Through the lens

This evening I watched the SIGGRAPH Electronic Theater. This is the part of the conference where they show the thirty top computer graphic short films, as judged by an esteemed jury.

It is a bit of a devil’s bargain, because intermixed with the brilliant shorts and student films are excerpts from the demo reels of the major special effects houses, showing how they did this scene from “Edge of Tomorrow” or that scene from “The Desolation of Smaug”. For the most part, these are all pretty identical and devoid of even the semblance of dramatic interest.

In fact, it’s hard to see who these are for. If you are into the technology, then you already know all this stuff, so it’s boring. If you aren’t, then an endless sequence of matte shots, wireframe models and lighting layers is boring and meaningless. I suppose it’s just one of those hallowed traditions humans sometimes get caught up in, like slavery or factory farming.

But the rest was totally delightful. I found myself thinking about how much of the power of the wonderful student films and indie animations came from the choice of shot — the exact placement and movement of the virtual lens, the artful edit on action.

And I realized that a lot of what I was seeing would not translate, in anything like its current form, to a Virtual Reality experience. Cinema relies, on a very fundamental level, on directorial control over what you can see when.

Which means that the art and craft of telling stories in VR, as that genre evolves and matures, will necessarily be something quite different. It will be fun to see what that turns out to be.

Fast forward

The SIGGRAPH conference kicked off today with the traditional “Fast Forward”. This is the event in which all of the technical papers — and this year there will be something like 173 such papers — are previewed by their authors, who get 30 seconds each to convey the gestalt of their work.

I was one of those authors this year, and it’s always fun to get up in front of this community at the kickoff of such a great conference.

In fact, I have an unusual place in the history of the SIGGRAPH Fast Forward. The first year it was held — in 2002 in San Antonio — I had a paper at the conference. When I was told about this radical new idea, that we would each give an incredibly brief preview of our papers, I thought that was completely ridiculous.

“How can you summarize a paper in such a short amount of time?” I thought to myself. So as a kind of punk response, I wrote and then performed a gangsta rap version of my paper.

Well, dear reader, my little rebellion ended up being embraced by the community. The following year there were a number of similarly edgy takes on the Fast Forward presentation. In the intervening years, the punked-out Fast Forward presentation has evolved into a proud transition.

This year for example, one young scientist performed a truly excellent gangsta rap version of his paper. I wonder if he has any idea that I am his original mack daddy.


Our symposium today, the culmination of our three days of collective brainstorming and preparation, was a success! A large group of very interesting people came out to see what we had to show and say.

The audience asked great questions, the student demos went spectacularly well, and everybody was energized by the realization that all of our respective work really did hang together into a kind of shared vision.

Perhaps the best part was the realization that these people — the ones in the room with us — are our natural community, our future collaborators, our “fellow travelers”.

Now we’ll just have to see whether we can sustain that momentum!


For our intensive little workshop, Wednesday was brainstorming, yesterday consisted of mutual demos (followed by several hours of trying to digest and understand each others’ demos), and today was all about synthesis.

Tomorrow we will present, in a symposium, the results of our three day effort. Until this morning it wasn’t at all clear what we would end up with. There was so much to work through, so many intersecting concepts to absorb.

But over the course of the day today it all began to take shape. Knowing that we will need to go in front of a largish group of people tomorrow and present our collective ideas was a very effective motivator.

As Samuel Johnson once said of his soon-to-be-executed acquaintance William Dodd: “Depend upon it Sir, when a man knows he is to be hanged in a fortnight, it concentrates his mind wonderfully.”


Today, at a very hip coffee shop in Vancouver, our barista was a young man with a porkpie hat and a beard longer than the beard on the lead singer of ZZ Top. He also had a very cool t-shirt — black, with a printed white message that just said “[image]”.

It was the epitome of hipness. The shirt was sort of saying “Sure, there could be an image here, but that wouldn’t be cool enough. So instead I am a shirt that simply refers to the idea of an image.

I really liked that. So when I went up to the counter to return my coffee cup, I said “I like your shirt”.

Whereupon he looked at my shirt, and said “I like your shirt too, but I’m not smart enough to understand it.”

That’s when I realized that I was wearing the cultural equivalent, in my world, of the same t-shirt. Mine was also black, and printed on it in white was the Rendering equation.

This is a wonderful mathematical integral, formulated in 1986 by Jim Kajiya, that describes the fundamental laws of how things are illuminated in computer graphics. I cannot begin to count how many technical papers in the field have been built upon this formula.

So I wasn’t actually wearing a computer graphic image — I was wearing the idea of a computer graphic image. Which means, I guess, that I am, in my way, just as much of a hipster as that barista. Only without the beard and porkpie hat.

I was happy for the compliment. “Thanks!” I said, nodding toward the word on his shirt. “One of these is a way to make one of those.”


Today I worked with a small team of colleagues on an all-day brainstorming session. And it was a lot of work.

One of the most difficult things about brainstorming is that you don’t really know where you will end up. Yes, there are general guidelines, ground rules, agreed upon paths, to be sure.

Yet by definition, you don’t know, going in, which ideas are going to evolve into something exciting, which are going to be controversial, and which might just turn into a sinkhole, leading to argument, confusion and mutual incomprehension.

But it’s always worth it. Even the bad parts are good, because what you are really doing is learning a map of each others’ minds. So not only are you generating ideas together, but you are steadily improving your ability to work together.

In the long run, that can be the most valuable part.

Building a library

Every day for weeks, here at the Centre for Digital Media in Vancouver, I have passed by an empty bookshelf in the hallway, just around the corner from the room where I’ve been staying.

I realize that this is a centre for digital media, but the sight of that bare bookshelf, looking so empty and forlorn, would tug at my heart every time. So finally I decided to do something about it.

This weekend, on a visit to Bowen Island, I came across a wonderfully illustrated book about Visual Thinking. I brought it back with me on the ferry yesterday morning, and put it on the shelf. Now at least there would be one book!

This morning I emerged from my room to discover that this first book had been joined by a second — a nice little paperback of short stories about Vancouver. Overnight, the number of books in our emergent library had doubled!

I am hopeful that people will continue to add books, and that our little shared bookshelf will grow and grow, one day becoming filled to the brim with good old-fashioned pre-digital media.

Wearable technologies

Clothing is such a successful and well integrated technology that we generally forget that it is a technology. For most people reading this blog, almost anything you are wearing is the result of highly mechanized and capital-intense methods of production.

And clothing is not a voluntary technology — it is mandatory, both socially and legally. If you try to walk out in public without clothes, people will think that you are crazy, and you will also get arrested.

There is much worry, as our society edges ever closer to universal wearable technologies, that we will lose privacy, anonymity, freedom of unmonitored movement. Once we pop in those convenient AR contact lenses, we might find ourselves tagged by the NSA, Google, rogue hackers in Belarus, or god knows who else.

But we long ago gave up our ability to remain “untagged”. The clothes we wear — those extremely mandatory clothes — tell a rather detailed story about who we are and where we fit into society. And rather than fight that ability to be identified, most people have been socialized from an early age to embrace it, to incorporate it into their very sense of self.

The social control exerted through fashion can be quite pernicious. For example, if you are too poor to afford expensive clothes, there are many places where you will never be welcome, or even permitted to enter.

So don’t be surprised by young people who are not at all freaked out at being constantly tracked through the newest wearable technology. Being continually monitored and identified, wherever you go, might very well come to be considered the height of fashion.

Learning curves

Every so often I am struck by the huge gulf that can exist between “knowing what something does” and “knowing how to make something”. The tools that we rely on can be every simple to use, and yet extremely difficult to create and to understand under the hood.

For example, everybody gets what television does, but it takes specialized knowledge to understand how it really works, and even more specialized knowledge to build a working TV set. The same goes for automobiles, pianos, smartPhones, and all sorts of other modern tools that we take for granted.

This came home to me yesterday when I set out to do something for a computer graphics project that is very easy to describe: Given two curves, find out how much to move and scale one of the curves so that the two curves become superimposed on each other as close as possible.

To do that, you need to change the x,y and scale of one of the curves. All those possible x,y,scale values form a kind of parabolic dish, and the best answer is at the very bottom of the dish — the place where a marble would roll to if you dropped it into the dish.

But to turn that picture into math, and then into a computer program, took me several hours, and at the end of it I had a little piece of code that would make sense to very few people:

      var n = min(P.length, Q.length), a=0, b=0, c=0, d=0, e=0, f=0;
      for (var i = 0 ; i < n ; i++) {             var px = P[i][0], py = P[i][1], qx = Q[i][0], qy = Q[i][1];             a += px;             b += py;             c += qx;             d += qy;             e += px * px + py * py;             f += px * qx + py * qy;       }       return solve([ [n,0,a,c], [0,n,b,d], [a,b,e,f] ]);

Like most software, this relies on other software. For example, the last line calls a linear equation solver, which is a standard tool that I already had lying around.

Everything that is interesting — and maybe difficult to understand — about these dozen-odd lines of code is in how you convert a simple idea like “make two curves line up” into another simple idea like “find the bottom of a dish”, and then into mathematics. There isn’t anything interesting about the computer code itself.

Sometimes when I think that we should be teaching everyone to program, I look at code like this and I realize that the real power-up doesn’t come from learning programming, it comes from learning how to express ideas using math. And that’s a whole different thing.

Under the watchful gays of the church

Every once in a while you run across a story which is so incredibly stupid, embarrassing, and “hit me in the face with a brick” idiotic that it becomes weirdly transcendent.

Such a story happened the other day, when Tim Torkildson lost his job at the Nomen Global Language Center in Provo, Utah, because his boss, Clarke Woodger, sacked him for posting a lesson on “Homophones”.

As everyone who reads this blog knows, a homophone is a word that sounds like another word. You can’t really teach English without covering this essential topic. But Mr. Woodger, apparently, thought that teaching about homophones would promote a gay agenda.

Around this point, you are probably thinking that I’m making this up. But no, I swear, this is actually what happened.

I suppose I could try to say something clever and funny about this. But what could be funnier, in a tragic sort of way, than what actually happened here?