stating the obvious archives | about

If you build it, will they come?
Dec 04, 1995 :: Michael Sippey

If you read Wired's recently published history of Java, you know its origins. Some folks at Sun wanted to build a language for consumer electronic applicances (a.k.a. Personal Digital Assistants), that would enable content providers to distribute applications (and the data that drives them) over wireless networks. They wanted it to be a "simple, object-oriented, distributed, interpreted, robust, secure, architecture neutral, portable, high-performance, multithreaded, and dynamic language." Dizzy yet?

Well, to make a long story short, Sun saw the bottom fall out of the PDA market. And basically had nowhere to go with Java, and no clue about the WWW, even though most of it was running on Sun workstations. Until (BANG!!!) all of a sudden Sun woke up, got the net.religion in a big way, and Java was pronounced our savior.

Now, there's been a good deal of hype about Java over the past several months. The Wired article is just the tip of the iceberg. Business Week's cover story on the "coming software revolution" featured Java, and described it as a Microsoft killer. George Gilder praised Java in a recent Forbes ASAP article, calling it a "fundamental shift in the history of technology." For Gilder, Java is another step in the inevitable "hollowing out of the PC."

People "in the know" say Java will be everywhere, and soon. Netscape has licensed Java technology to include in version 2.0 of their Navigator. And since Netscape has the browser market locked up (so the argument goes), users will start screaming for Java apps to run on it. Right?

OK, reality check. I downloaded Hot Java (and Netscape 2.0). And I played with some Java apps. And it usually works. And it's usually pretty cool. You get animation (cool). And you get "ticker tape" (cool). And you get neat text effects (cool). And you get calculators and rudimentary spreadsheets and 3d molecular models you can drag around on your screen (cool, cool, cool).

But Sun is going to need to have a LOT of patience if they want to really see the PC get hallowed out (and Bill Gates topple in the process). Because Java (at least right now) ain't nearly as addictive as the stuff it gets its name from.

  • In their Java language white paper, Sun talks about Java being "a system that could be programmed easily without a lot of esoteric training." But they go on to say how Java is based on C++, and how wonderful it is that they've removed poorly understood "features" of C++ like "extensive automatic coercions." Earth to Sun (oooh, I like the sound of that): wake up. Java is not "easy." HTML -- now that's easy. Visual Basic -- that's easy, too. Make it like those languages, and you'll have geeks round the world beating a path to your door.
  • There are no compelling Java applications. C|net has a Java app on their front page. It scrolls computer industry headlines in a "ticker tape" fashion. I could scan those same headlines in about 5 seconds on a standard page. And how long does anyone stay on their front page anyway?
  • It ain't truly cross-platform yet. If you check out any of the Java application archives, there are always two versions: one for Hot Java and one for Netscape Navigator. Now wait a minute -- wasn't Java supposed to solve all of those problems?

Now, I'd like to see some decent competition to Microsoft's hegemony as much as the next guy. But I have a hard time believing it's going to come from Java -- at least not in the next few years. Yes, there will be specific applications of Java (probably in academia, maybe in the financial world). And yes, sites like c|net will hype their "Java enabled" sites.

But until there are tools that are easy to use, a platform that is truly independent, and a whole HOST of killer applications, Sun ain't going to make a bundle of money on this thing. They are going to need a good deal of faith in (and undying support for) their vision, and the money to back it up. Because the stakes are high. And because this is just the opening volley in what will be a long battle, and an even longer war.

And because I'm pretty sure that James Gosling doesn't come cheap.



Other pieces about client-side software: