SFNL Home Page


 The Forums

 Product Reviews

 Let’s Fight Sp@m!

 NetBEUI and Win XP

 PayPal Donate

 Letter Mail Donate


Scot’s Newsletter

Windows and Broadband Information You Can Use!

Get free email delivery of this newsletter!

Recommend Scot’s Newsletter to a Friend!

October 4, 2001 - Vol. 1, Issue No. 13

By Scot Finnie


  • Build a Better User Experience ... or Get Out of the Way
  • Subscribe, Unsubscribe, or Change Your Address.

    This issue of Scot’s Newsletter is a one-time, temporary departure from the usual format. Next issue the newsletter will be back to what you know, and hopefully love -- or at least like. This week I'm bringing you something that's been on my mind for a good long while now. I think the time is right for it, and I hope you agree. Whether you do or don't, I welcome your thoughts and reactions.

    An Open Letter to the Computer Industry:

    Build a Better User Experience ... or Get Out of the Way
    I'm issuing a challenge to the people behind the computer industry to open their minds and begin thinking differently about quality of experience. I'm not talking about user interface, but about the entire end-user computing experience. The way people think about and interact with personal computers. What they do with them, what they try to do. What works. And what doesn't.

    A computer is only a tool, but at the moment personal computers are very poor tools indeed. That's because the industry has invested little effort, money, or time to improve on the basic idea. The industry's development power has been focused on features, function, and form instead of usage, experience, and raising the bar on what a computer can achieve for people. A PC's most important strength -- its ability to be programmed to do a wide variety of tasks -- is also its Achilles' heel. It's a jack of all trades and a master of none. In order to be a master, it needs to be programmed, very carefully, by masters.

    During the early 1990s industry people bemoaned the lack of a "killer app." Here we had this powerful tool all gussied up but we were bored with what it could do. The Internet came along and filled that void. But the underlying problem remains. We've expended far too much effort honing the innards of this significant machine, and nowhere near enough doing the really hard work: Dreaming up important things for it to do.

    There's another problem. Most computer users, whether corporate or consumer, are more mindful of what can and does go wrong with their PCs -- be they Windows, Mac, Linux, Sun, Palm, or whatever -- than they are of the powerful and productive things they could be doing with them. In other words, we're so focused on avoiding trouble that many of us compute defensively, sometimes to the point of being frozen and unable to try new things.

    There's a pat response the industry likes to pass out on this point: By its nature software has bugs, and there is no such thing as perfect software. That is a true statement, but it doesn't give commercial software-makers a license to churn out sub-standard code that leaves customers in the lurch. It's just that sort of thinking that the market has rejected. The heady days of ever-increasing sales are over, perhaps forever. If there's a silver lining, a severe loss of profits is the only kind of correction the industry ever would have paid attention to. Here are some of the lessons I believe the industry should be learning.

    Course Correction
    In the early 1980s the personal computer was the cause celebre of a computing revolution that wrested control of computer power away from an exalted few and placed it into the hands of the many. By 1981 anyone who had the money and inclination to buy his or her own computer could do so freely, and there were even choices, such as the Timex Sinclair, IBM PC, Apple II, Radio Shack Model I, Commodore, and many others. It was a quintessential American story about freedom, as corny as that sounds. But I was there, and that's what happened.

    Now a generation later, some of those players who rode under the banner of delivering computing to the masses appear to have forgotten what it's all been about. Or perhaps they never knew. The 1980s saw the beginning of a new way to market and distribute computer capacity. The first decade of the 21st century should be another turning point. Frankly, we're long overdue for a sea change.

    The sad truth is that the computer industry has completely lost its customer focus -- even lost touch with a big chunk of its customer base. Over the last 10 years we've gone through wave after wave of consumer versus corporate focus, and neither one is the answer. When you look at the way companies like Dell and many others segment their markets -- Home/SOHO, Small Business, Corporate, Government, and so on -- it becomes clear that computer makers are completely out of touch with who is buying their wares. For example, I work for a corporation, out of my home, and I run a small business. That's not all that uncommon. That Dell expects me to self-select who I am when I buy a PC from them its Web store is ludicrous. Every time I price out a PC I feel alienated by that heavy-handed market segmentation. But it also doesn't work because I always price out in three different "identities" to see which one offers the best mix of features at the best price. There's no long-term advantage to the PC maker treating its customers this way. The computer industry is lying to itself and its stockholders.

    I believe those companies that successfully pursue a new direction must start with an understanding of what customers truly need. And I don't mean what they need next month, next quarter, or even necessarily next year. That kind of vision is mighty hard for the average public company to gain any traction on because it's focused almost exclusively on raising the financial bar (or not lowering it so bad as Wall Street thinks) *this* quarter. So, entrepreneurs take note: This could well be the best time there is to dream about technologies, products, and services that spur an order of magnitude shift in the direction for computing. Do you have what it takes?

    Perfecting the PC
    It sounds laughable now, but I can tell you one thing computer users truly want: Software and hardware that works. Microsoft coined the marketing phrase "It just works" early in the Windows Me development cycle. Well, Windows Me doesn't just work. It's the most banged-out for corporate bucks operating system Microsoft has produced since DOS 6.0. But the notion of making PCs work far more reliably without bother and hand-holding is, I think, a strong avenue to success. The industry thinks this is impossible, but then, the mini-computer companies thought millions of PCs around the world was a stupid idea too.

    Suppose just for a minute there were a radical new PC and operating system out there; one that ran thousands of existing applications, never needed rebooting, had many fewer hardware and software conflicts, eschewed bells and whistles for crash protection, and was wonder of wonders, easy to use. Is it Linux? The Mac? Windows XP? GeOS? One of Larry Ellison's network computers? No, in this little fantasy interlude, this new people's computer is different, something not based on existing technology or today's thinking. It's more than just a promise, it's the real deal. The question is: Would you buy it? Do you think other people would?

    But to PC makers and operating system programmers that's a very tall order. The Windows operating system is one of the most widely distributed pieces of software in the world. The sheer numbers of installations and hardware, software, networking environments it runs in dictates that tens of thousands of unforeseen problems will crop up for each separate version of the operating system. Many will never be solved or addressed. And so far, we're only talking about the operating system. Application developers and hardware makers issue a regular stream of products, each of which will run into its own set of troubles out in the real world of computing. When you really stop and think about this, it's a wonder our computers work at all.

    One of the keys to delivering an operating system that works is having a strategy that extends well beyond the software. Operating systems are also about drivers, application compatibility, standards, user experience standards and expectations, and a lot more. Microsoft backed into the operating system business by buying DOS to fulfill an IBM contract supporting the original IBM PC. Over most of the years since it's done a darn good job of being what a good OS provider must be -- an industry leader. But in recent years, Redmond has given up on a lot of things. The company is far more short-sighted, more self serving, less supportive of pure technology, more interested in promoting only what it thinks it can dominate. That shift has hurt the industry.

    To be a hugely successful operating system leader for the long term, your company must expend thousands of people hours and millions of dollars building consensus among hundreds of partner companies who make software, hardware, drivers, firmware, and so on. It means having the talent and doing the research to pick the right direction for the industry, and then working persuasively and in an aboveboard way to get other industry players in step. It entails hand-holding and compromising, and many standards and practices must be dictated. The Linux community is facing the exact same problem now that Microsoft, Intel, and hundreds of other companies faced in the 1980s and 1990s. To do it well, a company must lead by example above all else. Some other will attempt to undermine you just because you are the leader. Programming a complex desktop operating system is a giant undertaking, but leading the computer industry is much more than a project.

    There's another key to building software products that don't break and that interoperate well. It is simply making those goals paramount. The term planned obsolescence has been used to describe products designed to fail after a certain period of time so that companies can resell newer but little better products to those same customers. The Detroit automakers were, I think, rightly accused of that in the late 1960s and 1970s. What caused it to change was the American people voting with their wallets by purchasing slightly more reliable Japanese vehicles. Japanese engineers realizing that was the impetus for their success in the huge North American markets redoubled their efforts to increase product longevity and reliability. Later they did the same with gas efficiency and safety, and eventually performance. Now Japanese automakers are building bigger cars and trucks. Detroit has countered with drastically more reliable and more fuel efficient vehicles. They waited too long, and some say they will never recover (but in the spirit of disclosure, let me say that I own Ford Motor Company stock).

    I am not saying that Microsoft and Intel and the rest of the personal computer industry have purposely made computers unreliable or in need of upgrade. I am saying that they haven't done a thing about the fact that they are. The people who are experienced users, or who have experienced users helping them take care of their PCs, have far fewer problems. When they do have trouble, they're usually able to diagnose and make repairs quickly and easily. What that means is that the logic processes for dealing with computer problems exist. They just need to be encoded into the operating system. Again, studying how people use hardware and software is the key to making that software much better. The "usability testing" that most software companies do really just tests the new features -- not things like problem resolution, troubleshooting, and the like.

    In Microsoft's defense, it has made reliability and software compatibility a mission-critical target of Windows XP. It's too early to know whether it truly succeeded on those points, but I have a hunch that it did, at least by its own measure. But Microsoft copped out on hardware compatibility. Corporations are far less concerned with hardware compatibility than consumers are. My point: Windows XP is no next-generation operating system. It's evolutionary from Windows NT. In many ways, the company is just now finishing off NT. Windows XP is a step in the right direction, but far more is needed.

    The Existing Players
    Most people believe Microsoft was never a nice company, that it was always a bad boy. I disagree with that view, but there's a sickness about the spirit of the company now that has caused a lot of talented people to leave it, and caused many of its loyal customers to question that loyalty. Microsoft was once, however, a company with a pure vision about the future of computing. Yes the vision included Microsoft holding a dominant position -- and raking in a lot of money. But that's business in the real world. Leading an industry well and making pots of cash are not mutually exclusive endeavors. The problem may just be that Microsoft reached the end of its vision and had nowhere to go from there.

    There is one company that from its inception has typified the kind of company the industry needs now. In the mid to late 1980s Apple Computer's operating system technology was several notches above IBM's or Microsoft's. The Macintosh was just a better computer. The only people who doubted that were the ones who hadn't tried it. Part of the reason for Apple's technical achievement was that had a closed environment, one where it created both hardware and operating system. That gave it control, and it's true that as a result some types of problems common under Windows are rare on the Mac.

    But Apple made several costly mistakes. It forgot the first lesson of computer history: Proprietary hardware is the path to eventual ruin. Selecting the Xerox PARC-derived, mouse-driven user-interface principles behind the Mac OS showed significant vision. But Apple was terrible about attracting and fostering a software developer community. What's more, the company proved unable to update its operating system to take advantage of emerging technologies. And, of course, Apple was in competition with third-party Mac hardware makers. Later it even competed with the software makers. Just as soon as the Mac was firmly established, Apple should have sold the hardware license to as many takers as it could -- allowing the price of the hardware to erode markedly. That would have spurred a growth in marketshare, which in turn would have attracted both independent hardware and software makers.

    Instead, Apple is essentially still trying to execute on its original plan -- which requires the market to prefer elegance over value with its hard-earned dollars. Something else, the Macintosh user experience is only marginally better than the Windows experience. And that advantage has eroded over time. Apple's hardware and software is no less subject to bugs, conflicts, and the need for upgrades than Windows software and hardware. And there are few significant things you can do with a Mac that you can't do with Windows, whereas there are far more applications for Windows than for the Mac. So, I don't think we need Apple Computer now. But we may need a more business-savvy 21st Century version of Apple.

    What about Linux? Some hail open-source software development as the answer. But I have some doubts about that. It's essentially software by committee, and there's no incentive for participants to react quickly to changing market demands. The Linux open-source movement seems to be something of a contradiction, but "react quickly" is still the operative phrase. I'm hopeful the Linux community can create a truly Windows-competitive desktop OS. It has made some strides but isn't there yet. We'll have to wait and see.

    First and foremost, PCs aren't dead. Not even close. Whether a computer slips into your pants pocket, slides into your briefcase, or sprawls across your desktop -- it's still a personal computer. The notion that every PC user has a single boxy computer affixed to a desktop may be a trend on the wane, but I wouldn't bet the farm on that. The steep decline in sales over the last year doesn't signify the end of the PC era; it signals that the market leaders have become fat and complacent. They've forgotten how to focus and innovate. Don't get sucked into a negative mindset.

    The market drives technology, but it can't take the industry there on its own. The leading companies must divine the proper direction. Microsoft, Intel, and other industry players have in the past proved themselves perfectly capable of that kind of thinking and action. But Microsoft nearly missed the Internet revolution, and it apparently believed itself immune on antitrust issues. Don't get lulled into a false sense of negativity. Personal computing opportunities have never been more prevalent -- especially for any company that has vision. None other than Apple's longtime evangelist Guy Kawasaki agrees.

    In this business, vision comes from studying what customers need and then buying, stealing, or developing technology that fulfills that need. Right now the market doesn't quite know what it needs (so what else is new?), but smart minds can put it together.

    Today's PC isn't the be all and end all, the last in the line, the most evolved and most refined of the species. Keep an open mind. IT managers have fewer dollars to spend, right? Consumers are fed up with problems on their PCs? What if you could deliver a personal computer that breaks 70 percent less often? You think there isn't pent up demand? There is. We all just don't want more of what we've already got. Who needs a new headache?

    Finally, there many uses for PCs that the computer industry hasn't even begun to consider or develop. Some require technology we haven't developed yet. But at least a few others just haven't been properly developed, packaged, and marketed. Communications is one area that we've only scratched the surface of. But there are many others. Hunt out the new technologies.

    Those computer companies that apply these thoughts and approaches stand a good chance of reaping serious rewards. This will sound naïve to many, but forget about market segmentation. What's good for consumers is good for corporations, because personal computers that are more reliable and that make people more productive will provide serious return on investment for corporations and institutions while naturally appealing to consumers. I don't have the answers; if I did I'd be busy retooling my garage to develop the next-generation operating system and PC. But I do have some suggestions, both about standards and practices, and about areas that seem to me to be fertile ground for development:

    1. Forget the upgrade treadmill. Users don't need another version of your product. Selling us an "upgrade" every 12 to 24 months as a business model that feeds P&L statements isn't going to wash any longer. It's worked, more or less, for 20 years, but the market is sick of it. Microsoft has introduced landmark OSes every four to five years. I count Windows XP and Windows 2000 as one operating system, Windows 95 and 98 as one, and Windows 3.0, Windows 3.1, and WFW 3.11 as one. (NT was a test-tube OS that was really a network operating system in the 90s. It's only now bearing financial fruit on the desktop.) When you look at it that way, Microsoft should charge more for an extremely well tested new version of Windows that doesn't need SP1 to make it worthy every four or five years. Everything in between should come for free, or as part of an relatively inexpensive subscription upgrade package.

    2. Plug and play 3.0. You who would make hardware, software, operating systems: Make sure it works together before you put it up for sale. Make reliability, compatibility, and interoperability the number one goal. Hardware and software should plug in seamlessly, without the user having to think twice -- the way you would plug a lamp into a wall socket. It should plug in the same way that every other hardware or software plugs in. Making this happen requires a lot of companies holding hands. Firmly. Or maybe Steve Jobs has been right all along?

    3. Users should control of what's on their PCs. Uninstalling software, whether it be for hardware or applications, or something deemed to be part of the operating system, should remove 100 percent of themselves, reverting every single change they made on installation or by user configuration as they leave. It should be optional to do a light uninstall that leaves many settings behind. This should be a strict rule that every company in the computer industry follows. The leading operating system maker, above all others, should be held doubly responsible to meeting this standard.

    4. Don't create interfaces geared mostly toward first-time users. Computers aren't like toasters. The people who use them may use them all day long. You had to learn to drive a car, right? You weren't born knowing what the shift handle does, which pedal is the brake, or how to make the cabin warmer. Yet, step into any car now and those things take almost no time to figure out. People can be taught the basics. But hitting 90 percent of the audience over the head with overly pronounced novice-oriented control surfaces should be legally actionable. Nevermind the fact that there's no surer way to alienate your existing customer base. If you're leading an industry, or trying to, this should be obvious.

    5. Computers are transformers. Current operating systems user interfaces are incredibly mundane. There's no reason why a computer can't take on a special purpose interface designed around specific usage applications. The problem is identifying worthy usage applications, and that can be tough because new technology and user trends dictate important new applications -- such as those of the past, including spreadsheet analysis, word processing, publishing, browsing the Internet, downloading and manipulating music files, and managing digital images. Are these activities more the province of applications? Not necessarily. There's opportunity in identifying entire operating system configurations based on usage. It's yet to be proved that works, but I'm convinced that someday it will.

    6. Consumer isn't a dirty word. Consumer applications have been barely touched by software and hardware makers. Digital cameras are hot, right? Hotter than PCs, perhaps. You have to have a PC to make get the best of one though. Wireless networking, household controls (video monitoring, power controls, security), shared applications, shared broadband access, and even neighborhood communications are just some of the many things people will use PCs for in their homes. The whole consumer thing doesn't look hot right now, but that's temporary. One thing that's needed is intense customer focus on form factors that blend with the home environment without compromising expandability.

    7. Broadband applications. If you're in this for the long haul, at some point broadband connectivity will reach critical mass because the pent up demand for broadband Internet connectivity is huge and growing. Whenever the Internet connectivity industry finally figures its way out of the paper bag and starts delivering in a meaningful way, computers could take on significant new roles and uses. Timing is everything. It is very easy in this business to have the right idea three years too early. But is there something related you can do until then?

    8. Networking and peer services. The Internet turns the whole world into an unruly, uncooperative, unbelievable local area network. Distributed processing, peer storage and retrieval, and network communications are just some of the already realized applications that global networking will present.

    9. Communications. Love it or hate it, the single most important application to rise in popularity over the last several years isn't the Web browser or AOL's client software -- it's live chat messaging, more commonly known as instant messaging. If you have any doubt whatsoever that computers are at their very heart a communications medium, you should look at this phenomenon more closely. People like communicating with their PCs: Email, Internet newsgroups, and instant messaging are some of the more visible examples of that truth. Note that all three of them are very low tech solutions, which also happen to be unique among pre-existing media. That's an inspired formula that bears inspection.

    10. Radical new input devices. It's a huge challenge that may not be adequately solved for decades, but the main thing holding back several important computing advances is the way we input information into computers. Someday keyboards and mice will probably seem as archaic as punch cards do now. But the companies that invent the technologies and sell the products that spur that transformation will be household names. Voice recognition is one idea, but there are others. If we can achieve nearly effortless interface with our PCs that's also highly portable, huge changes will follow. The killer applications will follow and could be a dime a dozen.

    Do you have a comment, response, or suggestion? Send me your two cents.

    The Fine Print
    If you like this newsletter, I need your help spreading the word about it. Please share it with friends and co-workers, and encourage them to sign up! It's free.

    Visit the new Scot's Newsletter Forums.

    Subscribe, Unsubscribe, Change Email Address or Message Format
    You can unsubscribe at any time; I don't believe in captive audiences. The website subscription center is the easiest way to manage your Scot’s Newsletter subscription. Changes take only a minute or two. You must select your message format — Text or HTML — even for address changes or unsubscribes.

  • Unsubscribe
  • Unsubscribe Help and Options
  • Subscribe
  • Change Email Address or Message Format (HTML or Text)

    To help with the cost of creating and distributing the newsletter, I accept contributions via PayPal and Letter Mail. For more information on donations:

  • Sign-up for PayPal (if you don't already have it)
  • Option #1: Donate via PayPal
  • Option #2: Donate via Letter Mail

    Send comments, suggestions, or questions about this newsletter. Don't be bashful about telling me what you like or don't like. Send emails related to editorial content (only) to

    Please address advertising inquires (only) to:

    Sign-up for PayPal.

    Support this Newsletter by Donating Today.
    Or donate via Letter Mail.

    How to Link to Scot’s Newsletter

    Copyright © 2001-2007 Scot Finnie. All Rights Reserved.
    Ten Myths About Copyright Explained.

    You are subscribed to Scot's Newsletter HTML EDITION as: $subst('Recip.EmailAddr')