Video games are the only medium in which longer length became an inherent selling point. Is it any wonder that even their best narratives are generally bloated, flaccid, and poorly paced? And then combined with bland, repetitive grinding gameplay activities?

You can see a similar pattern in the serialized novels of the 19th century: When authors are paid by the word, they have an incentive to produce more words. But this impulse, at least, was counteracted by the fact that their readers still wanted a good story and weren’t particularly concerned with length.

Only in video games do you see media consumers focusing on length-of-play as an important feature in and of itself.

A couple years ago I thought this trend might actually be reversing itself as Final Fantasy XIII came under criticism for being too long. But it doesn’t seem to be sticking yet.

How much should a single character be allowed to accomplish in 1 turn?

What you’re looking for is a sweet spot.

On one end of the spectrum, you have: “That guy is able to do way too much before I get another turn.” (Munchkin Quest suffers from this, as described and fixed in my house rules.)

On the other end of the spectrum, you have: “I’m not able to do anything interesting this turn.” (Imagine a game of Monopoly in which you could either roll the dice to move or buy the property you landed on, but not both. The balance would not be appreciably altered, but I suspect the game would suffer nontheless.)

The reason games use interrupt actions — such as attacks of opportunity — is to widen the tolerance of the sweet spot: It allows you to design turns that are long/large enough to maximize the potential for “I can do something fun with my turn” while simultaneously preventing the “why can’t I do anything to stop him from that long sequence of actions?” problem.

The disadvantage of interrupt actions, however, is the complexity which arises from tracking the triggers which allow those interrupts to be used. (Attacks of Opportunity in 3rd Edition gave you a single interrupt action, but had a very lengthy list of possible triggers. Swift actions were later added to the game, giving you a plethora of interrupt actions to choose from; but these almost universally had the trigger of “whenever you want to do it”, which is very easy to keep track of. 4th Edition simplified the list of triggers for Attacks of Opportunity, but drastically increased the number of different interrupt actions in the game and gave most of them different triggering conditions — thus radically increasing the complexity of the game.)

THOUGHT EXPERIMENT: Imagine a 3E-style combat system in which every character gets a single standard action each turn. In addition, they have a single interrupt action which they can use at any time. (For example, to hit someone running past them or trying to run away from them.)

(If you wanted to keep full actions, you could allow them to be taken by any character using their interrupt action on their turn.)

The point is that you don’t define any trigger conditions for those interrupts: You don’t need to keep track of them (or try to avoid doing them). You simply use them whenever you want to and can do whatever you want with them. Theoretically, this would give you the flexibility to respond to actions on the battlefield as you choose, without the difficulty of trying to keep in mind all the possible trigger actions.

Certain effects could vary the number of interrupt actions a character has. For example, a haste spell might simply grant a character an extra interrupt action. (Although on a similar note, you’d probably want to prohibit spellcasting with an interrupt action. An added layer of complexity, but probably necessary for balance. There are almost certainly some other wrinkles to work through if you wanted to use a system like this.)

Obstacles in roleplaying games do not exist in order to prevent a PC from doing something. They exist in order to challenge the players to come up with an interesting way of doing it.

(This thought occurred to me as I was reading Flawless by Scott Andrew Selby and Greg Campbell in which they describe a diamond heist in ’76 in which the thieves tunneled into the vault from the sewer. They tested for the presence of a seismic alarm using an alarm clock and hauled away the excavated dirt in a Landrover they drove through the sewer tunnels. Then they welded the vault door shut from the inside and threw a Bastille Day looting party. That vault door didn’t exist to protect the jewels. It existed to make those thieves look cool.)

Not that long ago I wrote about the Long Con of DRM. As a follow-up to those thoughts, I talked about the fact that Valve’s Steam is often seen as an exception to the general vileness of DRM systems, largely because it (a) added value through instant delivery and by allowing you to access your Steam account from any computer in the world and (b) Steam is generally not as intrusive as it might be.

Valve apparently decided I hadn’t made my point about how goddamn awful DRM is strongly enough, so they decided to give me some compelling supportive evidence: As of September 1st, I can no longer play any of my games — games I have been playing for a decade — because Valve decided to retroactively make those games incompatible with my computer’s operating system.

Ah, DRM, how I despise you.

(And if you think I’m ever buying another piece of software through Steam then you must think I’m a goddamn idiot.)

The Playstation Move and the Kinect will both fail.

This isn’t because they aren’t worthy technology: The Kinect is potentially revolutionary and slagging the Move because it’s dupiing the Wii’s controller is like slagging the Sega Genesis because it duped the NES controller. It’s obviously true. It’s also irrelevant.

But they will fail. Because add-on controllers for video games will always fail.

THE SIMPLE MATH

There have been 40 million X-Box 360s sold worldwide. The cut-off point for the Top 10 games sold for the X-Box 360 is Fable II with 2.6 million copies. Which means that if you can sell your X-Box 360 game to just 6.5% of your potential customers, you can break into the Top 10 list (which would obviously qualify your game as a huge success).

Now, let’s assume that the Kinect is a huge success as a technology platform and sells to 25% of X-Box 360 owners. This would mean 10 million Kinects sold with somewhere between $1 and $1.5 billion in total sales. Huge success for Microsoft.

Despite this massive success for the Kinect, however, the developer of a Kinect game is still going to be struggling: In order to sell the same 2.6 million copies of a Kinect game, they now need to achieve a 26% market penetration.

In other words, under this incredibly rosy scenario for Kinect, a developer has a choice: If they develop a non-Kinect game, their potential audience is 40 million customers. If they develop a Kinect game, on the other hand, their potential audience is 10 million and they’ll have to literally quadruple their performance in order to achieve the exact same success.

That decision is practically a no-brainer. Which is why game developers rarely develop games for add-on controllers and virtually never bother developing AAA titles for them.

IT GETS WORSE

But in practice things are even worse for the Kinect.

The second best-selling game on the X-Box 360 is Call of Duty: Modern Warfare 2, which sold 7.5 million copies on the platform. But total sales for that game are almost twice that (at least 14 million, possibly more than 15 million). If it had been a Kinect game (and thus exclusive to the X-Box 360 platform) none of those additional sales would have happened. One of the most successful video games in history would suddenly only be half as successful.

Which means that a game developer doesn’t have to just quadruple their performance with a Kinect game, they actually have to increase it by seven- or eight-fold in order to match their potential success with a non-Kinect game.

THE (NOT SO) BIG PREDICTION

Of the two technologies (Move and Kinect), I trivially predict that Kinect will be more successful. Not because it’s cooler or more innovative (although that may attract a few developers in its own right), but because I believe it will be easier for designers to incorporate Kinect-enhancements into games which will not require the Kinect (and can therefore still be marketed to the total X-Box 360 market and ported to other platforms).

For example, in Assassin’s Creed 2 there’s a section where one of the NPCs suddenly stops talking to the protagonist and instead turns to the camera and begins directly addressing the player. (Which, in itself, was a pretty nifty bit of meta-narrative since you’re actually playing as the guy who’s playing the Assassin’s Creed 2 simulation.) The effect is pretty cool. But it would have been even cooler if the game had a Kinect-enhancement which allowed the NPC to look directly at me no matter where I was sitting in the room (or even follow me around if I chose to get up and move around).

If I was Microsoft, I’d be doing everything in my power to convince AAA developers to include these kinds of subtle “Kinect Enhancements” to their games. If they could pull it off, they might even find the magic bullet to disprove my prediction: Making the X-Box 360 version of every AAA title into the “best” version of that game would not only help to sell the Kinect hardware (since every game you buy would make the Kinect more valuable to you), it could also prove to be a potentially devastating blow for Microsoft’s competitors (turning even cross-platform AAA releases into something akin to a “semi-exclusive” for the X-Box 360).