to be a lack of imagination about what users can do with a PC. A few years ago as hard drives began to increase in size exponentially at ever lower prices, I wondered why we didn't have the ability to use a computer to record television programs as we could with a VCR. Not long afterward, TiVo and Replay and what are called Personal Video Recorders (PVRs) arrived.

This type of unit is a natural. It's puzzling why such a feature didn't start on the PC. By that time, many PCs came with video cards that had "video out" capability. You could do it on a PC, but it was mostly aftermarket, the cards were unwieldy and the approach was clumsy. Innovation should begin at the top of the food chain. I'm not saying there isn't a place for a single use device such as TiVo. I am saying the technology should have been pioneered on the PC.

General Motors used to put their latest innovations on their most expensive cars, eventually moving those innovations to their other models. While PC makers are not quite in the same category in terms of models and structure, you would think new and innovative ideas that might someday spawn single-use gadgets would begin on the PC or the more expensive models, ultimately working their way down the product lines as cost comes down.

Even today, PCs have an advantage most televisions don't have, you can individually select the show you wish to record rather than going through a lot of input as you do with a VCR. Many people who haven't even considered a TiVo already have a PC and might find such an ability enticing. As was the case a few years ago, most PCs come with graphics cards with a video out connection. This would allow the user to view programs recorded on the PC on a television.

The above is just one example of a missed opportunity, but the point is, there's nothing new, no real hook that gives users a reason to move beyond today's setup. Further, I find it curious that few PC companies are in the aftermarket business. They make money selling desktops, laptops and servers. Only Hewlett-Packard and Sony sell a wide variety of devices that one might purchase after they've purchased a PC. Connectivity seems to be the mantra in the PC business today and H-P and Sony are well positioned to take advantage of it.

This is a part of innovation. It means PC manufacturers will have to dream up new uses, and if they don't make the add-on device they will have to partner with a manufacturer for that purpose.

The PC needs to become a full-featured communications center that users can access, not only from any place in the house, but also from anywhere in the world. Because it is stationary, it is the one point everyone can reach, assuming you leave it on when not at home. In that regard, greater attention needs to be given to a PC that can shut down in the face of a power outage but also power back up when it senses power has been restored and stable after a given period of time. Backup units are fine but each new iteration of the OS often makes the "Uninterruptible Power Supply" software incompatible, and this leads to issues for businesses and home users. It also means greater attention to security for "always on" devices.

I think we're going to need a greater use of wireless, although wireless network speeds are still well below that of common wired networks. I'd love to eliminate some of the wires in my own setup but I'm not willing to lose the 100Mbs speed I have, especially when I'm moving files of a gigabyte or more. It's all the more critical in a business environment where gigabit speeds are becoming ever more mainstream. Personally, I'd forego the wired network if I could just achieve 100Mps.

As the potential of connectivity has become more apparent, Microsoft has included UPnP (Universal Plug n' Play) capability and I certainly would like to take advantage of it with my TV and stereo. I have wires all over the place and wireless capability for these external devices, including digital cameras and video cameras, would be great. I realize legacy devices aren't ready, but if there was a reasonably priced device to which I could connect my stereo, TV, VCR and cameras, and have them interact with my Computer, that would be great-- all the more so if that interaction were wireless.

In some respects, I think we need to fix a few things. One of the most often mentioned problems seems to be the inability to play DVD movies on desktops and laptops. Users often don't have the correct software player or they don't have the correct Codecs. For something that is often used as a selling point for computers, it's still too difficult.

In the wireless arena, I note that Microsoft and others are now offering wireless keyboards and mice. Perhaps speakers should be next. The only problem is battery use, so I'm not sure any of these will truly take hold until some improvement in the battery is made.

One of my bigger complaints is the inability of CD cataloging applications to be able to pull information off of the "TOC" (Table of Contents) on the CD. If you've created a CD that has a counterpart in an online database, it's no problem because it goes online and downloads the information. But if you create a mix, probably the biggest use of music recording to CD, you have to manually type in what's on the CD if you are trying to organize your collection.

I've found a couple of applications that now claim to take the information from the "TOC" and I'll be testing a few shortly. Nonetheless, this is a glaring oversight. Something which has become one of the most widely used features in the digital world requires almost infinite patience if you intend to truly catalog and categorize your collection so that you could easily determine with a modest search if you have a particular song, who sang it, what track on what disk. It's just surprising that only now am I starting to see CD organizers that don't need to go online in order to automatically catalog a disk.

In line with that, many people have large vinyl record collections that they'd like to convert to CD, much to the chagrin of the record companies, but it is legal under fair use. Nonetheless, this is another needlessly complicated task. If you want the CD copy to have individual tracks, then you must record your LPs (vinyl records) to disk, one track at a time. Yes, there are track splitters, extremely complicated and too much work, and some programs sense the silence between tracks but also end up splitting any track with silent passages, hence this is unacceptable. It would be easier if we could simply click a button between tracks to alert the software of a track break. It would digitally make note and automatically split the tracks. Still a lot of work but much simpler than what we face today.

It just seems the computer makers either are not using their own product or they just don't take note of the very things they do in every day life that might be a task better performed on a PC. Where's the vision, where's the imagination? Find a need and fill it. It doesn't have to be something we've never done before, just an improvement on how we do things.

Build a better mouse trap.