I got into a great email debate with intrepid reader Chuck White (Director, Creative and Product Development at ADVANCE Recruitment Advertising Inc.) over last week’s article on Onflow. “Not another proprietary plug-in!” bemoaned Chuck. “What we need are standards like SVG, not more proprietary plug-ins.” SVG stands for Scalable Vector Graphics, and it is a proposed standard developed through the World Wide Web Consortium, the major standards body for the web.
So what’s wrong with standards? Some of you might remember my columns on VRML and how it is my position that it was actually the religious adherence to the VRML public spec that helped kill VRML as a viable 3D format. In my opinion, the open market, rather than open standards, is the best arbitrator of graphics standards. Want an example?
Ever heard of Flash? Now, ever heard of SVG? Unless you’re a graphics maven like Chuck White, chances are your answer is “Yes” to Flash and “No” to SVG.
And why is this? SVG has been kicking around for more than a year and a half and has the support of most of the major players in the field, including Macromedia and Microsoft. And yet, after 20 months, it’s still in the “draft” stage of development, it has no released player (outside of a few beta plug-ins), it has no tools, and it is not bundled with either of the major browsers. That’s problem number one. The development of complicated open graphics standards is slow.
Problem number two is that standards are, by their very nature, general. They are not geared to solving any particular market’s needs.
When I worked at Cosmo Software, our VRML player was more than 3.5 megs in size in order to be able to support the entire VRML spec. We realized too late that advertising would be a good market in which to develop products. But to support things like banners and page effects, we didn’t need a 3.5-meg browser. We could support the advertising market with a player one-sixth that size. But we were stuck with an albatross around our necks due to the spec we had to adhere to and a religious constituency, both inside and outside Cosmo, that wouldn’t hear of deviating from the standard.
Of course the counter argument is that these types of standards have the support of large software and hardware manufacturers. The big guns have already agreed to support and incorporate them into their products. Once a browser/plug-in/player is supported by Netscape and Internet Explorer (IE), the game’s over, right?
Well, not quite. Again we go back to history. Both Netscape and IE supported and bundled a VRML player in their products so they could each claim to be supporting open standards. Unfortunately, they did not support the same VRML player. And there’s the rub. Because open standards are just that: standards. A piece of text on a page. Standards are not a piece of code or a hunk of hardware. With something as subjective as graphics, it turns out there are many different ways to interpret the same set of guidelines.
For instance, the spec might say that the standard must support lights. But how those lights work, how they attenuate and reflect and refract, can and will be interpreted differently by different plug-in manufacturers. Which is exactly what happened with VRML. Content looked completely different in IE than it did in Netscape, forcing designers to design different content for the different browsers.
Multiply this by the fact that should there be bugs in the player (and believe me, there will be bugs), then a third and a fourth and a fifth version of the same content must be created to support a spec that was supposed to make things easier. (Bugs have to be fixed to support the open standards.)
Although this sounds improbable, it’s exactly what happened with VRML. Just ask anyone who tried to develop content using it.
Chuck, however, did make some good points. Chuck wants to develop great products, and he wants to know that the sand isn’t going to begin shifting under his feet if some company decides to change its format. He wants standards. We ended by agreeing that I would reconsider my position on open standards, and he would take another look at Onflow. Not a bad way to end a debate.
Until next week, keep it rich.