Babak Bagheri

I’ve been absent from these pages tending to other matters. One of these is now interesting enough to write about. So here it goes.

For the past year and few months, I’ve been conducting a real-life experiment in bootstrapping a business. The original impetus for the experiment, aside from wanting to make money of course, was Apple’s creation of the App Store. The App Store is a remarkable market place. It provides a structured environment where I thought the kind of startup experiments I had in mind could be carried out cheaply.

I wanted to see if a series of applications could be made at minimal cost to generate enough interest and revenue to become self-sustaining. This would be a very low-cost way to try out my ideas about the experimental way of bootstrapping a business. The business entity I set up, Mentiva LLC, has now created three applications that are available for sale at the App Store: Swoop, Galaxy Wars, and Doodle World. The first two are games, the last offers a combination of drawing and physics simulation. The sales and reviews of these apps have been less than stellar, but the patterns and lessons that emerge are very interesting and valuable.

Swoop

The first game, Swoop, was aimed at pushing the iPhone and iPod to their limits by creating a true 3D environment. This actually turned out to be fairly challenging given my self-imposed cost and time limits. The app was done in about 3 months by one person (yours truly) working on it off-and-on using the least expensive tools available. The out-of-pocket costs for the app were less than $2000. The experimental question was: would a true 3D simulation be enough to generate interest and sustain upgrades? I had my answer within a few days of releasing the app: No. It became clear that App Store customers were interested in other aspects of games. Pushing the devices to their limits with 3D simulations was actually a negative, not a positive attribute. I’ll expand on the lessons below, but one interesting point was that the object of the game itself, parachuting, did generate interest, as in “good idea, but…”

Galaxy Wars

The second game, Galaxy Wars, was aimed at using the multi-touch capabilities of the devices in an interesting way. I had never considered myself a “gamer,” but I had enjoyed games like Asteroids, so the next experimental question became: would a multi-touch game in the Asteroids genre generate enough interest? Galaxy Wars was created at an even lower out-of-pocket cost: a few hundred dollars for an additional second-generation iPod. The time to create it was also fairly short: about a month from idea to app. Again I had my market answer within a couple of days of the release: No. The multi-touch controls were unusual enough to actually turn people off instead of generating interest.

Doodle World

The third and most recent app is Doodle World. Here the question was: would an unusual combination of already-successful features generate interest? The two features I chose were “glow doodling” and physics simulations, both of which had been successful in existing apps. With Doodle World, a clear answer hasn’t emerged yet. There is one very negative review of a version of the app, but it says more about the approach than the specific market question.

Three experiments are not enough for sweeping generalization. But some patterns are emerging. I’ll expand on one of these that I think is most important: it’s challenging to carry out low-cost experiments in a commercial setting. This is true even in a very structured environment like the App Store.  The reasons have to do with the way we make purchasing decisions. We go into a transaction with a set of expectations about both what’s on offer and the seller. Low-cost experiments taint these expectations, they can “damage the brand” even if there is no brand yet. The challenge is to do just enough initially to pique interest and not damage the brand. To do this one needs to know the market well enough to push the right buttons and avoid pushing the wrong ones. But those are really the questions we want our experiments to answer, hence the challenge…

Computer programming, “software development,” has been a large enough field to qualify as an “industry” for some time now. Most programmers spend their time figuring out and writing out, in excruciating detail, procedures for machines to follow. To do this effectively, they have to know, in fairly excruciating detail, what procedures other programmers have coded so that their procedures can interact correctly. Can this field continue into the future along the same lines? I don’t believe so.

The reason is actually fairly simple: piling detailed procedures on top of each other is not a scalable activity. This is so even if we use the latest and greatest techniques for organizing these procedures using objects, aspects, patterns, and the like. As the industry has found out, large piles of code become unmanageable, or at least very difficult to manage, whatever organizational techniques are used to create them. In fact, the most successful software projects have managed to contain their most important procedures within a small “kernel.”

Cube Farm awaiting a "Bank of Programmers"

Unfortunately, the current bank-of-programmers approach to software development does not encourage this kind of small-footprint development. There is a deep reason for this. We have been missing an important class of computational kernel: machine learning. As long as the systems we create have to be programmed for every procedural variation, we’re stuck with hordes of programmers and brittle software.

The usual story goes something like this: a beautiful small system is developed; users interact with it and generate wish lists of what changes they’d like, sometimes very loudly; programmers are put to work piling code on top of the original system; the system gains more fans but gradually becomes more and more difficult to maintain; at some point, someone decides that the system needs to be “re-architected” and the cycle begins all over.

Some have argued that the solution is to embrace this cycle, speed it up, and monitor it to make sure the system evolves in “the right direction.” That’s an improvement over the traditional rudderless approach, but it misses the main point: if systems could learn, then much of the piles of code would be unnecessary. In the next few decades, either we take machine learning more seriously, or the software crisis will get worse and worse. And no, “the cloud” isn’t going to help with this in and of itself.

The value of a good guess can hardly be overestimated. Many have tried to make a science of guessing. The fields of probability and statistics are the results of these attempts. Two men’s ideas in these areas have gained currency in recent years. The two are the Reverend Thomas Bayes and Andrey Andreyevich Markov. As with all ideas that gain popularity, there is a danger that these will become dogma, or at least second-nature, and replace critical thinking. Their approaches are similar in that they both take an axiomatic view of likelihood and its measurement. It’s important to know their assumptions to avoid mis-applications.

The British Reverend’s axiom is also known as “Bayes’ Theorem” or “Bayes’ Rule”. It can be considered a theorem because it can be derived from other “laws” or rules. It says that new observations should affect our probability estimates for guesses in a simple multiplicative way: the probability that a guess is true after a new observation is the simple product of the likelihood of the observation, with the guess taken for granted, and the ratio of the prior probabilities of the guess and the observation. This provides a very simple way to “update our state of knowledge” based on new observations. As you can guess, it is extremely difficult, maybe even impossible, to verify this general rule empirically in any reasonable way. Is the universe really such that uncertainties propagate in this simple way? There is evidence that it may be, but that’s a far cry from taking the rule for granted.

A Markov Model of Congestive Heart Failure Treatment (National Library of Medicine)

The Russian mathematician’s axiom is easier to state. It simply says that the probability of one set of observations immediately following another set is independent of history. At first blush this seems counterintuitive. After all, how can we ignore the past and hope to make a good guess? Markov, of course, never claimed that we could. His axiom just generalizes properties of simple chance events: the probability of winning the lottery does not depend on what the winning numbers have been in the past (in the absence of cheating). His generalization simply says that the right set of observations gives a snapshot that summarizes the history that led to it. In the case of the honest lotto numbers, observations of previous drawings are irrelevant. Markov’s theory then builds on this axiom to consider chains and networks of these history-independent observations (or “states”); systems like these are said to have the “Markov Property.” One can then find interesting conclusions about these Markov chains and networks in the aggregate. It is important to keep in mind that the Markov property is a hypothesis and should not be used indiscriminately because it simplifies the math, or, worse yet, just because it sounds sophisticated.

Again, what does this all have to do with you and me? Calculations based on Bayes’ and Markov’s theories are fairly common now. Some of these calculations are in very critical policy and engineering areas. Just intoning their names gives a claim or a theory a certain level of respectability. We should guard against accepting such derivative claims and theories uncritically and consider whether the underlying assumptions are really applicable, at least in our own work.

Why do we consider some products to have high quality? This is a simple-sounding but actually complex question. The common answer is to repeat the legalism about obscenity: I can’t define good quality, but I know it when I see it. That answer, of course, just begs the question, but there’s a kernel of truth in it. High quality is in large part assigned by a community, sometimes in fairly arbitrary ways.

During the 1990’s Toyota was in the process of establishing Lexus as a high-quality luxury brand. They had noticed that fit and finish (as the Detroit auto makers would have put it) were central to the American perception of quality in an automobile. They had in fact designed the Lexus to have just about the best fit and finish possible at the time. A famous TV commercial touted this attention to fit. In the commercial, a ball bearing rolls over the gaps between the car’s body parts to show how well and consistently they fit together.

Clearly, Toyota was right about the car-buying community’s views on quality; they had made other decisions about quality based on their observations that also hit the spot.

This social nature of judgments about quality means that abstract “six sigma” approaches to quality will likely be ineffective in isolation. One has to focus on what the target audience considers to be of high quality. Statistical measures should be based on these judgment, if it makes sense to use them at all. This is where the interactive web can help. The challenge there is, as mentioned here before, to cut through the large number of voices and be heard.

We know very little about how our brains work. Only recently have we begun to learn what goes on inside the cells and connections that make up the brain. These are good first steps, but the real challenge is to understand the network effects that actually lead to behaviors and how these network arrangements are formed. One group made a splash recently by claiming to have simulated the brain of a cat on a computer. This heroic computational effort needed one of the largest and fastest multi-processing computers on the planet. Even so, the simulation was limited and abstracted away from the real cat brain.

The RoBoard RB-100, a nano-itx board

Given this state of relative ignorance, can one take a practical engineering approach to construct an “artificial brain?” My contention is that this is possible. We need to combine insights from what little we know about the brain and complex computation in general. To these insights we should add something like the Nike tag line: “Just do it.” Fortunately, off-the-shelf processors are now powerful enough and networks are fast enough that we are less constrained by computational power than ever before. The key, as has been pointed out here before, is to avoid direct simulation of biological brains and think creatively about  computational kernels and network arrangements that can learn.

Engineering efforts of this kind will not only have direct benefits, but they will also enhance our understanding of biological systems. We should keep in mind though that these are only first steps. Ray Kurzweil’s singularity will not suddenly materialize tomorrow.

Network effects had been neglected for quite a long time. The classical “reductionist” view of the world was simply this: if we understand the rules that govern the pieces of the universe, then we can easily deduce large-scale behaviors. The past century has shown that this is wishful thinking at best.

Much has now been said about the level of complexity that interconnections among simple components introduce. This appreciation for network effects has not yet been exploited for engineering purposes fully. The old reductionist views still dominate in most engineering disciplines. The same is true about other practical arts; we do not yet have good ways of purposefully exploiting network-introduced complexity.

The reasons for this neglect are deep. To exploit the kind of complexity that network connections bring about we need to look at design differently. We need to view design as guidance instead of control. Complex systems do not respond well to direct control; they can, however, be guided and steered. We know very little about techniques to do this kind of guidance and control effectively.

Japanese Zen Garden (National Geographic)

There is one group, though, that has been struggling with these methods for some time: teachers and trainers. Living systems, especially people and groups of people, present exactly the kind of complexity that requires guidance and steering. Unfortunately, the track record for teaching and training is not stellar.

Perhaps the most effective of trainers have been spiritual teachers, for example Zen masters. Zen spiritual leaders have instinctively understood that guiding a complex mind requires meticulous attention to the environment. These environments are arranged to evoke the proper frames of mind that encourage the right mental changes.

We should start thinking of the machines we build as potential trainees and students. Our task is to design into them the kind of complexity that can be guided as we need. If a machine is simple enough to be controlled directly (“programmed”), then it’s probably too simple to do anything really interesting.