September 8, 2004

Linux cannot compete with Longhorn

Author: Joe Barr

Wait. Hear me out. I'm not saying that because Longhorn is a superior platform in any way. I'm saying it because Longhorn is not real, it's just the latest codename for the next version of Windows. As everyone knows, the next version of Windows is always the best operating system of all time: it's always faster, more stable, and more secure than anything the world has ever seen before. Comparing a real operating system like GNU/Linux against the marketing dreams of the malignant monopoly from Redmond is like comparing your geekiness with Helen, Sweetheart of the Internet: a futile and unfulfilling exercise.

Still, such false comparisons are one of Microsoft's favorite marketing techniques. Remember, Microsoft has never competed on the technical merits of its operating platforms. Not from the day IBM gave them a corner on the market until today. Given the quality of their products, that's probably a good thing for Redmond: DrDOS was a better DOS than MS-DOS, and OS/2 was far better, far more advanced than Win95.

Microsoft has always preferred to compete with its core competencies: FUD, astroturfing, false advertising, rigged benchmarks, funding "independent" studies, being channeled by submissive scribes and sycophantic industry analysts, and attempting to freeze the market long enough for them to catch up with the competition. Longhorn will probably fit into all those categories before it materializes, but its primary purpose is the latter. It's just another train-load of BS designed to freeze the market until the next version of Windows gets there.

In the beginning

Consider this quote from a ComputerWorld story, from May of 1994, by Ed Scannell and Stuart Johnston:

Microsoft Corp.'s consistently poor track record for delivering systems on time continues to disrupt developers' product development cycles and, ultimately, the purchasing plans of corporate information systems shops.

While this failure to live up to one's word is endemic in the microcomputer software industry, when a key provider of systems software like Microsoft does it, it often creates waves with crippling effects.

"If you believe their press releases, then you probably deserve whatever happens to you," said Vadim Yasinovsky, president of Clear Software, Inc. in Brookline, Mass. "If you don't learn from history, then you are an idiot by definition."

Those words were written because the due date for Cairo -- which at the time was the codeword for the next release of Windows, just as Longhorn is today -- had just slipped two years, and it wouldn't be available until 1996.

The great promise of Cairo was to be its Object File System, OFS for short. OFS was going to be a native database file system, similar to what IBM had been shipping in OS/400 for years. But that promise was trimmed from the feature list before NT 4.0 -- the release once called Cairo -- was launched in 1996. So instead of OFS, the Win95 UI became NT 4.0's most cherished attribute.

The undead file system

OFS may have been summarily yanked out of NT 4.0, but it didn't really die. It was merely pushed back into Microsoft's favorite OS. You know the one. That's right, the next release of Windows. In this case, that was NT 5.0.

Bill Gates did an interview with PC Magazine between the launch of NT 4.0 and NT 5.0. Here is the answer he gave when asked what had happened to Cairo. Notice the subtle shift: Cairo is no longer a product, not even one still over the event horizon. Instead it's a "vision." Or a collection of visions. It's difficult to know for sure.

The only thing of all that vision that's not in the marketplace is the file system and directory -- the rich file/system directory combination which is now part of the NT 5 product. We actually put a developers' release of that in people's hands in November in a professional developers conference we had... And so later this year that'll go into beta testing.

Having the rich storage system with the directory -- that was part of that Cairo vision. And so although a lot of the Cairo things have been done, that's the one that we're still working on. Today when you think about storage, you think about storing messages as one thing or addresses as another thing or user objects, machine objects as another thing. Anyway there's just too many ways that people are storing things and having to learn utilities and different security, different replication, different enumeration, query. Right now there's two grand unifications taking place: all the presentation is being unified around a sort of a super browser that takes over the shell, and then all the storage is being unified around a sort of a super file system that takes over a lot of those functions. The storage unification is the harder of the two, but they're both very important and will make the system more powerful and easier to work with.

Far be it from me to say that Gates was lying, but he was certainly spinning faster than the political pundits on cable news. Please note the fact, however, that the OFS went into a beta version of NT.

La plus ca change

NT 5 was renamed Windows 2000. In case you've forgotten, W2K has come and gone with nary a sign of the promised file system. Ditto for Windows XP. Some claim that XP -- the letters Chi and Rho in the Greek alphabet -- was the real Cairo. But if it was, it didn't have the new file system either.

But guess what has been hyped as part of Longhorn the last year or so? Correcto-mundo. With just a new coat of paint hastily slapped over it, OFS has become WinFS. And if you pay attention to Windows news at all, you probably also know -- or have guessed by this time -- that Longhorn's ship date has slipped and some features have had to be trimmed. Oh, no! Not the promised file system, again! But, yes, in fact. The centerpiece of Longhorn lo these many months,
WinFS has just been dropped from the production release of Longhorn, which is now scheduled for late 2006.

Luckily for the MS spin-machine, it was a simple cut-and-paste job to patch their excuses for not shipping OFS on its original schedule in 1994 and re-use them for Longhorn. I think they've really gotten the object-oriented concepts of inheritance and re-usability down pretty well, don't you?

Here's what MS exec Jim Allchin had to say about recently in a story in ENTNews by Scott Bekker about WinFS waving bye-bye to the cattle car just as the Longhorn train was pulling away from the station:

The first change is we're going to go hard-core for a Longhorn client in '06 and hard-core for a server release in '07...

In order to become more crisp about the dates, we've had to make some hard trade-offs....

What that means is, given the hard focus on date, that WinFS won't be in the client release in '06. It will be in beta at that time.

Pretty amazing coincidence, eh? That the OFS -- oops! --- WinFS is being pulled from Longhorn. But don't worry! They're going to put it in a beta. Just as Gates promised, about ten years ago. With promises like that, you have to figure that guy in 1994 was right, and trusting Windows users really are idiots.


Actually, I think my conclusion to this story was written last year, by someone else. Roger Howorth wrote in a story called Longhorn's long haul which appeared in PC Magazine UK last December: "Of course, the real question is why does Microsoft seem to focus on promoting Longhorn rather than products that are available now. Perhaps Microsoft hopes that if it focuses on next-generation technologies we might forget about its rivals of today."

Let's not make the job any easier for the spin-meisters at Microsoft by trying to compare a real operating system, available today, with whatever happens to be on the coders' desk at the witching hour two years down the road. It only serves to lend Longhorn some much needed appearance of substance.

Click Here!