Where is my "Made for Linux" computer system?
by Doug Dingus | Osopinion.com | November 2005
"When the technology community considers Linux, we know what sets it apart from the competition. But if you were to ask the average consumer, few could give you a legitimate answer other than maybe it doesn't cost anything. (However there will always be those that look in vain for the free beer everyone keeps talking about.) Editorial contributor Doug Dingus offered the following opinion piece to osViews which proposes some interesting ideas to help differentiate Linux to consumers by way of hardware rather than just software."

Lately I found myself doing a lot of thinking about Apple computer. Frankly, the Mac running OS X seems especially attractive as compared to the average PC, no matter what operating system it happens to be running. Ask around, and you will find a lot of Mac users (not to mention more and more PC users) who will agree with you.

Clearly the value proposition Apple brings to the marketplace goes beyond simple price and feature comparisons.

The question on my mind, is how much of that can apply to the growing Linux marketplace? I have been running Linux for long enough now to wonder if the ability to run W.I.N.E.'s win32 API is even necessary anymore? As a matter of fact, I've come to
believe that hardware targeted for Open Code would make more sense than that which is commonly referred to as "open hardware."

Apple produces both hardware and software. Recently they also began to successfully leverage Open Code toward building more value into their products in a way that is both efficient and better suited to their relatively small, but loyal, share of the market. This synergy between hardware, operating system, and core application set allows the company greater control over the user experience than is currently seen in your typical PC. At the core, this is what Apple users will typically say is why they chose Apple rather than going with the flow. Put quite simply, Apple computers generally just work? with a minimum of hassle compared to most other computers. Isn't this what most Linux users want also?

The Open Source community has made great strides in the last few years in both quality and quantity of software being made available for use today, but we have failed to produce an environment that is simply useable as it is powerful and cost-effective. Perhaps hardware is an important, but overlooked piece of the overall Linux usability solution.

An important part of the Linux experience today is choice. You can choose any part of a Linux computing solution. Your choice of hardware, desktop environment, display manager, supporting applications is wide open and generally regarded as a good thing. Free software is, after all about freedom, and I agree strongly with that core philosophy, and I suspect a fair number of you reading this will agree as well.

All of this choice does have its downsides however. I believe many users do not want too many choices, or do not want to learn to make them. The majority of computer users want many of the tough choices that matter made for them so they can go ahead and make the easy ones as the need arises. Clearly this is a service worth paying for as both Apple and Microsoft continue to show, each in their own way. Why not do the same for Linux in order to increase the value proposition the platform offers to the mainstream computing public while preserving the core values that allow users to do what they want as their skill and or needs grow?

Linux has an identity crisis of sorts today. What is it really? Is it the kernel? Maybe it?s a particular distribution? Are all Linux distros the same? If so, then why so many? If they are different, which one do I choose? These are all questions easily answered by people who understand what Open Source software means, but they all are showstoppers to people who don't. Rather than bother with the answers to these, they are likely to choose one of the more clear and simple choices, such as OS X or Windows.

I believe Sun has identified this as an issue as well. They chose to sidestep the identity problem by not using the Linux name in their newest products. Linux becomes "Java Desktop System", which is fine for them because they have established Sun and Java in the marketplace to a degree that makes sense. However, Sun is not a brand associated with consumer level products. Java is, to a degree, but the relationship between the two is less than clear for anyone but corporate buyers.

We need a reference Linux platform that is both inexpensive and feature rich. People need to be able to buy a "Linux System", or "Linux Computer" and understand what that means. What makes a Linux system different from the Apple and Microsoft ones you can get today?

Marketing operating systems is tough because most average users don't understand what an operating system really is, or why they should care which one they have. Once they start asking questions along these lines, the sale quite easily boils down to which one is easier to buy, with the answer being the same one my neighbor has more often than not. This is not going to get us anywhere.

Apple and Microsoft have another thing going for them that Linux does not; namely, they stand out clearly. Apple is about cool looking computers that are easy to use and that offer popular features right out of the box along with the ability to run most any application you need. Microsoft computers are generally easy to use, are the standard because almost everyone has one and they run almost everything.

What are Linux computers? Better question: What should Linux computers or systems be? Here are a few of my ideas to get the discussion started:

Feature Rich

Rather than support all those add-on internal hardware cards, offer most features built right into the machine. Each machine should come with TV Video In and out, ethernet, USB / FireWire, DVD, 3D graphics, hardware encryption, 5 channel audio, or at least good stereo audio with multiple inputs and outputs, serial and parallel ports. The idea here being to ditch all the legacy stuff in favor of a nice base platform suitable for most home / hobbyist tasks. Take as many shortcuts as needed to build the features people want without support for things nobody cares about.

Limited but sensible upgrade choices

Everybody wants some degree of expansion, so make that easy. IDE storage devices are easy enough, so provide for those. Internal cards? Forget those. If the onboard feature set is reasonable, most folks won't need them. For those that do, they can get a PC because Linux runs on most everything. Ram should be easy to change without taking the whole thing apart.

Clear and useful user experience from power on to power off

This means no bios checks or mystery black screens. Users should see a welcome screen that gives them what few choices they may need. A nice gui here for system startup, some testing or advanced options that make sense, point 'n click. People should know what is happening in a clear comfortable way at all times.

The hardware available today will last most users a long time. This used to not be the case, but it is now. A good solid Linux System should do most things... well, for a long time. Remember, a lot of folks buy new and start over. We should take advantage of that to get started. Once people are using the machines, it will be possible to offer services and upgrades that make sense in the future. Maybe trade in your old model, get your environment moved to a new one ready to take home again. Given the hassle of getting a new machine, this might be an option most people would welcome if they had the chance.

Problem areas

Which distribution to use?

Perhaps a choice of them combined with the work of projects such as Desktop Linux, or Ximian could work as well as one distribution would. Remember, most of us already using Linux have made the jump. We can make any choices we want. This effort would be about bringing more folks into the fold to our benefit as well as theirs. Given a sensible choice of base platform, our ability to choose would not be reduced much, if any at all. It would likely increase as the number of applications and users does.

Who makes the hardware?

Could be one supplier, or many depending on the nature of the design and market demand. Suppliers could compete on form factors, or on bundled external hardware supplied with the system. They could also compete on upgrade services or support policy, for example.

Design agreement / feature creep general design issues

Community designed hardware is likely a tough proposition compared to software. Nailing down base specifications will be subject to a lot of wish-lists and wistful thinking. Prototypes will need to be funded and tested. Perhaps these things can be overcome if enough of the community sees enough value in this type of platform.

Advantages

Uniform base environment for testing and new software development Clear target for application developers, consistant user experience, clear brand identity for the mass market, potential for high consumer value perception, platform favorable to free / open source software development without having to play catch up with new development Potential to be DRM free to a high degree (Better now for this than later.)

Desktop Linux will continue to be an enigma for most people until we solve the identity problem. Perhaps a hardware solution is too expensive and other solutions will prove more worthy. I just can't help but wonder about the positive effects such a development effort might yield. ::

"This article was syndicated under osopinion.com's Open Content License."