New Dreamweaver, amazing HTML4 support!!

True story. I got hold of the specs for a secret Kevin Lynch project. As JavaScript improves more can be done with it — even though the company is famous for its browser plugins, that’s only one type of technology. This project’s been designed to bring full rich-media interactivity and authoring support to HTML.

If you’ve used Flash Pro you’re familiar with timelines, which can set up straight ahead animated presentations, or branching interactive paths triggered by user events or system events. In these internal Dreamweaver plans I saw timelines and encapsulated behaviors ripped out from existing motion-graphics tools and applied directly to the new abilities offered by upcoming browsers.

“But don’t all the different browser brands have different capabilities, while a plugin adds predictable capability to any browser?” I objected. Another set of specs held the answer: Dreamweaver was to include its own conformance-testing database, and would alert you whenever a feature was not offered by one of your target browsers. Matter of fact, you could specify the minimum browser you needed to support, and the authoring interface would conform itself to display only those possibilities supported by your chosen audience… you could make an experimental site which required a particular recent browser, or build for a general realworld audience… your choice, Dreamweaver would handle the browser-variance details for you.

I was flummoxed. Giving similar visual timelines and development amenities to HTML tooling, what effect would that have on plugin adoption? Weren’t we cannibalizing our own business? If JavaScript added rich-media interactivity, wouldn’t it kill off plugins?

Ripping off timelines and property panels and the rest… why would Macromedia introduce an authoring tool which would compete directly with Director and its Shockwave plugin…?

… yup, this all happened a dozen years ago, when HTML 4.0 was but a gentle rising glow upon the eastern horizon, back in the days when one browser vendor did things one way while another browser vendor did things another way and you still had to deal with all those dumb hicks out there who didn’t use your own favorite browser. It’s from a time very much like today, just a bit more historical.

Macromedia already had Backstage Designer, an HTML editor which interacted with serverside CGI components, whose native database files could export out as HTML. Adobe had Pagemill, very popular… Microsoft had bought Vermeer. None of them addressed the new rich-interactivity support in HTML4, much less the browser variance which went along with it.

It was pretty gutsy for Macromedia to build an HTML authoring tool with standard multimedia authoring features. The timelines and behaviors were particularly controversial inside the shop, because they directly paralleled existing Director authoring conventions, as well as those of the nascent Flash. There was a special Macromedia site, dhtmlzone.com, which showcased pages using the new browser technologies.

In a sense, Dreamweaver provided the first cross-browser JavaScript framework, uniting all the competing browser brands and conflicting versions, with behaviors which were themselves written in JavaScript and were inherently extensible by anyone.

The oddest thing, though, was that the timelines and behaviors and extensibility and what-all good stuff ended up not mattering as much as how “Dreamweaver just leaves your code alone”. This turned out to be the big feature which drove adoption. Dreamweaver was the first HTML authoring tool to use HTML itself as its native file format — the first to respect that an author might overrule the program. This made the difference.

Those glitzy animation timelines were eventually removed from Dreamweaver, and JavaScript extensibility has always had a hard time developing a market… good thing we never tried to refactor for later rich-media specifications like SMIL and HTML+TIME and such. The fancy features weren’t the real draw for using Dreamweaver 1.0 for HTML4… it was the respect for a creator’s workflow which eventually made Dreamweaver the closest thing to a “standard” HTML authoring tool.

The relevance? Even back before Macromedia joined Adobe, there was implicit acceptance that changes in customer needs overruled existing investments — even though Dreamweaver could “kill” Shockwave, the project went forward. And Adobe has a lot more emphasis on cross-channel publishing than Macromedia ever did.

DHTML seemed like a good bet to make. The flashy multimedia features were heavily hyped, yet never really proved practical for most content developers. Macromedia also continued to invest in providing all browsers with predictable advanced capability, by plugging into their various extensibility architectures. Both Dreamweaver and Flash were very successful, fulfilling complementary jobs.

But I still remember how gut-wrenchingly shocking it was, reading David George’s secret timeline specs while on CalTrain to the Redwood Shores office, back towards 1996, looking at timelines and behaviors and stuff coming out of Director… people were saying that HTML4 would kill Shockwave, and here we were helping it along…. ;-)