I spent last week in muggy New Orleans, home to Windows Hardware Engineering Conference (WinHEC) 2003, a surprisingly exciting show given its rather stodgy and seemingly limited name. But WinHEC has become the place to be for people such as me, who are interested in upcoming developments in the PC industry. So I was treated to a long list of briefings and sessions detailing upcoming Windows client (Longhorn) and server (Blackcomb) releases, Next-Generation Secure Computing Base (NGSCB--formerly Palladium), and various other technologies that won't see the light of day for months or in some cases years. You might wonder why future Windows developments are relevant to you in your day-to-day job, given the problems we have keeping today's systems running. But WinHEC wasn't just a crystal-ball look at the future. The show made obvious some technology trends that will affect all of us. Here are three topics from WinHEC that I think are important.
Windows XP for the Long Haul
No one from Microsoft explicitly said so, but during all the talk about Longhorn--which, incidentally, won't ship until 2005--one thing became clear: XP isn't a short-term OS. Nope, we'll be using XP for many years. And although we'll see various service packs (Service Pack 2--SP2--is due in late 2003), Quick Fix Engineering (QFE) releases, support for new technologies (e.g., Bluetooth, USB 2.0), and other updates, such as Windows Media 9 Series, the core XP OS will have a retail shelf-life of at least 4 years, far longer than the previous several Windows versions. This news is good for several reasons. First, XP will be more supportable--it's a known quantity and the base for several product editions (e.g., XP Professional Edition, Tablet PC Edition). Second, users will grow comfortable with this Windows version as it becomes pervasive across most business and consumer systems. Third, for the first time since the mid-1990s, we might have a chance to master a piece of software in a time when competition and the Internet boom have caused OS and office productivity suite development to surge, with frequent product revisions.
Longhorn Installation Technologies Will Be Staged over Time
I also learned a lot about Longhorn last week, and if you're morbidly curious, you might want to check out my new Road to Longhorn article on the SuperSite for Windows ( http://www.winsupersite.com ), which details some features and changes we can expect in the next Windows desktop OS. But don't be turned off by the long-term nature of Longhorn--one piece of very exciting Longhorn technology is relevant to all of us in the short term. That technology is Longhorn's Windows Preinstallation Environment (WinPE), a relatively new imaging-based setup routine that Microsoft says will dramatically shorten Windows installation time. In fact, the company says that unattended Longhorn installations should take less than 15 minutes.
Why does this news matter now? To prepare system makers, IT administrators, and anyone else who needs to roll out massive numbers of Windows desktops for the Longhorn version of WinPE, Microsoft will provide an interim WinPE version later this year that works with both Windows Server 2003 and XP. The interim WinPE will provide more modern (i.e., not DOS-based) tools for rolling out Windows now and help system makers prepare their factories for the changeover when Longhorn does ship. The interim WinPE version, which will ship this fall, will support 32-bit and 64-bit Windows.
Secure and Private Computing
Another technology that you should be aware of is NGSCB (or Palladium as I still like to call it). Palladium is one of the most misunderstood technologies Microsoft has created, largely because of the misinformed opinion pieces that have sprouted up on the Internet. But Palladium isn't evil, it isn't about Microsoft being Big Brother, and it certainly isn't about coercing control over your computer. Instead, Palladium, like many Microsoft technologies these days, is based around a few core principles, which the company's marketing folk call pillars or tenets (I provide these terms for humor purposes only).
In Palladium's case, those principles are security, privacy, reliability, and business integrity. "From a process perspective, we're changing the way we're building software," Mike Nash, vice president of Microsoft's Security Business Unit, told me last week. "It's cultural." Nash compared Palladium to the early days of NT, which he said addressed problems that, in some cases, didn't even exist when the product was released. Palladium consists of four components: Clear process isolation, in which you can have applications running inside the Palladium context; sealed storage, which determines who or what can access which data; secure hardware paths, so that the system knows that users, rather than a snooper tool, are sending keystrokes or mouse movements; and attestation, a system that verifies things are what they seem to be (e.g., that the applications you run are the right applications, the user is the user, and the systems are the systems).
So why worry about Palladium now? In an increasingly connected world, keeping private corporate and personal data is getting harder and harder. So Microsoft expects businesses to be the catalyst for Palladium-based PC sales (Palladium requires Longhorn and special PC hardware). Looking forward, it's not too early to begin examining Palladium and thinking about whether this solution might be appropriate for your needs. If Palladium is successful, we might one day take for granted data, personal, and corporate security, as we do with electricity or running water. I suspect many companies want to see that day hastened.
If you're interested in more Palladium information, please let me know--I'm not sure whether it's too early to begin discussing specifics. But one thing is clear: These technologies are coming down the track like a speeding freight train, and they're going to appear more quickly than seems possible. Don't be caught unaware when everything suddenly changes.