The Future of the PC

This month we’ve seen a number of vendors present their views of the next generation of the personal computer, starting with the traditional PC vendors at CES and ending with Apple at MacWorld. Let’s spend this week talking a bit about the different camps. While it goes without saying that each is absolutely convinced it’s right, each also has issues that need to be addressed.

Linux: In Search of Innovative Hardware

One of the problems with an initiative driven largely by programmers is that they feel the world would like what they like. This is similar to what happened when Windows first appeared: The technical programmers thought it was stupid, and Unix, for the most part, retained a heavy command-line interface capability to address this technical need. Competence is king with Linux, and those that lack that competence have found it difficult to install, maintain, secure and use the system.

The first true commercial desktop version of Linux was from Lindows, now called Linspire. From a distance the product looks a lot like Windows, and many things even seem to work the same way. However, this is a hardware-driven distribution, and if you want a decent game or third-party application to run on it, you have to use an emulator. Emulators are getting better, but they still consume about 40 percent of your available performance.

Novell, Red Hat and others have a number of hardware partners that could make this situation interesting, and IBM’s divestiture of its PC unit may position it for a repeat of the relationship it once had with Microsoft, only this time with Novell or even Red Hat. The end result may be very close to a Linux version of the Mac OS on Intel, with a different application set but a similar user experience. This is probably one of the main reasons Apple is refusing to let a version of iTunes for Linux see the light of day.

The problem with products associated with this platform include fragmentation, the lack of a common strategy and an almost exclusive focus on low price. The lack of commonality suggests that interoperability problems will increase over time, and the excessive price focus suggests that products may not have the quality this segment demands. Still, TiVo is a good-quality offering, it is sold at an attractive price, and it runs on Linux. That shows Linux is viable, just as TiVo’s lack of profitability shows the risks.

Apple: A Trip Back to the ’90s

Apple’s latest offering, the Mac mini (I’m assuming Apple isn’t planning a Big Mac), is similar to the old Apple Cube in many good ways. It has a clean industrial design and is easy to set up and use. The old Cube was well received for its looks, but it was horribly designed, which led to its market failure. The mini is well designed and arguably the best value in an Apple PC today.

It is also very limited, having been crippled intentionally so it doesn’t cannibalize iMac sales. Apple machines now come with a fairly complete software bundle, which is good because third-party software support has been dropping over the last several years. Apple is still executing on the “what’s mine is mine, and what’s yours is mine” rule when it comes to third-party developers, and as a result developers are largely staying away from the platform. Reportedly HP is no longer ordering iPods, suggesting that that relationship is on the rocks as well. And we have discussed Apple’s litigation with retailers in a previous column.

Still, the Mac mini and iMac are the closest things to an appliance PC we have in the general market, and that market has been screaming for an appliance PC for some time now. Unfortunately, Apple is so horrible at partnering that it’s a wonder anyone resells their products. Even though the mini concept may be the future of the PC market, Apple probably won’t be driving it unless the company fixes its anti-partner and anti-sharing attitude.

HP, IBM, Clear Cube and PC Blades

Coming from left field we have HP and ClearCube (resold by IBM Global services) and the concept of PC blades. This takes the blade server concept that slammed into the market during the dot-com years and applies it to desktop computing. A blend of the thin-client market (now largely led by Wyse), blade servers, and traditional PC technology, it is one of the most forward-looking offerings available.

The product is far from consumer-oriented today, but it promises the user the closest thing to a utility-like experience in which hardware failures, for the most part, can be fixed by simply rebooting — the switch from a failed blade to a new blade happens automatically in the background. Data and the expensive bits are physically secured and fully redundant, so disasters have less likelihood of destroying the business, and theft is far more difficult.

A variant could be supplied by a service at some future point with a cable TV-like revenue model and user experience in which the user paid a monthly fee for secure, reliable and easy-to-use technology. Although these currently run Windows, there is nothing to say they couldn’t run Linux, and both IBM and HP have shown increasing interest in the alternative platform.

This has two problems: It currently lacks the common cross-vendor standards needed to make this move widely in either the consumer or corporate markets, and it is still relatively expensive to deploy (though it is vastly less expensive for most to support).

Microsoft: The Embedded Stealth Strategy

One of the most successful groups in Microsoft is the embedded group. It is most focused on vertical clients, it currently dominates the markets it serves, and its customers are some of the most content in any company. This, I’m sure you would agree, is all somewhat unexpected for Microsoft. If we were to use this group as a template, the future of the PC would be filled with unique flavors better targeted at what you wanted to do — rather than the Swiss army knife approach we have today.

In the living room you would have a TiVo-like PC — much like what HP built with its new Linux-based media hub — that did a few things very well and lacked even the ability to run most applications (and viruses) so that it could be on seven days a week, 24 hours a day without concern. For those who wanted to play games, another box, networked to the first, would be optimized for games much as the Xbox is now. (By the way, I wonder how many realize that the new Mac mini is similar to the Xbox in many key ways.)

In the office, the machine would be optimized for e-mail, general document creation and Web services (with a huge focus on search). In the kitchen, another variant designed for that environment would be used. The only general use product would be a portable computer, but one that would vary even more greatly in size and capability then what we have today.

The problem is that this is not currently the path that Microsoft is on, and recalling that HP went with Linux for its offering suggests Microsoft isn’t yet saying the things that at least one vendor wanted to hear. (Though, as mentioned last week, that may have suddenly changed, thanks to HP).

The vendor that will win will be the one that drives the standards and stays focused on the emerging customer needs in this segment. This probably says a lot about Intel’s recent customer-focused reorganization and the recent moves by IBM/Lenovo and HP.

Next week we will look at some of the actual forward-looking products, both from vendors you know (such as Dell) and some you don’t (such as Kloss) in our quest for the future of the personal computer.

Rob Enderle, a TechNewsWorld columnist, is the Principal Analyst for the Enderle Group, a consultancy that focuses on personal technology products and trends.


  • Without universal broadband at a cost of about $15/month for 45MB/s to the home, doctors, small business manufacturing/offices, I suspect that many other nations then the USA will take the lead in technology innovation and capitalist economy (possibly China without democracy/freedom).
    On the "personal" PC more innocuous, wearable, resilient, reliable self-diagnosis modular-repairable for owners/users (like swapping a battery today) and user and environment intuitive.
    The hardware will be more single-chip+IO with encrypted flash-memory (or write-once [throwaway] media) for security of OS configuration and encryption-keys and Bio-ID/Signature …, RAM/DDR2 (maybe flash) memory drives large enough to run a complete OS and all applications on the PC. Wireless/WiMax will allow applications/OS updates and downloads "on-the-fly" as environmental interactivity becomes more common place in the High-Technology countries like Japan, Korea, Norway, ….
    The USA is more like a "Technology country" (not high-tech) with a techno-deprived class of citizens … look at the education and health care systems across the USA as prime examples of the general/developing economy (very big is not very great when compared to most smaller High-Tech countries and the EU).

  • MacMini–yawn, yawn. I was a Mac user for years. I now love being on a WindowsPC. There are lots of new and exciting software apps out there. I wouldn’t want to give them up.
    Future of PC. Well, I along with a number of friends are finally moving from desktop to laptop. The laptop is definately the direction things are going.

  • First off.. There is the Linux fragmentation myth. Yes, there is variation, however this is application fragmentation, not system. The GUI is considered an ‘application’ on Linux, not an integral part of the OS. I.e., it is a *working* version of what Windows 3.1 was, but running on top of something that makes DOS look like punch cards. Because this sort of GUI level fragmentation does exist, you can do three things that are impossible with Mac, Windows, etc. You can go without a GUI, you can pick one that fits on the system you are trying to use, without completely sacrificing what actually runs on it and any changes needed to use one over the other are relatively trivial, compared to the complete rewrite required to go from MAC to Windows, for example. Yes, more consistency and some level of agreement one feature ‘might’ make things a bit easier in some cases, but that also limits the options as to ‘what’ features will exist in that single standard. That can either mean one bloated mess like MS Windows or an oversimplified nightmare that can’t do anything, without patches to add in stuff that suddenly splits the standard again. Right now the biggest issue isn’t what GUI you want, but how easilly each distrobution lets you install and manage things. That I would like to see a working standard emerge for. Instead we have several variations and even when you can convert between them, there is no guarentee that the RPM install for Redhat will install quite so easilly as one native to the version you use, which has a different package manager.

    Point is that yes, there are some aspects of Linux that could be simplified a bit to make it easier to install things and provide less worries about ‘what’ GUI you need to use something, but having the choice means I can run the damn OS on everything from a wrist watch to a supercomputer, without paying some company for a different version on every damn device.

    And that is where the issue with the appliance computers comes in… Yes, I have a VCR, a DVD player, a TV, etc. But I AM not real excited about the prospect of some clown selling me ones that cost $100-$200 more because it runs MS Windows Media Centre. Why? Because the damn thing does what ‘they’ want it to do, not what I want it to do. If someone comes up with a new codec tomarrow that could let me add twice the video and the same quality on a DVD, I might be able to use it on my ‘computer’, but short of buying a new appliance with it (which might take 4-5 years to reach the market), I won’t be able to use it on anything but my own computer. And if the codec is first introduced with software that converts existing DVD sets into a single DVD (which is the geek thing to do), then it might never see the light of day, because the MPAA will have to sue everyone using it first, while all the companies scramble to come up with proprietary, non-compatible variations with DRM, so instead of taking 4-5 yers to get a DVD appliance able to use the original, it might take 9-10 years for someone to buy enough licenses to sell *one* player that supports all of them. Heck, the one I have they left off old VCD standards, which means I have to buy a $100 application to burn true DVD or SVCD (which might not play either), even if I AM making my own movies…

    Frankly, I for one don’t want or need a dozen different devices that only do one thing. If I wanted to have to use one machine to search the internet, a second to record something on TV and a third to type letters, I would have gotten a WebTV, kept my old typewriter and still be using a VCR, which at least lets me bloody keep a copy of the show for later reference. But being able to refer to both the internet and TV programs on the same thing I AM typing up a report on is 1,000 times better than when I didn’t have that option.

    What scares me is that all these companies are so damn sure that appliance systems are the wave of the future, that they don’t get the real problem. That competence ‘is’ key. The Linux people are right about that. My parents have probably once a month talked about getting rid of our cable box and going back to the original cable. Why? Because every time they want to put in a VHS tape of DVD they get confused about what to do to make the thing work. At least once a month one of them picks up the wrong remote, accidentally turns something off or hits the wrong button. Figuring out what it wrong is simple, and should be especially for my father, who use to be the head of the manufacturing and repair facility for a ski resort. They get confused because they don’t know how or why those particular things work, not because they haven’t all been stuffed into one box with lots of pretty menus.

    Windows does the later with everything. You know the result? You get some clown going into a Windows user group screaming that he might have a virus and describing flickering video and streaky lines across the image. Actually knowing what realyl did this, I feel sorry for him, it means he will not only have to replace his failing video card, but do to the way most CRTs are made, it has probably damaged that too from continuing to use the bad card to post a stupid question to the forum. And that is just the tip of the iceberg when it comes to things ranging from the blatently dumb questions to completely legitimate ones that we can’t help them with, because Microsoft decided you should be allowed to change/fix. Now we are going to go to appliance systems where maybe 50-60% of the people using them have to call an 800 number, talk to a tech and be told, "Sorry, it just won’t do that, there is nothing I can do.", or, "Yeah, that is easy to fix, but you will have to send it back to us, since the device can’t be modified by the person that bought it."? Give me back my damn PC, then tell what address to send the shredded remains of all my proprietary, "You can’t do that, and even if you could, we refuse to tell you how.", software for it, along with the desicated remains of all the appliance PCs you forced on me!

    I AM seriously disturbed by the fact that companies want to operate on the principle, "People are morons, so lets not even bother to try to teach them how anything actually works." I AM also extremely glad that they don’t use the same standard for airplane pilots, architectural engineers or even, as bad as some are anyway, people with drivers licences… Most of the things in the world require basic skills to prevent you from electrocuting yourself, blowing yourself up, crashing, lighting yourself of fire or generally doing something stupid. I AM not real clear on why computers are considered to be something you should be able to, "just plug in and have it work." TVs do that, and people still can’t figure out how to use them properly sometimes. We need fewer clueless people, not more of them. And letting companies dictate what everyone uses, while catering primarily to the clueless is not going to improve the situation.

Leave a Comment

Please sign in to post or reply to a comment. New users create a free account.

Technewsworld Channels