Sunday, April 19, 2009

Improving Readability: Empathy

Last week, I claimed that game readability can be improved with increased emphasis on empathy and data. I'll be discussing the importance of the former first, but there's quite a bit of overlap between the two.

I neglected to offer a solid definition of readability previously, so I will correct that now before we progress. Consider readability to be a measure of the player's understanding of what outcomes are possible and how they can affect those outcomes. Basically, the inputs and outputs of the game's systems. We can distinguish this on both a micro, minute-to-minute level and a macro, long term level, but it's the case that problems in one often affect the other.

To be clear, readability is a quality, not a value judgment. Some games are intentionally quite complex and unclear. Internalizing the game's systems is the primary source of satisfaction (and even they can often improve upon some unintended obfuscation). But when the readability of a game is diminished, it needs to be deliberately and with great care, which is rarely the case.

This is related to the concept of visibility that Donald Norman discusses in The Design of Everyday Things. The distinction in the interactions Norman presents are almost never explorative in nature. The object in question provides a specific function (or set of functions) and it is the designer's responsibility to communicate these things as clearly as possible, i.e. "If you want to achieve X, do Y." Readability in games more takes the form of, "If you do X, Y will happen. If you instead do A, B will happen." Visibility is about objectives, readability is about decisions.

Finally circling back, what does readability have to do with empathy? Creating a readable game system requires empathy for the player, namely the acknowledgement that your perspective as a designer is completely and irreconcilably different from the players. Readability is about the speed and accuracy at which the player can internalize the game's systems. For the designers, and just about anyone else involved with the project, some amount of this knowledge is already present. This makes their perspective on the game's readability completely inaccurate.

Empathy means acknowledging this truth, which is not isolated solely to games. Anyone involved in any sort of interaction design is simply too close to what is being designed to evaluate it objectively. Having one year of experience or twenty does not change this fact (although ideally a veteran designer has come to grips with this). In my life before games, I worked at a mobile media startup and saw this error made time and time again, to disastrous effect.

Bill Moggridge wrote a fantastic book called Designing Interactions, where he compiles essays from dozens of interaction designers. Suffice to say, it begins with Doug Englebart discussing the invention of the mouse and goes from there. It's very interesting reading for anyone interested in how good interactions are created, and game designers should count themselves amongst that group.

One commonality between most of the book's contributors is the importance of empathy for the user. Time and time again, the designers interviewed discussed how essential a user-centric approach was. Keeping the user's perspective consistently in mind and evaluating possible solutions are the building blocks of creating excellent interactions.

The evaluation part is as important as empathy, and that's the data bit I discussed prior. I'll be addressing this next, and explain what it's important to never listen to your users.

Labels: , , ,


Blogger Kirk Battle said...

Great read, I've been playing Ocarina of Time recently and that nails what I've been trying to explain bugged me about it. I never played many games from the N64, PS1 era, so I'm not used to reading the graphics or understanding what it's trying to communicate. Now that I'm almost at the end, I'm used to it and don't need gamefaqs, but there was a while where it was a real chore.

April 19, 2009 at 7:07 PM  
Blogger Nels Anderson said...

@L.B. Thanks. It's interesting how the "vocabulary" of interaction in some games has quickly become an archaic dialect.

Old RPG might be the most extreme example of this, although that could be because most other games of that era were very linear and has to misinterpret. I remember playing the re-release of the original Final Fantasy on GBA and ye gods, I almost never knew where to go or what I was supposed to do.

Maybe that was part of the appeal, but I can't help but worry those foundational games are only going to become increasingly inaccessible. But then again, how many people read Beowulf or The Divine Comedy these days? Certainly not the folks at EA Redwood Shore (oh snap!).

April 20, 2009 at 9:27 AM  
Blogger Alan Jack said...

Excellent work Nels.

It brings up an interesting parallel with Clint Hocking's GDC talk on Improvisational game systems - there is a balance between predictable and chaotic systems that allows for maximum enjoyment. As you said, some systems are made to be confusing - if we knew exactly what would happen from our actions, it would be less fun.

This cuts into the core of "fun" which Ralph Koster explains is essentially an abstract form of learning - as he puts it "an exploration of a possibility space".

When you discuss how old games were harder to play straight-out-the-box, you have to remember that there was a time (and I'm talking mid 80s here) when game systems were so obtuse they required you to consult the manual before you could stand a chance of playing the game. Nowadays, even a dedicated "training" section is frowned upon. Some have argued this is impatience or laziness on the part of the player, but in reality its the simple case that "fun" should not incorporate the kind of learning that is more like work.

There is a delicate balance of readability with deliberately introduced chaos that produces a fun game. Looking forward to the "data" blog post now!

April 22, 2009 at 3:25 PM  
Blogger Nels Anderson said...

@Alan Looking at Far Cry 2, you can even see examples of good readability in its random systems. Rusted guns jam far more frequently than shiny new ones do. The guns' models communicate that very clearly, so the player understands they're using crap gear.

If the models for awful and pristine guns were the same, but they had different jam probabilities, this would confuse the player. Worse yet would be if broken/reliable weapons had models that didn't correspond to their chance to jam at all.

I realize I may have been a little unclear, as I tried to clarify here. A game with good readability isn't necessarily obviously, it is consistent.

April 22, 2009 at 6:12 PM  
Anonymous Anonymous said...

I can't agree strongly enough with this Nels. Empathy makes us better designers, but it also has the added bonus of making us much better collaborators.

Some people will argue that understanding what other people are thinking was the entire reason for us evolving our large brains. I wouldn't be surprised if it were true. Almost any human failing can be improved by improving empathy, as can almost any game design problem.


April 30, 2009 at 11:03 PM  
Blogger Nels Anderson said...

@Mike Every time I hear someone say, "Players are stupider and lazy these days. They just want easy games with big explosions" I reel and seethe. While it's certainly true at times, this is whinging more often used to excuse unengaging gameplay.

Nobody can control the audience. If something is working, the only option is to find a way to fix it.

May 1, 2009 at 4:56 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home