Click "Sleep" for a dark background.
Click "sleep" again if text isn't dark.



Abstract on Abstraction pt.1

The following article series takes a close look at how abstraction is used in craft and design by mainly focusing on the negative effects of abstraction. We'll examine what happens when creators abstract off of what has already been abstracted. Though doing so may seem harmless enough, with our understanding of art, design, and complexity we can shed some new light on the subject. Hence the title works on two levels; what happens when we make abstractions of abstraction and a concentrated examination on the concept of abstraction. Not to layer it on too thick, but I will say that part 1 of this series will be fairly abstract in that I'll start off very general with statements that pertain to creating art of various kinds. So you could say the title works on three levels. I suggest finding something solid to hold on to, which is exactly where we'll start off.

The image in the middle is my original abstract picture. Notice how the meaning becomes less and less clear when you abstract on it. Click to enlarge.


A Solid Foundation

Previously I urged us to Embrace the Abstraction, which means we should be willing to embrace simplifications and interpretations of all kinds within the art we consume; it means not holding particular styles or features as "standards" that other works should aspire to achieve; it's about understanding that video games, like other forms of art, don't have "archaic" design elements, rather some features simply go in and out of popularity. 

Since writing Embrace the Abstraction I was compelled to cover two other topics that are very important to video game design. First, I wrote A Defense of Gameplay explaining how gameplay is a unique and sufficient experience to convey ideas and experiences. Then I presented the idea of art's Infinite Undiscovery and how the limitations of artistic mediums, what art is, and the nature of complexity limit the meaning we can derive from art. In the article I also explain that supporting the core of gameplay involves proper tutorials as well as removing elements that distract from the core. Abstractions, or simplifications, are perfect tools for reducing the complexity of elements, pruning away excess so that what's left strongly supports the core. Put it all together and it's clear why we should embrace the tool that is abstraction. What I haven't explained is, like any tool, abstractions can be misused.

When software developers try not to abstract experiences, they end up creating simulations which are clearly interactive systems designed to be as accurate to real life as possible. Even with these extreme examples, simulations can't simulate everything. Real life is simply too complex. If elements lie outside of the realm of the simulation, they are often abstracted or cut altogether. On the other end of the spectrum it's possible to abstract too much. If a designer continues to simplify the complexity of gameplay elements, the elements can become too simple to express complex ideas or support complex skill based challenges. Abstraction is obviously a balancing act, but the question is how do we improve at it? 


How to Abstract Well

To properly abstract for the purpose of simplifying elements to strengthen a particular effect or core idea in an artwork, we have to be highly knowledgeable. For abstraction inherently involves taking something and making cuts. Think of abstraction like baking, surgery, or programming. If you make cuts haphazardly, you'll likely end up ruining everything  (or killing someone). Before making abstractions for our games we need to know what we're dealing with (the medium), what the goal is (conveying specific experiences and ideas), and how this happens (gameplay, interactions, systems, emergence, etc.). We need this knowledge before we can make informed decisions about what to cut. 

Complexities cannot be compressed, as I often say. This means that every individual element or unique bit of data cannot be conveyed in any other way other than explicitly and individually. You can try to invent clever ways to convey multiple bits of information with fewer "statements." But ultimately, even the most "successful" attempts to compress complexities end up either creating vagueness and ambiguity, or they simply load some of the complexity onto the clever construction. To add to this idea that complexities cannot be compressed, complexities for coherent, meaningful works are not of equal value. In other words, every complexity in a video game is not essential or core to its gameplay. We already understand this basic concept because we recognize that games have nuances. Nuance by definition encompass the extra, unnecessary, and generally unnoticed complexities.  


The Source

Assuming we have a strong understanding of game design that gives us the knowledge we need to make informed decisions about abstractions, we must consider what we're abstracting. Call it the source, a model, an example, or inspiration; for our purposes they're all interchangeable. It's important to draw your inspiration from rich experiences; experiences you ponder and mull over; experiences that affect you deeply; or just experiences you think are worth sharing with others. These experiences don't have to be complex experiences. But real life experiences tend to be inherently rich and complex.

Like I stressed above, before you abstract, before you begin to make cuts in attempt to craft your art, it's best to be very knowledgeable of the situation, which includes the source material. As we already know communicating something meaningful through a medium takes more than throwing in a bunch of elements you like. How the elements work together and work for the audience is important. Along the same lines, it's important to study the source material thoroughly to become an expert on it. Then you can take that expertise and convey the best most necessary parts of the experience without requiring an audience of experts to appreciate your work. Designers must do the hard work (learning/study) so we can spare our audience the trouble. And this is all because complexities cannot be compressed, complexities aren't all equal, and the conveyance of art is limited to a medium of objective complexities. 

It's important to understand that the most impactful and salient reason that most of us find meaning in abstractions is that they point to more relatable, more commonly lived, more "real" experiences. By pointing to their source material we can better understand the source material. This is the case even when the source is real life experiences in all of their complexity. Simply being "like" the real deal is what makes abstractions so compelling and art so magical. This "likeness" is also why metaphors have such immense artistic potential in the conveyance of ideas. I will certainly go in depth about metaphors soon. 

The original image abstracted 4 times. By the end, all meaning is lost. 

Decay of Meaning

The big takeaway that I wanted to end this article on is by making abstractions on abstractions there is a decay of meaning. Like the children's game of telephone where a message is passed between people so many times that it comes out completely different and nonsensical in the end, or like inside jokes made from inside jokes, repeated abstractions tend to produce meaninglessness.

Each time you abstract, you stress your design skills. The better designer you are, the more reliably you can create abstractions that retain the important complexities from the source material. But even if you're the best designer in the world, each time you abstract you must reduce the complexities you have to make cuts from. Each time you abstract, there's less and less of the incompressible complexities that made the source material rich. And though you can strive to cut the least important complexities with each abstraction, eventually you will abstract all the meaning away starting with the most complex ideas. 

Though there is no right or wrong way to go about designing a video game, there is definitely probable and improbable ways of reaching a quality, coherent, tightly desigend goal. Randomly typing code into a system is not going to work. The nature of chaos and order prevent something so complex, elegant, and meaningful from emerging from such randomness in all likelihood. Not playtesting and checking for bugs is another great way to produce a sub-quality product. Likewise, I believe game designers severely limit their ability to make unique games and express meaningful ideas when their main or sole inspiration come from other games. We've all played games that seem to lack "soul" or that seem to be completely uninspired and derivative of other games. For one reason or another, the end product is lacking. I can't help but think that this is mostly because the developers were aiming for an abstract result. From what I hear, publishers dread the kind of pitch that follows the formula: "our game is like X popular game but with Y small difference." Though these ideas may seem great in theory, it takes an extremely talented team of developers to independently reproduce the design greatness of a great game and then add to it. (listen here how this professional artist echos this idea of drawing from rich, original sources rather than other artistic works).

I believe there is plenty of room in our industry for games that are highly iterative of other games and for games that are abstractions of the abstractions that are video games. Regardless of what kind of game you make, it's still your to craft a meaningful experience with a strong core and foundation. The more you abstract off of things that have been abstracted, the harder it is to define and maintain a clear line of conveyance for the audience so they can walk deeply into the chaotic world of near meaningless abstraction and still find their way back to the core, back to the coherency, back to something real and relatable. I'll explain more of this idea soon in my update to my About That Indie Feel article series. 


If you found part 1 of this series to be a little too "abstract," rest assured. The rest of the articles will focus on specific examples. In part 2 I'll explain a few new concepts on the specific ways artists fail to properly abstract.

« Abstract on Abstraction pt.2 | Main | Pushmo: Puzzle Perfection »

Reader Comments (4)

Hi Richard, I have recently discovered your blog and really love it. I really like the ideas in this post, but it seems like you are using "to abstract" to mean different things. To me, the act of using another game as source material and changing this or that is a very different thing from taking a real world mechanic (JUMP) and translating it so that Mario is "jumping." I know you are going to go into greater detail in further posts, but could you clarify what you mean when you say "abstract."

May 12, 2012 | Unregistered CommenterDean

I'm finding this article really relevant to my life right now, both in terms of a project that I've been involved in that immediately jumped to mind when you were talking about games inspired solely by other games, but also in terms of my own ambitions to make games about real-world systems like soil erosion or fractional reserve banking or whatever.

I've been wondering whether I need to find experts on these subjects, or do the research myself, and wonder whether either is worth the trouble. Then I read:
"Designers must do the hard work (learning/study) so we can spare our audience the trouble."

Also, the latest post in Tale of Tales' dev blog connects with yours very well:

May 12, 2012 | Unregistered Commenteraxcho

@ Dean

Sure Dean. To abstract is simply to make a representation of something else. Unless you make a perfect replica, this representation will be a simplified version of that original thing. So if you're writing about an apple in a novel, your words describe some of the aspects of the apple, but there's no way they can describe every aspect of the apple (because real life is just that complex). It's the same when you paint a picture of an apple. Or put an apple in a video game.

When you abstract, you can pick what "original thing" you want to abstract. You can pick a real life thing. Or you can draw your example from something that was created to represent a real thing (abstraction). You can abstract anything really.

I hope that helps. Let me know if you find the links I make in my articles useful. I know that I explained abstraction in much greater detail in my article "Embrace the Abstraction." Also, let me know if you have any other questions/comments about my blog. I'm all ears.

@ Axcho

I remember sitting in my fiction writing class when another student was writing a murder thriller story. My professor talked about research books that other writers use to make their murder/crime scenes as realistic, consistent, and compelling as possible. Even when the murder/crime is such a small part of the story, it is the job of a good writer to do the work/research to get it right.

But yeah, if you find an expert use him/her. If you can't do as much research as you can.

Hi Richard, that does help. The links you make are useful, I started reading this blog at "Defense of Gameplay", and it had plenty of links to all the supporting posts. I also found your blog archives, which was quite helpful.

I guess its tough to figure out the line where abstraction in the form of deriving one's work from a previous person is harmful or helpful. I have read that sculptor's used to train by attempting to duplicate in detail a master sculptor's work. I guess this phase is like the piano player learning by repetition.

I do feel that games like Super Meat Boy and Braid suffer in some ways because they are abstractions from Super Mario. The whole theme of Braid, the finding of the princess, has much less impact if someone in the audience doesn't realize he is playing with princess quest of Super Mario. Super Meat Boy is clearly trying to abstract the platformer genre, which make work for long time gamers, but it may also exclude people getting into gaming for the first time.

Anyway, I look forward to the rest of this series.

May 13, 2012 | Unregistered CommenterDean

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>