nettime's dusty archivist on Wed, 6 May 1998 20:06:20 +0200 (MET DST) |
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
<nettime> contribution for a net.history |
A Piece of the Action It's bad enough when technical squabbles percolate from the bowels of the technology sector into the mainstream press. But when a mere war of words turns urgent distinctions into more of a tactical squabble, all the wrong people usually win. These days, those who can least afford to lose are being splayed over the increasingly banal brawl of "Java vs. ActiveX." As early as a year ago, the average observer confronted with this dualism might infer an obscure courtroom battle between Juan Valdez and the makers of adult undergarments. These days, it's threatening to mean much, much less. Feuding between Microsoft and Netscape is nothing new, but at some point, the sundry issues of competitive corporate positioning collapse into fundamental questions of access: users' access to software, software's access to computers, and companies' access to our attention. The untold story of Netscape's ascent was their serpentine throttling of distribution - of content, plug-ins, software, everything. It's a tale worth telling, if only because the looming success of Microsoft threatens to mark an end to this stillborn era. But the telling begs a rewrite of the industry's adolescent history... Already the darling of Silicon Valley after a few short months, Netscape's browser had attracted interest from Sun, Macromedia and Adobe, all of whom promptly announced deals with Netscape to integrate their respective technologies (Java, Acrobat, and Director) into future versions of the browser. And every software start-up on the net wanted in on the action, too. The folly in the Valley stems from these original deals. We'll put forward this entirely plausible scenario: Orchestrated behind the scenes by John Doerr of Kleiner Perkins, the deals with Netscape were more about building a "keiretsu" (which in our Japanese dictionary translates as "Old Boys' Club") than about creating a new marketplace for small-time software developers. The deals were easily inked and duly submitted to PR Newswire, but some of the partners - Macromedia in particular - ran into "execution" problems. Namely, their code wouldn't be ready for Netscape's 2.0 release. The Netscape Plug-In API offered a convenient solution: a standard way for programmers to hook their apps into the browser. When Macromedia finally released Shockwave, the code could be plopped into the browser at any time. Even better, the Plug-In API would be published, so all the other developers could theoretically whistle their ways towards similarly auspicious goals. Posting the specs on an obscure part of their site, the boys at Netscape considered the problem solved. And they were almost right. Overnight, there were a hundred plug-ins, with more on the way all the time. One can imagine Clark, Barksdale, Doerr, marketing VP Mike Homer, and a grinning, drooling Andreessen around the boardroom in Mountain View company. After an appropriately dramatic pause, the CEO intones: Homer, who had a bit of experience with marketing operating systems at Apple, no doubt outlines a strategy straight from the pages of an Apple business plan: The roster of plug-ins listed on Netscape's website grew, as did the number of newspaper and magazine articles proclaiming that "the browser is the OS." And if you asked any plug-in developer how business is going, you'd hear the same reply... "Not much yet, but we're going to get CNET to use our plug-in on their website, and downloads will go through the roof." At the major websites, the lines trailed around the corner for plug-in pitchmen touting an end to the limits of HTML or "even better compression than Shockwave." Most smart sites sent them packing; the ones that didn't soon found their customer support desks flooded with email and phone calls from perplexed users, whose browsers presented them with the cryptic error message... At Netscape, though, all was going according to plan. Plug-in developers were sorted into two groups, the ones who got money and the ones who gave it. To be sure, revenues were going to exceed expenses in this new profit center. Played right, the protection money paid by ISVs, "independent" software vendors, could dwarf the haul from Netscape's search engine payola scheme. The concept had ingenious simplicity: Step 1 - Open the technology. Step 2 - Close the distribution. Step 3 - The tricky part: Convince everyone that distribution is more open than ever, but only if Microsoft would stay out of the way. The plan worked so well that Sun decided to use the same strategy with Java. While Java applets appeared to get around the distribution scam, the real action sprang from the APIs, implemented in native code that only Sun and Netscape can distribute. Macromedia would create the multimedia API for Java, and Adobe would beef up its graphics capabilities. You, too, could enter the lottery for the price of a Java license, with the jackpot of having your code built right into the language - but most of the poor saps who ponied up the cash would turn up as ghosts in the Java Virtual Machine. Meanwhile, in Redmond... Microsoft was finally coming to its senses, seeing in the net a crowded, raucous stadium, with the teams already on the field well into the game, and a sold-out notice at the box office. Of course, they owned the place; they consulted their sky-box seating for perspective. The code boys drew up feature sets, the top brass started to look for holes in the business models of the key players. It didn't take long to discover the Silicon Valley keiretsu's dirty little secret... The solution - eliminate the distribution scam, and put all the net software developers on a level playing field. Easier done than said! Microsoft had long been promoting OLE (object linking and embedding) as a way to embed applications within other applications, but being able to stick a spreadsheet into a word processor somehow lacked the appeal of a chat window on the Playboy page. Best of all, for Microsoft, OLE ran on all the platforms MS cares about: Windows 3.1, 95, NT, and, oh yeah, the Mac - sorta. Presto change-o - OLE morphed into ActiveX, a part of Internet Explorer. As far as technology goes, it was a major improvement over the hacks Netscape peddles. But then, Netscape has never been known for its brilliant R&D. The real impact of ActiveX rests in a feature called AuthentiCode, which is a way for developers to digitally "sign" their ActiveX controls, guaranteeing that they haven't been tampered with. This allows the browser to automatically install the "trusted" ActiveX controls as needed. Most users saw it as a convenience feature, just one of the many reasons why they prefer Internet Explorer. But when Netscape 3.0 is introduced, it's the one feature Netscape doesn't copy. It didn't take long for plug-in developers and content sites to catch on to the benefits of ActiveX controls vs. Netscape plug-ins. Start-ups with unlikely prospects, like FutureWave, were suddenly on an equal footing with Macromedia, as website developers took the plunge and started using ActiveX controls on their web pages. At Netscape, the dreams of tall dollars from plug-ins began to fade. But worse news was on the way. Microsoft found a way of merging ActiveX and Java, allowing developers to build ActiveX controls in Java, and letting the "trusted" Java code gain greater access to the system and the ability to mix native code with Java. While developers puzzled over what this means, and editors saw their next year's worth of "ActiveX vs. Java" cover features angling toward meaninglessness, Netscape and Sun saw the writing on the wall. In Microsoft's world, software will play together without licensing dollars changing hands first. No plug-in bundling deals, no auctioning off a place in the Java API to Macromedia. The Internet Old Boys' Club found itself in big trouble, as the web of deals weaving KP's keiretsu together come apart. The solution, while not foolproof, exercised the home turf advantage: a war of disinformation. "Microsoft wants to kill Netscape plug-ins" was discarded as a dud, and Netscape instead settled on the "ActiveX is great if you've got a PC" line, which warms the hearts of the Mac and Unix crowds, who aren't exactly drowning in a flood of plug-ins anyway. How about Java? "Microsoft wants to kill Java"? Sublime. The "ActiveX vs. Java" meme still gained momentum, disingenuity be damned! Security is the favorite rhetoric of Sun. "Native code is inherently dangerous and insecure," they said (the corollary: only Sun and Netscape can be trusted to put native code on your machine). This got the academics to stop looking for security holes in Java, and concentrate their efforts on ActiveX. Meanwhile, the MIS boys started beefing up their firewalls. In Redmond, Gates was beside himself. A recent convert to Java, the smear campaign was the final straw. So, once again, it's off to war. But in web infowars, casualties are more likely to play dead than stay dead. The irony of this theatre of combat is that a company like Microsoft can be bullied by circumstance into doing the right thing and still see their tactics blow up in their face. With only the tiniest hint of a smirk, they release a Netscape Java plug-in, 10 times faster than Sun and Netscape's implementation. They hire away the Mac Java people from Natural Intelligence's Roaster group, and put them to work on building a better Java for the Mac. But there's no such thing as benevolence anymore - only cheaper and more cynical marketing ploys. Can the cool buzz of satisfied users douse the shrieks of quivering netheads and panicky competitors quick enough to calm the spectre of a bespectacled Big Brother, even one that brings good things to life? The web's technology underwriters will keep blasting away until the question is settled. And in an age where few people care whether the clip is half-full or half-empty, but rather who owns the gun, it might take a while to clear the floor of spent shells. [S U C K], 6 December 1996 --- # distributed via nettime-l : no commercial use without permission # <nettime> is a closed moderated mailinglist for net criticism, # collaborative text filtering and cultural politics of the nets # more info: [email protected] and "info nettime-l" in the msg body # URL: http://www.desk.nl/~nettime/ contact: [email protected]