Is it finally time for 3D online?
Interest in 3D technology for the Web is picking up, and this time it's no false alarm, says VRML co-creator Tony Parisi.
VRML veterans like Parisi last month marked the 10th anniversary of the language's first commercial implementation. And after a decade of waiting for a computer graphics Godot, they're used to encountering skepticism when they herald the imminent eminence of Web 3D.
Bodies littering the Web 3D landscape include that of Microsoft's Chromeffects effort (shelved in 1998), Adobe's Atmosphere title (killed in November), and Intel and Macromedia's joint venture to popularize Shockwave 3D on the Web (which dissolved along with other Intel Web 3D alliances).
In 10 years of turmoil and tried patience, both VRML and Parisi have changed. VRML, after achieving ISO standardization, in recent years has been reborn, under the auspices of the Web3D Consortium, as an XML-based ISO standard called X3D. Parisi has kept the Web 3D religion with a San Francisco start-up called Media Machines, whose clients include the U.S. Navy and Joe Firmage's ManyOne portal.
Last month, Media Machines said it had acquired, in an all-stock transaction, Cincinnati-based Virtock Technologies, which adds a Web 3D authoring tool to Media Machines' Flux 3D software development kit, publishing tool and player.
Parisi spoke with CNET News.com about why the Web is finally ripe for 3D content, why the Web browser is yesterday's paradigm, and what the architects of the next Internet interface have to learn from gamers, among other topics.
Q: You've said before that 3D was about to take off on the Web. Intel and Macromedia said it was going to take off. When I look at the market for Web 3D software and content, I'm reminded of what they say about second marriages--it's the triumph of hope over experience. Why is this next period going to be any different from the last few false alarms?
Parisi: The experience we're having at Media Machines is that we're no longer seeing the primary focus on the government client or the corporate client where we've been making our living, but on a daily basis we're getting inquiries from people who want to deploy their content on the Web. We're doing very little in the way of promotion, and people are asking themselves, how can I deploy a product showroom over the Web? How can I, an independent game developer, put together a game title and not have to cut off a limb in terms of payment?
So it's Web stores and gamers--the usual 3D suspects?
Parisi: One of my customers is building a science and education portal where objects are being represented in a photorealistic way.
This would be Joe Firmage. The last thing CNET reported about that was that he'd bought Media Machines.
Parisi: This was back in the summer of 2003, and we had signed a letter of intent for the acquisition. After several delays, due to logical reasons, both parties concluded we should stay as independent entities. But they're one of my top customers. They're licensing the Flux technology for distribution, and they're using it for the 3D content engine for their rich media portals.
So one thing we're seeing in very immediate steps is serious early adoption.
What do you mean early? It's been 10 years.
Parisi: It means it's not someone kicking the tires anymore. They're deploying it, they're buying it, they're doing something with it. We're past the demo stage, but we're not seeing widespread adoption yet.
This is the third swipe at this. The first was VRML, which was too much too soon. We had visions of all these different applications you could build with the technology--many of which are now being deployed--when computers didn't support 3D. The software libraries you needed weren't installed, so along with the 2MB download of (Intervista Software's 3D player) WorldView there was 10 megs of (Microsoft's) DirectX and Direct3D to install--over a 28.8 connection.
Now everyone has 3D on board if they bought their machine anytime in the last four years. And now you have the bandwidth; you have broadband.
There have been several proprietary formats for 3D on the Web, like Macromedia's Shockwave 3D and Adobe's Atmosphere. What happened to those?
Parisi: They were too little too late. In 1999, the Internet bubble was in full swing, and the multimedia giants introduced these extensions to what they already had. Times were good, it was the middle of bubble, budgets were big, and people had time to look at those things. But the technologies were pretty limited. They were weak, not as fully functional as VRML.
So that's the too little. The too late was that Internet bubble burst and those budgets just evaporated. The foundations of those businesses went away.
Now we're in a position where the timing couldn't be better. Every user has a computer that has 3D. Most people download some kind of game, or download music, or download a chat client--things that are outside the sphere of the Web browser. Right now we have these proprietary, 3D gaming universes, 3D chat worlds, like Linden Labs' "Second Life," and they're analogous to the Prodigy and Compuserve of a decade ago.
So you're saying the new 3D technology is the equivalent of Netscape, in 1995?
Parisi: Yes. You will see a disruption of those proprietary businesses because there are now open ways to do the same kinds of experiences. For the user it's free, and for the developer it's almost free.
How are you taking advantage of what you see as this looming opportunity?
Parisi: The dynamic I see is a lot of usage of rich media client applications, music players, video players, chat clients, integrated chat and gaming clients--what they spell is the end of the era of everything being delivered through a browser. That and the difficulty of deploying applications as plug-ins because of extra clicks required in SP2. We're seeing the hegemony of the browser coming to an end as the sole distribution vehicle and the user interface. And we're seeing an appetite for richer experience.
So how does this play into the market for 3D?
Parisi: We believe 3D is going to be a better organizing principle for information. Take the social network. Go on Orkut and you've got all your friends, and the interface is absolute (junk). It's tiles with pictures of your friends. There has to be a better way to organize that. My
What's the best 3D interface you've seen? Can you point to something in particular?
Parisi: My favorite, which was presented in a VRML tech conference in 1998, was a visualization of the workings of CERN's particle accelerator. The accelerator is a big honking thing, and you could fly through it, and within that you had a visualization of what was happening at the physics level. You could see the particles streaming through. You had the abstract integrated with the concrete. It's about being able to combine the verisimilar with the abstract.
So to get this to work, you need to have a new generation of people who are thinking this way. Where are they going to come from?
Parisi: That generation is coming out of the art schools knowing tools like 3D Studio Max, Maya, LightWave, and many of them are learning how to take their design skills to the Web, much the way Photoshop jocks learned how to develop for the Web when their jobs demanded it.
The next generation needs to have been raised on 3D user interfaces. And you've got that with gamers.
Parisi: That's the best example of people raised on 3D interfaces. When you move one of your characters around in a 3D game, it already knows how to walk. In VRML, you could walk through the same space and not have any sense that you had any identity. No one did that level of design. You wouldn't hear the footfalls. So that has to happen.
The bulk of the interface design will come from (the) gaming community, with additional innovation through these proprietary 3D chat worlds. But in most of these chat rooms, there's nothing to do! You see someone's avatar, and they're picking their nose. It's a piece of glitz attached to text chat. In an application like "Everquest," you have exactly the same environment design and you're there to do something. There has to be a purpose.
How does Adobe's acquisition of Macromedia bode for 3D?
Parisi: It's obviously a boost for multimedia in general. You have two powerhouses getting together with serious multimedia tools, one coming from the document print world, the other from the interactive CD-ROM world, trying to make the Web a broadcast medium. That's huge. It's continued evidence that Macromedia is relevant as a business, that this is not a sideline to information technology or the Web. More and more, it's going to be the main gig on your computer. So to me, that's all goodness.
But isn't there a tension between Adobe's vision of the world and what Macromedia is trying to do?
Parisi: Yes. Is the WYSIWYG (what you see is what you get) document culture of Adobe going to win, or the interactive broadcast culture going to win? Typically in these mergers, one culture is going to win out. Do you think they are going to be able to maintain their separate identities? I'm dubious.
As for 3D, I don't believe they can ignore it. Will they embrace open standards? We've seen a glimmer of open-standards friendliness with Adobe and their support of SVG (Scalable Vector Graphics), but those days might be over now that they own Flash.
What's your opinion of (Microsoft's upcoming graphics system) Avalon?
Parisi: At the risk of sounding too technical, it's BS 3D. You can drop in a cube or a sphere and drop some video on it, and they're calling that 3D. If you want to develop any kind of 3D that has complex objects or behaviors or a rich 3D environment, what they don't tell you is that you then have to write code to that. You're basically writing C# or C++.
Microsoft pioneered a lot of the 3D technologies you say have prepared the Web for this 3D explosion. They created DirectX, Direct3D. They have the Xbox. What's to keep them from putting that technology into Avalon?
Parisi: I can only assume it's internal politics keeping that from happening. I can't see it happening any time soon. You have to give Microsoft credit for one big thing: They are very focused. They want to win at this console market. Why would they distract themselves with this market that's going to take five years to blossom? By the time this becomes a mainstream phenomenon, they can buy their way into it or build it then. They're taking minimal baby steps to say they're doing 3D, but it's not real 3D. So if there is going to be any relevance to Avalon, they need X3D to succeed.
Aren't open-source gaming engines one of your biggest competitors?
Parisi: From a technology standpoint, the most significant competition would come from open-source gaming engines, because the technology is the most similar to what we have in Flux. It's about real-time objects animating, moving, about fancy rendering techniques and being able to interact with them. Those engines, traditionally, have not been packaged to be deployed on the Web. They've been 20 or 30MB downloads, though that's changing. It's also a business model issue. They're developing games and the engines to support the games. You don't have a motivation for development of this kind of content on the Web. If there were a freely available open-source or open-source-based ubiquitous playback engine that did 90 percent of what the commercial ones do, you'd see a lot of development directed there.
But if an open-source game engine took off, that would be the real competition.
