If there's one issue raised by the Talkbackers for which Microsoft deserves some grief, it's Redmond's overselling of the Win32 API behind NT, 2000, and XP as a form of investment protection from the pain experienced when we weaned ourselves from Win16 API dependence. But those perceptions were created before Microsoft faced the most challenging public relations fiasco in its history: the barrage of reports about security and reliability problems in its operating systems and applications.
Chairman Bill Gates launched his "Trustworthy Computing" initiative to clean up the mess. But considering the depth and breadth of Microsoft's software portfolio, it's unrealistic to expect any initiative to cure all that ails Microsoft users overnight. While progress has been made, problem reports still seem to surface weekly and will continue to do so until Microsoft can put its legacy of untrustworthy products behind it. While some view the initiative as the work of spin doctors, Microsoft is taking the trustworthiness of its products very seriously. It has to. Except for a few defections, its customers seem willing to put up with the transgressions as long as something's being done about it.
That expectation takes two forms. First, the company must do everything it can to publicize and correct problems as they arise. To that extent, Microsoft hasn't let the threat of more ugly public relations nightmares (the sort that can bring down other companies) interfere with this goal, and that is to be commended. The company has been relatively forthcoming. IT managers tell me that most successful exploits of Microsoft's software come as a result of users or administrators who don't heed vendors' warnings.
Second, and more germane to the topic at hand, Microsoft cannot under any circumstances launch a new generation of products that repeats the unsavory track record of its current generation. This second challenge will be exacerbated by a defining characteristic of next-generation operating systems and applications: seamless interoperation with systems and services inside and outside firewalls by virtue of standard Web services protocols.
So new are Web services that the standards for their deployment are yet to be finished, and won't be for some time. Microsoft had to make some tough decisions for the sake of the predictability of its own environments, regardless of what they are connected to--Web services, the Internet, or something we haven't thought of yet.
If Microsoft fails to meet either of these expectations, the customer defection ratio could change drastically. Gates and his management team are well aware of the stakes.
In a confirmation that Office 11 (which is expected to ship in mid-2003) will not run on Windows versions earlier than Windows 2000 Service Pack 3, a Microsoft spokesperson cited a variety of technical reasons for the decision One reason: Microsoft's latest software installer, version 2.0, can be found only in Windows 2000 SP3 and Windows XP. According to the spokesperson, this installer "greatly simplifies the installation process for both consumers and large corporate end users."
Although installation hasn't been a serious issue for me, I can see how a more robust update and installation infrastructure can contribute to the security and reliability of any platform, which was the second point that the spokesperson made. "Windows 2000 and XP include improvements to the auto-update features as well as firewall and encryption enhancements. Office 11 will leverage those strengths of the operating system. To the extent that Microsoft wants to make the most secure products it can, it's fair to say that that this decision [falls within the auspices] of the Trustworthy Computing initiative."
Fair enough. But, will Office 11 rely on other services in the operating system that are connected with XML, Web services, or compatibility with advanced computing initiatives such as pen computing via TabletPC or voice recognition? Microsoft will only say that there's no Web services connection.
But even if there isn't any connection, the expectation that Microsoft would freely support the interoperation of all its applications with all its operating systems for the foreseeable future is absurd. Given the choice of compatibility versus security and reliability, the choice is obvious. Microsoft knows that it has to make a clean break in order to turn this corner. If we, in one breath, continue to criticize the company for its security and reliability problems, while with the next breath refuse it a chance to make that break and clean its own house, well, that just makes us a bunch of whining hypocrites.
Does this mean that Microsoft should be absolved for its failure to produce something more secure, or that the company should be allowed to extort money from us before it provides the security we were entitled to in the first place? Probably not. But, there will be more to future versions of Office and the Windows operating system than security improvements; and, for some of us, those enhancements will be worth the cost of upgrading.
As for people like ZDNet Talkbacker George Johnstone, who apparently doesn't want to upgrade, and wrote that "If it ain't broke, you have to upgrade it whether you like it or not," that's not true. Microsoft recently disclosed information about the lifecycle of its products and the support for them. If you're happy with what you have and it works for you, nobody is forcing you to upgrade. I know people still running Windows 95 who are perfectly happy to reboot their system several times a day.
Not only that, but older Office products will continue to function on newer versions of the operating system. That means that people who are satisfied with the current version of Office but want to upgrade their operating system to one of the newer editions of Windows (for the sake of stability and security) can do so without having to buy a new version of Office.
What do you think? TalkBack to me or e-mail me at firstname.lastname@example.org.