By November 1998, Microsoft had shelved Chromeffects indefinitely, moving Engstrom over to MSN as a general manager for Web product development while Deborah Black took over as general manager of Windows presentation technologies. There were several “official” reasons for the failure of Chromef fects, mostly having to do with lack of developer support, complaints that Microsoft was ignoring web standards as provided by the World Wide Web Consortium, and the negative view around the industry that Microsoft was trying to take over web standards and make them proprietary.
Ludwig saw Chromeffects as a missed opportunity for Microsoft. “I was excited about the Chrome project, and it’s too bad that it didn’t happen at that time, because it took many years to reintroduce some of those ideas on the Web. And that’s unfortunate because I think Microsoft could have led the way there years ago.”
By the time Chromeffects development was halted, Ludwig had moved over to work in the MSN division and David Cole had taken over his previous position. He believes that the project was halted because there was just too much work left to do for it to become viable. “It was a lot of work back in those days to ship a full product. These days you can send out a beta forever and live with it, but back then that was not the case, and I think the judgment was that there was way too much work left to finish this thing.” On the other hand, he concedes that if he had still been managing the project, “I probably would have figured out a way to do something.”
According to St. John, none of that tells the real story. He contends that the DirectAnimation approach that Chromeffects used just wasn’t suitable for the market, but that was only one reason for its apparent failure… and not the one that mattered. He says, “Note that if lack of developer support or web standards was a reason for Microsoft to cancel anything, IE and Silverlight and C# would have been abandoned years ago as well. None of those are a MICROSOFT reason for failure.”
The real reason for cancelling Chromeffects, St. John states categorically, was the imminent threat from the DOJ. “Eric finished the 1.0 Chrome on schedule. It was done… not fast, but it worked. Nobody had the power to stop Eric at that point except Ballmer himself. Eric had to walk away from it because he knew that he was going to be pegged hard in the DOJ trial and was going to end up testifying. The DOJ trial was not a distant possibility at the time Chrome was canceled; it was an absolutely real threat to the company’s existence, and Bill and Steve knew it. The jets were scrambling to deal with it, and Eric had a bull’s-eye on him because he as the most prominent Apple, Netscape and RealNetworks assassin left standing at MSFT.”
From the perspective of WildTangent’s offices on the Microsoft campus at the time, St. John was able to observe what was going on first hand. “Eric was in constant DOJ trial prep meetings in the weeks before Chrome was canceled. It was all super-secret. Eric knew it had to be canceled. He didn’t stop it, and it was his baby.” St. John characterizes the MSN general manager position Engstrom was moved into as a “retirement job” that was given to him directly by Steve Ballmer.
In retrospect, St. John describes the pros and cons of the Chromeffects technology. “Chrome was utterly revolutionary, more powerful than even HTML5 today nearly 20 years later. But there’s a reason it’s taken 20 years for anybody to ‘want’ HTML5… it’s not clearly useful for much. Suffice it to say that the idea of human web monkeys authoring this stuff manually was the dream of smart people who don’t understand that ‘enabling’ monkeys with powerful tools won’t turn them into Thomas Edisons. They’ll just be monkeys banging around with expensive tools. HTML 5 and Chrome suffered from that condition. You can’t expect junior engineers with simplified tools to actually build extremely rich interactive web experiences just because it’s now possible for a genius to do it. It was the wrong ‘formula’ for media on the web. It either needed to be done with tools OR as the early DX prototype embraced, you just had to enable the full power of DirectX and let the people smart enough to use it make the products for everybody else.”
Accomplishments and Security Holes
St. John recounts the contributions of the DirectX team to online media. “Our contributions to online media after DirectX included naming the ActiveX API, doing the deal with Macromedia to ship the Flash plugin with IE, building the Windows media player and server, shipping the VRML 3D plugin for IE, and starting development on the technology that would ultimately become Windows Update… however our most exciting and significant contribution may have been a technology that never shipped… Chrome.”
St. John offers some interesting personal perspective on the DirectX team, and a shocking admission: “I worked with some amazing people and they’re maybe humble, and they’re also quiet. History should remember some of the things they did and that went on because there were some amazing achievements and some amazing risks taken that people at the time would never have thought were a good idea.
“Some amazing mistakes were also made that had consequences through the ages that I really hate confessing to. ActiveX. Don’t get me started. They wanted to call it Direct Internet or something because Direct was such a cool API, and this was when Microsoft was transitioning to the internet revolution and we were going, oh god, don’t screw up our branding. We don’t want whatever your plugin nonsense for allowing media apps to run on the browser to be called a Direct API because you have no idea what it means. So I was sent with the job of convincing them to call it something else. And everybody liked the “X” name, so I gave them a list of other blah blah blah X names and they picked ActiveX for it. And it was originally engineered to allow Flash to run on the browser, but it was also the pathway for enabling media apps, which I became in charge of. A lot of people don’t know that. And so the ActiveX was the single, biggest, most destructive security hole ever created in the Internet that caused viruses and spyware problems in Windows XP down through the years. I eternally sit in my basement and go, ‘God, I hope when history books written, nobody associates me with being responsible for helping create that API.’ Oops. I opened my mouth.”
WildTangent Implements Plan B
Fully incorporating in August 1998, WildTangent successfully implemented the features Engstrom had hidden in IE4 to realize St. John’s vision of what Chromeffects could have been. The result was that the first time, full real-time 3D games were playable inside a browser, resulting in a great deal of press coverage and providing St. John with a lot of personal publicity. Today, playing full 3D games on a browser may seem mundane, but when WildTangent first released their technology, it was a huge step forward.
Over time WildTangent’s website became host to numerous games and more than a million subscribers. St. John ran the company from 1997 to late in 2008. In addition to the success of the WildTangent services and technology, St. John developed multiple patents, including the patent for asynchronous updates on computers and a patent, which he sold for $500k to Google, for streaming real-time maps, which he had first prototyped while at Microsoft and rewrote at WildTangent.
According to Travis Baldree, senior engineer at WildTangent, St. John had intended to create a Web3D platform that would become universally adopted, and that at one time the company had about 200 people, many of whom were trying to pitch the technology to various different business and military operations. He also recounts the many experimental projects they engaged in, which he found often useless, such as a 3D stock ticker. But Baldree’s real interest was in making games, and after Chromeffects was canceled, much of the company’s non-investment revenues came from making games for various companies, such as Toyota.
WildTangent also ported mainstream games to work in browsers, including Need for Speed and Tony Hawk 2. Baldree says he had only two months to port Tony Hawk 2 and that he was not allowed to even look at the source code because they wanted to develop their own code which could be reused in future projects.
“So what I had to do was sit with two computers. I had one next me running actual Tony Hawk 2, and then I had mine, and I would count under my breath to see how long it would take for me to get from one location to another or how high I could jump. And just tried to gauge it and screenshot it so I could figure out these distances and times, which is really difficult for games like Tony Hawk 2 that are trick based, and where you have to very precisely jump from here to here.” He was also prevented from using quaternions* for his camera work, because the original game used them, and so he had to use matrix math instead.
Eventually, Baldree believed that he had enough tenure at the company to do a project of his own—an action role-playing game that he called Fate. He was given two months to do his project, but it took six. He says that some people at WildTangent had tried to discourage him from attempting the game, thinking it not casual enough to have broad appeal. However, Fate—an action RPG in the same genre as the mega-successful Diablo series from Blizzard—found a large and enthusiastic audience and helped WildTangent make a transition to games and away from the idea of a universal platform. Baldree eventually left WildTangent, and, after a brief stint at Flagship Software, he started Runic Games with Blizzard veterans, Max Schaefer, Eric Schaefer and Peter Hu where they expanded on Baldree’s success with action RPGs by creating the very popular Torchlight and Torchlight 2.
Sony and WildTangent
Among WildTangent’s investors was Sony, who invested $1 million in 2001. St. John relates how he received a visit from Shinichi Okimoto, chief technology officer at Sony Computer Entertainment, before Xbox was announced. Okimoto-san was asking about the possibility of Microsoft releasing a game console, and if they did, what direction would the likely take. St. John told Okimoto-san what he thought and advised that Sony should take its PS2 architecture, add memory, more graphics options and faster clock speed and call it a PS3, thus leveraging their PS2 developers and existing tools—advice that Sony clearly did not take.
After the Sony investment in WildTangent, Okimoto-san flew to Redmond again, and this time St. John decided to have some fun with him. “I couldn’t resist.” Because he was still doing contract work at Microsoft, St. John had a key card, and so he took Okimoto-san to visit the cafeteria in the Xbox building, which was only blocks from the WildTangent offices. “Nothing like that would ever happen in Japan, right? He came to do spying and intel, and I walked him right into Microsoft lunch campus and pointed out Stuart Moulder and all the executives, and so forth. I think he was mortified that I might introduce him to somebody. I had a real thrill doing that.”
St. John relates a very interesting conversation he had with Okimoto-san regarding the advanced cell architecture that Ken Kutaragi had planned for the PlayStation 3. Okimoto-san was not happy about it, calling it a “crazy cell architecture,” and noting that it would cost Sony billions of dollars. St. John adds that the Japanese have “a very nuanced way of criticizing their bosses,” so it’s probable that Okimoto-san said something far less blunt. But the gist of it was that not everybody at Sony was crazy about the direction Kutaragi had taken.
About Microsoft’s possible entry into the console business Okimoto reputedly said, “You don’t know how these crazy Americans are going to make consoles.” St. John predicted a strategy that had not yet been determined, suggesting that Microsoft would “run you to death on the software side,” but Microsoft could leverage its PC development base, make software as a service and add multiplayer, communities and online gaming. “Whether or not you have the best graphics and hardware capabilities, Microsoft’s going to define the competition for you based on online services. I said I could predict that with almost certainty because I shaped that strategy a lot. How they execute it, who knows?
“Okimoto-san said that they were basing their system on Linux. I said, ‘You know it’s really going to come down to, can you guys build operating platform software APIs, online services and online communities, online commerce… can you build that stuff? Because that’s where Microsoft’s going to drag you.’” According to St. John, Okimoto-san left Sony three months later—whether he quit or was fired was unclear, but it was clear, according to St. John’s conversations with him, that he was not completely onboard with Kutaragi’s strategy.
Ultimately, Sony put Phil Harrison and the U.S. based teams in charge of building much of the software and services, and at one point St. John received a phone call from Sony of America’s vice president of engineering asking for some advice.
WildTangent had a lot of experience in building online publishing and services, and when Sony’s VP of engineering asked, “What do I do?” St. John replied, “Here’s how you’ve gotta do it. Here’s what you’ve got to build. Here the way you’ve got to think about it. Here’s how you handle your developers… and this and that.”
St. John says the man was extremely grateful and said, “Thank you. You’ve been incredibly open. Aren’t you worried that I’m a competitor?”
“No, absolutely not.”
“Why is that?”
“Because if you do absolutely everything I just told you to do, you’ll go out of business.”
St. John says the man paused, stunned. “What do you mean?”
“Don’t you get it? Microsoft doesn’t care about consoles. They care about the Windows operating system. If they can drive the game business into an online business model where the hardware’s irrelevant, then you’re dead. And that’s what it means to do this. The minute the device is just an API and platform, and games can work everywhere, and the service is portable, and so forth, the console’s just another commodity box. And so, you’ll be gone as a competitor if you succeed at this. Microsoft doesn’t care because they’d be happy to see consoles be gone and go to the PC or laptops.”
Today, after a stint as CEO of High5, at the time of this writing St. John is CTO of Nyriad: http://www.nyriad.com/company/
* quaternions: a complex number of the form w + xi + yj + zk, where w, x, y, z are real numbers and i, j, k are imaginary units that satisfy certain conditions.