Should Microsoft dump Windows? Yah, right!

Should Microsoft toss out the millions of lines of code in Windows and start over from scratch? Most readers say: No way! Even those who think Windows needs improvement agree this is not the way to go.

COMMENTARY--How strong is your imagination? Can you see a time when Microsoft will toss out the millions of lines of code in Windows and start over from scratch?

Not me. And neither could my colleague David Coursey in a recent column, which was sparked by Bill Gates's testimony at the antitrust remedy hearing.

Coursey also cast a critical eye on the need for competition for various applications, such as the browser, word processor, and media player. "And what would this 'killer Web browser' be, anyway? Is there something IE is missing? Yes, you can add all sorts of features around the edges--as Netscape has done--but none of that adds up to 'killer' for me," he wrote.

THE POWER of the installed base was evident by the outrage expressed in many of your posts. For some of you, the very suggestion that Windows development would be altered to any significant degree was like a dark shadow passing over the computer screen.

"INSANE! The cost would be astronomical for business and industry," Dan Anderson argued. "It's not like Microsoft couldn't make radical improvements to Windows's functionality, reliability, and security, if motivated to do so. Yell louder! My chief complaint with Windows is reliability and security. We are better off pressuring Microsoft to do the right thing in upcoming releases. Tens of thousands of voices will influence the product evolution."

"Starting over would have us end up pretty much where we are now: competing ideas and philosophies, pulling us in different directions until somebody 5 or 10 years from now decides that, 'Gee, we need a new OS.' That's too much work to get a product, let alone convert all hardware, software, and users to a new paradigm. Ain't gonna happen!" Mike Drabicky wrote. "Microsoft has done us a major service by having a common platform to which others can design their hardware and software. This has done more good for the computer industry than many are willing to admit. Microsoft has also done us a major disservice by being so arrogant that they would squash any and all competition, by whatever means."

ON THE OTHER HAND, some of you thought the idea had merit--especially Linux and Mac owners. In a long post, Lynn Withrow offered a list of goals for such an effort. Still, even those who claimed to be ready for big changes couldn't quite make up their minds where and when to begin.

"I'm all for dumping Windows and starting over, [but] it doesn't have to be from scratch. Maybe Microsoft could buy back the rights to OS/2 Warp from IBM and use that as a starting point," Ray Reyes suggested with a note of nostalgia. "BUT FIRST, let's get Congress to pass a software lemon law, shall we? Otherwise, we'll end up with the same old s---. Outlaw this business of software 'licensing' and hold the developers to some decent standards!"

Windows programmer James Chaldecott speculated that Microsoft's .Net architecture could allow Windows applications to run on another operating system. "The important bit is that the way the code is hosted, and the fact that it isn't compiled to native code until it is on the end user's machine. This means that it is NOT linked directly to the OS."

"If Microsoft can get everyone to write their applications ONLY against the .Net APIs, then they can switch out the OS underneath, and the applications should not care (in theory)," Chaldecott observed. He also pointed TalkBack readers to the Mono Project, which aims to clone .Net for other platforms. "Is MS getting ready to pull an OS switch on us? It would be nice, but probably not."

MEANWHILE, A NUMBER OF YOU wrote that improvement was still needed for most software applications--and pointed to missing features found in Microsoft's former and current competition.

For example, David Shealey missed WordPerfect for engineering documents, after his company standardized on MS Word. "Some Office items work reasonably well, but Word is still the most unfriendly word processor in common use today. Until MS wakes up (unlikely) and adds the functionality of WordPerfect's 'reveal codes,' it will remain a difficult-to-use program. It is a crying shame that MS has trampled a far-superior word processor on its path to world domination!"

The general notion that applications may have reached some ideal state gives me trouble. Some of you agreed with Coursey that there's no "real" difference between competing products. If that's the case, then why worry and why bother?

For some folks, one small detail--such as WordPerfect's reveal codes--can make a big difference. Most of the time when comparing programs, you'll find a large collection of such small features. Conversely, others could benefit from applications with fewer  features.

SO, MUST PROGRAMS offer some "killer" feature set in order to have worth? I seem to have a growing stack of applications that provide basic contact and calendar PIM management functions. For many people, that's good enough, and they appreciate the integration with their e-mail program or handheld. They also like the price--zero.

However, I find lots of value (that I pay plenty for) from a program designed specifically for that task. The software provides the speed, record flexibility, and layout features I prefer.

Is that difference "killer" enough? For me (and certainly thousands of others), it is. Somehow, that doesn't seem to be enough of a reason in today's market.

David Morgenstern, past editor of eMediaweekly and MacWEEK, is a freelance editor and branding consultant based in San Francisco.

What do you think? Should Microsoft dump Windows and start over? TalkBack to me below.