Wednesday, November 10, 2010

Games and Learning, part 2

Greetings, fellow vagabonds. I hope you'll excuse the absence, as I was off canvassing in Bucks County, PA, trying to salvage a few candidates for the Dems. So it goes.

 Anyway, as we were discussing last time, games can best educate when they put you in someone's shoes, and give you a better sense of the decisions they had to make, moreover of what their decision space was. You need to have the freedom to see all of the options available, and to take those paths as best you can. Historical simulation games are, almost by definition, also contrafactual alternate-history generating games. If you want to see figure out what was in Napoleon's head at Waterloo, then set up the table with a couple of friends (Or, the computer alone) and play it out.

The logo of the
 most famous war game
 fail of all time
.
But there are a few wrinkles involved. Particularly when it comes to wargames. Despite their attempt to include Clausewitzian concepts like "fog of war," a lot of wargames basically presume perfect command, control, communications, and intelligence. (This was part of the problem with Millenium Challenge 02, from what I gather.) When we play on computer, every epoch of war is translated into our modern information-centric model, where the commander has perfect knowledge of what's going on and perfect confidence that his troops will do what he says. This is because it's built into the interface to be responsive and transparent.

But any student of military history, (or indeed any warrior today) could tell you that reality is not necessarily so. You may issue orders that may or may not be interpreted correctly (rivers of ink have been spilled over whether Richard Ewell misinterpreted Robert E. Lee's command to "take the hill, if practicable.") I may be uninformed, but I haven't seen any really good way in games to model unreliable communications and signals intelligence, despite the fact that entire campaigns have turned on such things. And with a few exceptions, there seems to be little emphasis placed on lines of supply and communication, cutting of which is one of the main points of modern war.



 Interestingly enough, Guy Debord, after retiring from the Situationists, went on to develop a game based on Clausewitz that placed major emphasis on lines of communication. I haven't played his "Game of War," but I certainly intend to.

All I know, is that I don't know
if he's going to zerg rush.
Perhaps it's a problem of signalling--oftentimes you don't know when your information is bad. So it might just seem like the dang system isn't working right. If it gives you straight-up question marks, then you actually know that you don't know, which means you do know something.

In some ways, Starcraft is a better simulation of the cognitive experience of leading an army than Civilization or Axis & Allies. There's resource depletion--if you run out of vespene gas, you're kind of screwed. There's fog of war, of course, but there's a real use for reconnaissance, because you as a player have to deal with the uncertainty of "What's he building in there?" Because once intelligence fails, psychological speculation fills in the gaps.

When approaching an opportunity that might be a trap, you must ask, is my opponent the kind of man to trick me, or is he simply foolish? That personal side of generalship, sizing up the opposing commander, his forces, and his intentions, starts with Sun Tzu and continues unabated to the present day. Gladwell devotes a good part of the chapter on Millenium Challenge 02 in Blink to discussing Paul Van Riper's appreciation for how and why Joe Hooker balked at Chancellorsville. There's constant epistemic doubt about what you think you know, and Lee was a master of inducing doubt, so much so that McClellan was convinced that he was facing armies three times their actual size.

So, getting back to games and learning, games definitely do have an opportunity to give us "Extra Lives," to give us mental experiences without the attendant consequences. Thus the entire 'simulator' industry, which kind of doesn't want to admit that they're basically games. A friend of mine in medical school tells me that they're doing lots of simulator work, not only for operations, but also in RPG-style dialogs with patients. That makes a lot of sense to me. On another front, one possible reason for the current profusion of true-freshman quarterbacks in college football might be that they played a lot of Madden.

This is why I'm looking with interest at the Game Show NYC, coming as it does with hopes of linking art, education, and games, and doing so under the theoretical aegis of Dewey's "Art and Experience." I may just submit an application, if I can get the right people together. I'm no Terrorbull Games, but I'm kicking around a couple of ideas.

1 comment:

Peter Shafer said...

Final Fantasy XII may be an interesting micro-level example of the idea of "lines of communication" in games. Final Fantasy has always been about giving instructions to parties of three to six characters that make up your party. And up until FFXII those commands were explicit. XII, however, increases the pace at which these commands must be issued, and to support the player the game provides the "gambit" system of to issue conditional orders in advance.

Each party member (three active, three in reserve) is given a set of conditional instructions. For instance: if anyone's health is below 50%, heal them; if the enemy is of a fire variety, attack with ice magic; and if none of these conditions are met, just stab at targets. The player is limited in how many of these conditional instructions can be issued, and he or she has to make a lot of assumptions about when those conditions are activated, as well as how the character's AI will behave when they run in parallel.

You have to issue these commands where each character fills a certain role (e.g. attack, heal, absorb damage, etc.) But the only way you really know that what you think you've told a character is what that character will actually do is to actually use them in battle. The challenge throughout the game is to better understand how you need to convey your strategy to the AI, and acquiring better instructions/conditions that allow them to behave more intelligently.

Final Fantasy XII is far from anything resembling modern warfare, but I think at a high level, it taps into those same sorts of uncertainty you're writing about between the commander, an army, and their enemies. It have been interesting to see the gambit system expanded upon to encompass larger units, or even to incorporate multi-player, competitive battles.