Specifically, Cascading Style Sheets, which allow fancy typographic and layout control without <FONT> tag infestations or over-intricate tables. It's a reasonable revolution for reasonable people. One day recently, though, the list caught fire.

Relatively speaking, anyway; against the backdrop of mild-mannered exchanges on typographic niceties, the slew of 20 messages in as many hours was a veritable flamefest. And it was all centered on three little words: Microsoft. Dot. Com.

"Microsoft will never be a leader in the Internet world," one angry message said. "Microsoft always has a way of redefining the standards," said another. The responses from Microsoft staffers were better argued than these attacks, but regardless of which side you're on there's plainly a raw nerve. To many, Microsoft is like a capricious elephant, smarter than it lets on, flattening little companies even as it trumpets its achievements.

The www-style exchange had started when Microsoft employee Scott Isaacs posted a message to the list to deny that Microsoft is working on a "proprietary superset of HTML." Isaacs, who among other things represents Microsoft on the W3C's HTML and CSS working groups, assured me that Microsoft is "fully committed to open standards."

They're certainly fully committed to playing the part, and it's starting to be convincing. They do throw their own widgets (for example, ActiveX) on top of standard technologies, but they are indeed implementing and helping to create open standards. Microsoft Internet Explorer is definitely leading the way in CSS support, for which we should be grateful. Imagine "Microsoft StyleX Style Sheets" or new <MARGIN> and <LINESPACE> and <DROP-CAP> tags, or yet another proprietary layout format from Adobe to add to the mess. Shudder.

Carl Cargill of Netscape Communications wears the term "standards weenie" with pride (though he offers "standards wonk" as a more sober alternative). In terms of sheer acronym emissions per minute he is, in my judgment, a champion. He clearly knows the territory, and lives in Netscape time. His job is to keep Netscape in tune with relevant industry standards, including CSS.

Netscape is an elephant, too. It's a thinning, nervous elephant, but it's still bigger than anything else in the browser jungle. Through sheer market dominance Netscape could continue the strategy of adding new proprietary tags to HTML for a long time. Cargill says he doesn't think that's the way to go.

He tells me a story: Many years ago, while the rest of the world was standardizing on TCP/IP (the Internet's present low-level lingua franca), certain companies were pouring money down a network technology hole called OSI. Never heard of it? Me neither. Yet Cargill says it cost the industry $4 billion. Even the most protective capitalist would find that story motivational. Ignore open standards at your peril, it seems to say. Cargill really believes that the tight feedback loop enabled by the Net will allow individual corporate innovation and industry-wide standardization to coexist and even converge.

Not everyone is as optimistic as Cargill that the mass market will reach open-standards nirvana. Jon Bosak, being Sun Microsystems' Online Information Technology Architect, is no standards slacker either. Bosak is active in promoting DSSSL, an ISO standard stylesheet language much more powerful and complex than CSS. He contrasts the two standards not only in terms of their technical capabilities but also in terms of vulnerability to manipulation.

A collaborative, consortium standard like CSS develops very rapidly, shaped by its actual use in the marketplace and frequently amended by its committee. People want tint percentages or non-scrolling backgrounds or better measurement units? We'll add them. The rapid feedback of this development cycle is where vendor manipulation or, to use a nicer word, tweaking, comes in. Netscape learns that users want scripting control of their style sheets, so they create a solution -- JavaScript Style Sheets or JSS -- in hopes that it will later become part of the public spec (which, in fact, it likely will). The consortium-style standard starts small and grows in essentially unpredictable directions.

A traditional international standard like DSSSL, in contrast, starts big and very well defined. Its life mission is to grow into its clothes. If the architecture is bad, this can be a disaster (perhaps a $4 billion one). If it's good, though, it's very powerful, especially for large organizations that can't take the risks small ones do. "Two or three years from now there may be just a few dozen DSSSL users in the world. But those users will have names like Boeing and Philips and Westinghouse and General Motors and NATO..."

There will be no "DSSSL for Dummies," Bosak predicts. He's probably right. And perhaps part of what that means is that the top-down standards model may never dominate the consumer market.

While a regimented ISO-standard HTML might not have developed at the breakneck pace we've seen, it might also never have produced the <BLINK> tag. The consortium process definitely has more warts, and the industry may not continue forever in its current motion toward open standards. But while it lasts, it's a welcome change. Instead of features designed to make us captive consumers, we have features, like CSS, which just get more useful as more vendors cooperate. Now that's style.