Startups, Zombies, and Film Photography
Perry Kivolowitz, digital effects guru, novelist, serial entrepreneur, and Oscar-winner, on his forty years in photography and programming
Polarr: What drew you to computers?
Perry: When I was 13 I read Michael Crichton’s The Terminal Man which is about artificial intelligence. The first program I ever wrote was on an HP 9835 desktop calculator, which consisted of a print statement that said “guess a number.” It read the number, printed wrong and went back to the first statement. So of course there was no correct answer whatsoever, it was a prank, and my classmates spent hours trying to guess the number! I thought that was great.
Polarr: That develop toward digital imaging and special effects?
Perry: My interest in digital imaging began prior to grad school, and then I did my graduate research in computer graphics. I had a lifelong interest in photography. I first held a camera in my hand at six years old and some of my earliest memories are in fact — it sounds silly — but they are in a dark room.
Perry: My father was a very avid photographer. And he was an early experimenter with push photography, he did a lot of night photography using the normal speed black and white films of the day and was shooting up at 3200 and 6400 ASA. The exposure provided by my father was definitely the formative exposure.
“Film is clearly permanent. What’s not clear is whether digital images will pass the test of time.”
And then in college I was the photography editor for the yearbook, and I was developing all the negatives myself. It was wonderful in those years because the yearbook publishing or printing company would develop all your film for free, so I was able to do a lot without a lot of expense. And you know you develop the negative in total darkness, and they have to be developed in different solutions. I knew which one to apply to each roll by tasting the film.
Polarr: How do you feel about digital simulations of film by companies like Instagram? Do you feel like the culture has lost anything, in the move to digital?
Perry: Certainly there is some loss and some gain. As in computer science, the answer that’s almost always appropriate is “it depends.” I think it’s wonderful that now the choice to use film can be an artistic or practical decision. And the opportunity to recreate the different looks of film allows you to have that type of artistic control again in the digital era. I would say the experience of producing photography is clearly better now.
On the other hand, I can go back to silver emulsions that are 150 years old and I can still develop them. It may be difficult, but I can still get prints off a picture taken 150 years ago. I doubt very much that 150 years from now someone will be able to get the print off of an arbitrary digital image. So the permanence of film has clearly passed the test of time; it’s not clear that digital images will pass the test of time.
Perry: A second thing that we’ve lost is, and this speaks to the objection that “it’s too easy,” is a film was, and I’m sure everyone who talks about this subject says the same thing, I had 24 exposures or I had 36 exposures. With a finite number of rolls of film, it was expensive to develop each shot and there was no immediate feedback. So the decision of what to shoot and how to shoot it was way more deliberate. The feedback and the technology within the camera was very limited.
But we have lost the need to be deliberate with each shot. During my college years I shot everything on a Minolta. It was the first 35mm camera to have a physical light meter on it, not through the lens but actually on the body of the camera. So in a dark situation it was clearly unusable, it was all done by eye. I’ve got these wonderful concept photographs of B.B. King. For those shots, it was just pure guess, it was pure guess in my experience, for what the exposure would be.
Perry: But coming back to the digital era, of course being able to look at the results immediately makes it easier to produce good photographs. But we have lost the very conscious approach that had you wonder, “should I push the shutter? Should I push the shutter release?”
But I think in a way it’s the same as it ever was. The same digital camera can be an enormously fun personal extension of your eye in the hands of one photographer, but in the hands of another person it’s a point-and-shoot. There were always some people who used film cameras as point-and-shoots, and there is no difference today.
Polarr: What are your thoughts on selfies?
Perry: I think there are way too many of them; they’re just gratuitous. People should be thinking about whether or not they want to permanently enslave billions of electrons to store that image. There should be an organization, ‘People for the ethical treatment of electrons.’ But I also think the idea of memorializing a moment is an important one. So, if you’re going to take a selfie, make a conscious effort to make the best selfie you can. Those are photographs too, so people should try to make them a piece of art, not a waste.
Polarr: I’d like to hear more about the companies you’ve founded. Is there any common theme among them?
Perry: Well, the later companies I founded came about in a very different way than my first company did, for instance. For that first company, I looked around at other products and thought to myself, “You know, I could do this at least as well as any of these other people.” Actually it was more like, “You know, I couldn’t do worse than these people so why don’t I do it myself?” And that’s what launched my first company.
“There should be an organization, People for the Ethical Treatment of Electrons.”
I started that first company with the hope of making hardware for data acquisition laboratory instrumentation. I produced a board that spoke GPIB, the General Purpose Instrument Bus, and that was going to be the first board that got to the heart of what I wanted to do with this company. So I went to a conference in Atlanta to show that board to Commodore Business Machines, the manufacturer of the computer it was supposed to go into.
Perry: It was pouring rain when I came out of the hotel on the second day of the conference, and by sheer chance I ended up sharing a cab with the general manager of a brand new division of Sharp electronics. It turned out he was at the show to introduce the first desktop color scanner. I asked him how the scanner talked to a computer and he said, “Oh, something you’ve probably never heard of, GPIB.” I said, “Really?” And I pulled out the board that I’d just developed that spoke GPIB.
“Oh that’s Commodore, isn’t it?” he said, “I used to work there. Jack hired me through a bathroom door.” There were only two scanners in the country at that time, but he had a soft spot for the company, so he lent me one for 30 days. He said they’d send some engineers out in the end to see how we did. So they finally came out, fully expecting to put the scanner into its box and take it back with them. Instead they saw the software we had developed and let us keep it so we could finish.
And that was the desktop publishing revolution, and from there it was image processing, and full desktop publishing. And then it became image processing for multimedia. We had a very early product, an image processor designed to process thousands of images. If you think of Photoshop as an image processor for one image, then our product was an image processor for thousands of images. So we were processing thousands of unrelated images, and it was a small step from there to process thousands of related images, namely desktop video. So we started, we rode the desktop publishing revolution, we rode the multimedia revolution, then we rode the desktop video revolution. Then finally there was a small step to working in film, and that’s where I’ve been ever since. It all started with a cab ride.
Polarr: What would you say is the accomplishment of which you are proudest?
Perry: I think the thing that makes me proudest is that I was able to build a company that offered good pay and benefits, in particular for an employee who had a very premature baby. From a technical perspective, certainly the development of warping and morphing, which resulted in an Academy award for scientific and technical achievement, is something I’m proud of. And then I spent another 10 years thinking about how to do it better. I finally crystallized the algorithm down to just a few lines of code and it actually is a more powerful, faster and better, so I’m very proud of that. Of course no one would ever see that, but I do. It’s now in Silhouette — when we added warping and morphing to Silhouette it was based on this new algorithm.
Polarr: What was it like to win the Academy Award?
Perry: First I just have to explain that the sci-tech category is unique among all of the Academy Awards categories. It’s not like best actor or best director; there is one category for all possible tech contributions to filmmaking. In any given year, there may be one award given or there may be 20.
It’s a rigorous selection process: There is a lot of research involved, you go through something similar to getting a PhD. Candidates can’t talk to the committee, instead a world expert in the area is assigned to be your surrogate and they research the background of the entire field. That’s cool too, if they’re researching my contribution and they discover someone else made a similar or related contribution, that person will also get the award. They actually can grow the number of people who might be considered in a particular invention. And then if the surrogate believes the invention is award worthy, they will go argue the case in front of the full committee.
It was elation, it was a thrill, and in some ways it engendered a sense of responsibility. As part of my acceptance speech, which of course only lasts 50 seconds, I said “I hope to do this again.” It has definitely impacted my career. It impacted everything — it’s made it easier in some cases to attract money for startups, it’s made it easier to sell products in the motion picture field. I think it opened doors in lots of ways.
Polarr: Can you talk a little bit about your role in developing one of the earliest key logging programs?
Perry: My principal interest in those years was in operating systems. The potential to record keystrokes was an interesting technical problem and I enjoyed the challenge. I enjoyed the feeling of succeeding in coming up with a way to do that. It began with an observation that these characters are available as typed, and the question of whether I could go retrieve them in real-time and then pull them out of the intestines of the operating system. I did with the intent from the beginning of sharing the source code and calling attention to that as a vulnerability in Unix. In fact I sent the source code to the first key logger out on to Usenet, which existed before the Internet, and said “Okay, here’s the problem with the way Unix works. Here’s the code that exploits it (and it was fun writing it), now maybe we should fix the vulnerability.”
Polarr: You were aware, even in the beginning, that key logging could be used for nefarious purposes. Do you think developers are morally responsible for future use cases of their inventions?
Perry: Well I think the, yes, people are responsible for what they create and it is a question of intent. If you are a member of a criminal enterprise and hope to develop a successful product for a criminal enterprise then very clearly you are a criminal. If you are a security researcher who is attempting to exploit and then bring attention to vulnerabilities, then you’re a good guy, and there is enormous amounts of gray areas in between. For example, working on facial recognition, that’s a great feature for a photography product for Facebook and it’s also an incredible tool for big brother, for oppression on a massive scale. That one is a totally gray area. I guess for me it comes back to intent, what is the intent in the development.
Polarr: You recently started a company in the music industry, which was a departure from your previous projects. Can you talk about what that was like?
Perry: I happened to watch musicians playing on a stage, and thought that really is an area that needs a new application of technology. And then I thought of the solution and figured, you know what, I know absolutely nothing about the music industry. I’ll start a company in it.
Perry: It’s a very difficult industry to break into as a technology startup. For one thing, the manufacturing of tangible products is not something that’s easily funded these days. The industry, and the technology that Polarr is working on, for instance, is more likely to be funded than a product that will be manufactured on an assembly line. So it’s very difficult to develop a new physical product that you hold in your hand. You are far more likely to get proper funding with an app, a piece of software of some kind. A financial person who is hearing your pitch can more readily identify with a social networking app proposal than they can with a new piece of audio workstation.
“I was yelling at the characters on the Walking Dead, hoping they would die. Because they were so stupid.”
Polarr: Do you think that’s at least partially because of the rise of crowd funding for physical products, or do you think it’s an oversight on the part of people doing the funding?
Perry: Well, venture capitalists are interested opportunities that might have a lot of zeros at the end of the potential. My music technology product has the potential of a small number of millions per year, but an app has the potential for a small number of billions per year, so the venture capitalists are not really that interested in my product. And that’s really just a question of scale. Physical products don’t scale as much as technical software products.
Polarr: So now I need to ask about your novel. It’s very different — what prompted you to write it?
Perry: I was actually watching The Walking Dead with my son, and I kept yelling at the TV, hoping that the characters would die…Because they’re so stupid. And finally he said, “You know what Dad, if you hate it so much why don’t you just write your own?” So I did. And my goal was, once you suspend disbelief that zombies can exist at all, to have everything else make sense. And I think I did that. I don’t think there is another zombie novel told from the perspective of a computer programmer. And if folks are wondering how a computer programmer can kill zombies, it takes a lot of GPU of course.
Polarr: What do you think about the potential of augmented reality versus virtual reality?
Perry: Well I think augmented reality is the next big thing. When things like Google Glass become ubiquitous, when people reach out with their hands into empty space to start manipulating a UI that only they can see, such as in like “Minority Report,” I think that’s the next big thing.
“Virtual reality is either the very next generation of gaming systems or at most its one generation away.”
I was involved with a very early developer of virtual reality, in fact the product was named ‘the virtual reality.’ The developer was Jonathan Waldern, of W Industries, and he started a long time ago with location-based entertainment systems. Virtual reality has come a really long way; it’s either the very next generation of gaming systems or at most one generation away. There’s a living environment lab at the Wisconsin Institute for Discovery, and they use virtual reality to experiment with designs for homes for disabled people, I think largely centering on helping people as they age. And it’s all virtual reality-based for research purposes. It’s very clearly an important part of the future; it’s an important part of the future with respect to entertainment as well.
Perry: There is a number of companies now working on products to capture real world environments and reproduce them in almost a holodeck fashion, by which I mean not that you can go running around in it and zapping things in holodeck, but you have captured real world motion picture which is responsive to your gaze. So you could actually walk around a bit in it and the three-dimensional perspective changes as you go.
Polarr: What are your favorite projects that you’re working on now?
Perry: Actually SilhouetteFX has a new generation of products coming out. A critical component is being able to exchange data with other substantial products in that space. We exist in a field that is dominated by a product called NUKE and our users expect to be able to interchange data with NUKE users transparently and easily. So the large project I’m working on now is an exporter to export our data structures into their data structures. It may sound small but it’s actually a very substantial project, it feels like reverse engineering file formats, which I did for fun back in the day.
Polarr: Do you have any advice for startup founders who are just starting out?
Perry: Well for one thing, it’s very important to look at what other people have done because you can learn a lot from their mistakes and successes. But a successful effort really has come from inside yourself and that may include different spins, different directions and (more to my heart), different fundamental approaches to the technology. So don’t be afraid to zag when everyone else is zigging.
“I’ve had success by doing something unheard of — Actually asking people what they want.”
Another thing is to be really careful who your partners are. I’ve been blessed with a fantastic partner. In ten years of business, we never had a disagreement that we couldn’t resolve. And in fact, in 10 years of business, here’s how we resolved all of our disagreements: “On a scale of 1 to 10, how strongly do you feel about this? You’re nine, I’m a seven, okay we’ll do it your way.” Since then, I have been involved in a number of startups where partners were a problem. You have to be very lucky to have good people on your team.
Polarr: How do you decide what kind of product to develop? How does that process work for you?
Perry: Some people have become extraordinarily successful by essentially predicting what people will want in two years or five years. It’s kind of an arrogant strategy for coming up with products, and of course fraught with risks. Steve Jobs and Frank Lloyd Wright are examples of creative types who used this kind of strategy.
But in my own career, I have had some success in predicting what my customers will want, but I have had greater success is doing something unheard of — and that is actually asking them what they want. Something that I’ve had success in doing is actually going into into post production facilities of a motion picture studio, and just standing in the back of the room and watching. And what I was looking for were the repetitive tasks that they had to do over and over and over again, all day long, that I could find a way to automate. So some of the things that I have made money on were the least sexy of technologies, but they solved problems that people experienced in my field every day.
There are two approaches to product design, two approaches to business. One is to look for what’s hot, and a lot of people focus themselves only in that area. But another equally viable path for business are the unsexy products that people must have for everyday living, that aren’t sexy at all but can become pervasive. It’s absolutely killer if you can do both at the same time.