Category: 01

Compton’s “10,000 Bowls of Oatmeal Problem” illustrates the possible problems a creator can run into while utilizing generativity into their work. The problems can be faced is when the creator constructs so many artifacts making each of them unique in their own way but it can become difficult to be perceived the same way by the user/audience.

We can look at this idea as we look at the construction of violins. Each crafted violin may produce a different or unique sound. However, on the outside, they are all designed the same way and are utilized the same way. Here comes Compton’s concern where the unique qualities of a generated object will remain unknown without acknowledging its good/bad attributes. In this case, depending on the user, an individual violin can maybe produce more richer, deeper tones than the violin next to it. Whether that is a good/bad attribute will be determined by both the user and the maker but either way it will help identify perceptual differences within the clutter of instruments.

In order for the artist to overcome this and generate a copious amount of artifacts that truly vary from each other, they must be able to recognize the different types of perceptual differentiation that can be experienced by the user and enhance them until the users find it recognizable.

In Kate Compton’s reading, the “10,000 Bowls of Oatmeal Problem” describes an issue in generative work when an algorithm can produce a ginormous amount of artifacts that are each unique but are not perceived as uniquely different from the audience perspective. Compton provides an example that if she created 10,000 bowls of oatmeal and each grain of oat was different, according to the algorithm, it is unique, but perceived from an outside view, it all ends up looking the same.

When you need to generate thousands of artifacts that will have to vary slightly, you can choose between perceptual uniqueness or perceptual differentiation. Perceptual differentiation is an easier level to succeed at if the environment doesn’t need the artifacts to be highly memorable (i.e. trees in a landscape or a very large crowd.) Perceptual differentiation is when the user can tell from a glance that there is a difference between the artifacts, but it isn’t that significant. On the other hand, perceptual uniqueness is the contrasting view between remembering a main character versus remembering a face in the crowd. The artifact must have a distinct character personality, making it more memorable than the other artifacts.

To overcome this problem, it is important to understand your audience and know the key characteristics of the artifact you are generating. Humans like readable meanings and identifiable personalities, and is a great strategy to start with overcoming this problem.

The “10,000 Bowls of Oatmeal Problem” describes when an algorithm can produce many artifacts that can be completely unique but aren’t necessarily perceived as different. Hence, Compton argues that if 10,000 bowls of oatmeal were generated and each oat was mathematically unique, the user might not perceive the oats as different.

When generating hundred or thousands of artifacts that vary slightly, creating perceptual uniqueness is difficult. Perceptual uniqueness is when an artifact has a distinct personality and is memorable amongst the other artifacts. Compton uses the example of a main character versus another face in the crowd. 

On the other hand, when creating landscapes or environments, perceptual differentiation is useful. Perceptual differential is when a user can spot a difference at a glance but the artifacts aren’t particularly memorable.

Artists could utilize perceptual differential and create environments while perceptual uniqueness can be explored through other means and techniques. 

Kate Compton coined the “10,000 Bowls of Oatmeal Problem” as a way to describe a common issue in generative artwork, differentiation between the different artifacts generated.

It is an interesting problem because it proposes to us the fact that while two objects may be mathematically different, these differences may not be perceivable at all for the user/player.  In general, there will be cases in which different levels of perceptual differentiation are needed when automatically generating objects.

For instance, assume you are trying to make a new race of monster from a set of attributes (some will have a tail, some wont, some will have legs and some might have wings), if you have a large number of varied parts to construct a monster, it is really likely that each monster you get might be completely different, with every single iteration creating a new race of monster.

But this could come with its own set of disadvantages, assuming you want an entire colony of monsters of the same kind, maybe this specific type of monster only has variations in eye color and tooth shape resulting in hundreds of monsters that could look similar but retain individual characteristics at a closer glace. I think in general, intention and context will determine the mutability of the object you are trying to generate and this is something that is at the core of what Kate Compton was trying to describe, you need to know what are the good attributes and the bad attributes (and the memorable attributes!) of what you are trying to accomplish when generating an object.

Using meaningful symbols and identifiable perceptual differences is a great way to avoid the “10,000 Bowls of Oatmeal Problem“, add various degrees of clear mutability when needed (but also know when not to go overboard!).

Kate Compton’s idea of the “10,000 Bowls of Oatmeal” problem describes the difference between mathematical uniqueness and perceptual uniqueness, and in my opinion, how our human perception of aesthetics and uniqueness is completely subjective.

Humans identifies objects by their key characteristics, for example, a bird by a feathered egg-laying animal that walks on two legs; and we differentiate different types of birds by what we consider to be their significant characteristics: their color, their size, their beak shape, etc. We do not identify them via their smell, or that one of them have an extra feather on its tail(though that maybe significant to birds) — these details are not what we look for when we perceive a bird. Hence, only changes on the significant characteristics ends up influencing our perception, and contributing to the perceptual uniqueness.

Thus as the article’s proposed creative process for procedural art suggests, it’s important to identify these key characteristics of whatever you’re trying to generate, and know your audience’s expectations. The article also touches on how associations and context could  contribute to the authentic-ness of the object, and that reminds me of the Bouba/kiki effect, where our associations of shapes/sounds/color/etc is constantly affecting our perception of the world– and we should use them to our advantage when creating work.

10,000 Bowls of Oatmeal Problem

The 10,000 bowls of Oatmeal problem examines a feature of generative art, the perceptual uniqueness – the  amount of distinction between artifacts and perceptual differentiation – the ability to perceive  two artifacts as different.

The lack of perceptual uniqueness does not necessarily hurt a project since depending upon the context a lack of perceptual uniqueness among a collection of artifacts can benefit or hurt the collection. For example, a game focused upon collecting an infinite number of creatures would be harmed by a lack of perceptual uniqueness since the game’s main draw is the infinite number of unique creatures. While the promised variety theoretically exists, but since the user can’t quickly distinguish said variety, it doesn’t really exist. On the other hand, procedurally generated crowd that wanders the background would benefit from a lack of perceptual uniqueness and merely having perceptual differentiation. The slight variation between each character would provide subtly uniqueness to each background character without causing them to stand out.

In order to increase perceptual uniqueness, the author recommends we generate objects that each have emphasized characteristics humans can easily distinguish.