Skip navigation

What's So Great About Longhorn?

Will its three best features be worth your cash?

I'm glad I'm not the guy at Microsoft in charge of Longhorn, the next version of Windows. The OS is starting to smell like the DOS 4.0 of the Windows world.

You remember DOS 4.0, right? It was "the DOS that no one except the vendors wanted." Its only real claims to fame were to support hard disks larger than 32MB, to give vendors a chance to talk users out of about a hundred bucks in upgrade costs, and to introduce a truly terrible text-based treeview of your hard disk called DOSShell—pronounced "DOS Hell" by those who've used it. Microsoft wrote DOS 4.0 and sold it as MS-DOS 4.0, then licensed it to IBM, which sold it as PC DOS 4.0. Lacking virtually any new worthwhile functionality, the OS was basically a "solution in search of a problem," and no one could figure out what the problem was. It was such a failure that Microsoft, in one of its more cowardly moments, blamed the entire debacle on IBM—"We didn't want this, they made us do it!"—but eventually the truth came out: Microsoft had the major design and development role.

I wonder who Microsoft will blame Longhorn on. Linux?

Ever since I started hearing about Longhorn, about 4 years ago, I've had trouble understanding its exact purpose. Then, a little over a year ago, Microsoft answered that question at its Professional Developer's Conference (PDC). Representatives told audiences that Longhorn would first ship as a workstation product (a Windows XP replacement) and that a corresponding server product would arrive about a year later. Redmond recently stated that it anticipates shipping the desktop version of Longhorn in late 2006, which would put it nearly 2 years away. That's a significant amount of time and would mark the longest period ever between Windows releases. XP shipped in October 2001, so we're looking at more than 5 years between the two releases. (Microsoft's spin-doctors will claim that XP Service Pack 2—SP2—is "basically a new OS," but that statement isn't really supportable. New OSs should offer a new feature or two, right? Then again, they also claimed that Windows 98 Second Edition was a new OS.)

At the PDC and thereafter, Microsoft has claimed that Longhorn's three big new features would be Indigo, Avalon, and WinFS. In case you haven't been following the Longhorn buzz, here's what those features will do.

  • Indigo. Indigo is essentially a newer version of the Microsoft .NET Framework. To whom will Indigo appeal? Developers, clearly. I'm not a developer, nor are most folks who would buy Longhorn Desktop, so at first blush Indigo doesn't seem like a big sales motivator for Longhorn. But ask yourself this question: Did Indigo's predecessors—COM, COM+, Distributed Com (DCOM), Dynamic Data Exchange (DDE), distributed transaction coordinator (DTC), OLE, remote procedure call (RPC), and a host of other acronyms—sell earlier versions of Windows? Yes, they did. Give programmers better tools and they'll build us better applications. That's what spurs OS sales. At this point, Indigo seems to hold the most hope for Longhorn. But let's be honest: Even an ultracool programming platform won't drive sales without a slew of new Longhorn-dependent applications, and ever since OS/2, developers have been reluctant to bet on not-yet-available OSs.
  • Avalon. Avalon is the new name for what was once called the Graphical Device Interface (GDI). Essentially, it's a new GUI platform. Yes, Microsoft wants to mess with the GUI yet again. Will that sell copies of Windows? Perhaps it will, but from what I've seen so far, it's nothing more than an Apple Computer "catch-up." It will be sad if all we can say of Longhorn in early 2007 is, "Hey, now it's possible to make a PC look like a Macintosh did in 2000." I'm not a huge GUI fan, and one of the largest arguments for niftier GUIs—to boost the game market—has disappeared. No one buys either PCs or Macs (or Linux systems, for that matter) expressly to play games: The Xbox, the Sony PlayStation 2, and the like have absorbed that market.
  • WinFS. WinFS stands for either Windows File System or Windows Future Storage, depending on which Microsoft source you talk to. What does WinFS do? As far as I'm concerned, it's the single most interesting thing about Longhorn, by far. I haven't got the space to really do it justice here, but in short, WinFS promises to radically change the way we store and manipulate data through "non-file items," and to significantly change the way we organize our files by largely doing away with the current decades-old method of organizing data—folders, that is. Instead, WinFS lets you view data as "stacks," or dynamic folders.
WinFS—when it finally ships in Blackcomb or whenever—is, at its base, a fairly simple idea. As some have observed, "It's crazy that we can use Google to search the entire Internet in seconds, but we have to spend 10 minutes searching our own hard disks." WinFS sits above NTFS and lets you associate attributes of all kinds—including attributes that you create—with files. It lets you tell the OS to monitor those files and perform certain actions when they change. It lets you create things that aren’t files, things that are actually stand-alone database records.

Sound interesting? Join me next time for more details.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.