Flipcode Interview by Kurt Miller

November 1999

Flipcode - Before we go any further, please take a moment to tell us about yourself and why you began work on the Nervana project (including goals and expectations).

I have been programming since I was about seven. I started with Logo and BASIC and finished up with C when I was about 14. I have been pretty much programming in C ever since apart from brief stints into C++ when necessary. I like the potential of optimised C to move from the conceptual realm to the low-level mathematical realm. I have written all kinds of things - compilers, anti-viral software, hardware-driver software and lots of games.

I started the Nervana Project in June 1996. I started it for two reasons. I wanted to develop an advanced and yet relative simple method of simulating human movement in complex environments and I had also been developing bits of technology all over the place - I wanted a means of unifying what I had been developing. I realised pretty soon, at that time, there was no way of displaying relatively complex landscape environments in realtime. These realtime landscapes were needed -technology because they could show what the simulated humans were seeing. You need to also appreciate I was developing in Australia and was quite removed from the hardware/software development channels.

Flipcode - This article [the Rushkoff article] that was posted about the Nervana project was the first exposure that many had to the project. I think its fair to say that the article was considerably hyped and was certainly not a technical article. What has the general response to the article been and how do you feel about it?

Very positive. Overwhelming. I didn't know the article was coming out. I was in the process of touring. I called a fellow I was hoping to stay with in Berlin and towards the end of the conversation he asked if I had checked my email. I got half a million hits in three days and I was no where near my email!

What is interesting is that it got people talking both publicly and privately about the direction of the technology industry. That in itself has to be a good thing.

Another interesting thing about the article was it was not a traditional interview. In fact the article was based on a demonstration I was invited to do by Rushkoff in New York for a few of his colleagues and the philosophy of the technology development was what came through at that demonstration and also in Rushkoff's article. The article is a combination of the demonstration and the three years Rushkoff has followed the Nervana Project. The article is a hybrid of a number of the technologies developed over the three years.

How do I feel about it? I think it is phenomenal. Basically I have been pottering around doing this kind of thing for the past three and a half years with virtually zero publicity and now in a two week period I have some of the biggest companies in the computer industry wanting to do collaborative work. I have companies promoting the fact that they will be working with me in the future. In some small way I preferred when no one knew about me. But I just returned to Australia and started work on my third album and ignored my email for a couple of weeks.

Flipcode - With as little "buzz-word" talk and dream descriptions as possible, please explain what exactly the Nervana project is and why it is significant.

The Nervana Project was developed in isolation. It wasn't developed with the economic development realities of the mainstream technology industry. It was developed with a central notion of childlike-wonder. A sense of discovery and ultimately the notion that in an industry where the speed of processing doubles every eighteen months, the technology should evolve double-fold. A projection of the future - a technological phenomenon like virtual reality for example - should change dramatically over a two year period. Yet through conservative economic forces, research and development is the last thing to factor into the shareholders' wants and needs.

The end users should reclaim technology evolution and demand that companies should evolve technology at a far faster rate. That is what the Nervana Project is about. Taking off-the-shelf technology and making it do seemly impossible things. This is where the Nintendo Gameboy comes in. Here I saw a piece of technology that was ten years out of date when it first came out ten years ago. Yet here we have a basis for the amount of processing power that is needed to generate a first person perspective landscape in realtime. No fancy 3D accelerator hardware, just pure low-level optimised code.

Flipcode - When expect a new demo that hopefully shows off more of what the engine is and can do?

I am developing a demo currently of a cube where every surface of the cube is a water surface and seemingly impossibly there are fish swimming inside the cube and the move to every surface to breath. No polygons, just low-level mathematics and optimised code. Diffraction, rippling, diffusion, volume rendering and movement - all the things that they claim require graphics hardware acceleration. I want to introduce frame interpolation so it will run in real-time even if it is 50 frames a second or 10.

Flipcode - Any other comments?

Before the Rushkoff article came out I used to get a lot of emails from kids asking about programming. A central theme in my development has been about embracing the multiplicity of solutions to find faster alternatives to historical legacies. I think the real challenge to programmers is to become mentors to new programmers because new programmers don't see histories, they see possibilities.

ApeSDK Docs --- Forward (Rushkoff)