PC Games To Help Public Policy Initiatives

Ben Sawyer writes: "The Woodrow Wilson Center's Foresight and Governance Project has published Serious Games: Improving Public Policy through Game-Based Learning and Simulation, a whitepaper. The paper illustrates how government agencies and non-governmental organizations (NGOs) can utilize game-based techniques, technologies, and approaches to produce innovative simulations, models, and game-based learning products that enhance public policy decisions. The Woodrow Wilson Center is distributing the paper on-line to a variety of agencies, organizations, and game developers to help foster greater discussion and cooperation between key public policy makers and game developers. Interested readers can find the homepage for the paper here."

Comments

In other news...

[jaxdahl]

Maxis has announced its new MMORPG, 'SimCity Live!'. Rumors are that it was brought up from the ashes of the Sims multiplayer addon. Participants will be able to live out a day of the lives of their favorite politicans. Be the President of the USA and get a lollipop. Be the Mayor of New York City and be the hero of the day. Be the governor of Minnesota and body slam voters! Will says "This will bring the hard money into people's own homes!".

Seriously, public policy? games? No, I didn't think so.

Re:In other news...

[Anonymous Coward]

Strange... the more optical networks to my city the more votes I get... gosh, thanks Maxis Telecom Inc.

Yea Right...

[mlknowle]

...The guy just wrote it because his boss caught him playing Quake!

New game for congress

[kir]

WOW! Now maybe someone will create a "learning" game that will teach congress that they can't take money from the MPAA/RIAA/etc. and give them whatever they want.

Soon to follow this game's release - a "learning" simulation for the Patent Office (just guess)!

Re:New game for congress

[NonSequor]

WOW! Now maybe someone will create a "learning" game that will teach congress that they can't take money from the MPAA/RIAA/etc. and give them whatever they want.

Will it make use of "force feedback?"

Re:New game for congress

[kir]

Will it make use of "force feedback?"

We can only hope!

PATENT OFFICE: gives stupid patent to XYZ Inc. on "their" method of powering on a PC using a unique device called the "power button".

FORCE FEEDBACK: ZZZZZWAPP!

PATENT OFFICE: OUCH! Uhhhh... nevermind. Our mistake. Repeal that last patent.

Re:New game for congress

[Anonymous Coward]

Yea, that'll be cool. However, to really impress such learning upon congress, I think some kind of force feedback implementation might be necessary. Or maybe just an on-screen "job security" bar. God knows the "right thing to do" and "public representation" bars have no effect.

SimPublicPolicy

[MiTEG]

This reminds me of the good old says during Maxis' peak, when they released a Sim version of everything, from SimTrain to SimFarm to SimCity, etc. My favorites were SimTower and SimCity 2000, and I think both could be easily tweaked to become more related to public policy.

In SimTower, you were the owner of a high rise building, and your task was to build the biggest possible tower while still pleasing everyone (elevator congestion, pricing, etc). An important part of public policy. Also, in SimCity, you took on the role of a city manager, and if that doesn't relate to public policy, I don't know what does.

Re:SimPublicPolicy

[sharkey]

SinPorkbarrelling, perhaps? Or how about SimKickback? SimBribetaking?

Or, more in line with the racial profiling tactics being kicked around for airports: SimJimCrow.

It's What We've Always Known

[Spencerian]

We can foster world peace and greater understanding by using the virtual tools at our disposal:

A railgun, rocket launcher, attack shotgun, and lots of Quad Damage. Oh, and since we need to simulate reality as we want it: Lots and lots of hot, redhead girls with infinite anime-style hair--with body armor and a willingness for world unity through, oh--I don't know--sex and blowing things up.

And this time, when They say, "Impressive", you WILL be.

Re:It's What We've Always Known

[mheckaman]

I don't think you're actually that far from the truth. I personally believe that video games such as Quake are great for releasing our stress in non-violent ways. I can remember as a young teenager getting upset and hitting walls from time to time, but now when I get upset, I just fire up Quake III or Red Faction and blast things around for an hour. When I'm done playing, I feel calm and relaxed. This is hardly scientific evidence, but I know that at least for me, these games provide a great outlet for violent impulses (which all humans have!) when I'm angry.

As for the sex, check out this: http://bodhibkb.tripod.com/bodhiq3a/id1.html

Matt

SimKickback

[FakePlasticDubya]

Hmm, perhaps a Sim that teaches government officials how to avoid corruption?

Sort of like what we tell the kids, you know, "Just say no!"

Re:SimKickback

[colmore]

How about "Winners Don't Take Soft Money" on the Congressional Mortal Kombat machine?

Riiiiiight.

[bobetov]

Not only is this a brilliant way to get funding for "research" that only a 13-year-old Sims fan could love, but it's clear that they have no idea what they're talking about. A good example:

"Not only is the game development community at the forefront of PC-based visualization, it is also a leading developer of applied artificial intelligence... blah blah"

Hahahaha. As a game developer myself, I can tell you exactly how leading edge game AI is. Let's all say it together now... Table Lookups!

Woohoo. Games are games. Simulations are simulations. Games are fun, simulations are not.

Bleh.

Re:Riiiiiight.

[flynt]

It's hard enough finding ACTUAL intelligence on a Wolfenstein server these days!

Re:Riiiiiight.

[bobetov]

Heh. You could almost make the case that games are not only not leading-edge artificial intelligence, they're proof positive that natural intelligence isn't all it's cracked up to be either, d00d.

Re:Riiiiiight.

[sharkey]

Games are fun, simulations are not.

Obviously, you haven't checked out the simulation called "Virtual Valerie".

Re:Riiiiiight.

[Bodrius]

Although I was about to point to "Black & White" and some others, I guess they're about as representative of the state of AI in games as Infocom-adventures are of the importance of substance over flashy graphics.

Re:Riiiiiight.

[carbon1]

Yes, and IF is some of the most substantial gaming there is. I've actually been surprised by plot twists in a few particularly good IF entires at the annual competitions.

Re:Riiiiiight.

[MAXOMENOS]

Let's all say it together now... Table Lookups!

What? That's it? Not even a heuristic function thrown in for good measure? What a rip!

Re:Riiiiiight.

[Rogerborg]

• As a game developer myself, I can tell you exactly how leading edge game AI is. Let's all say it together now... Table Lookups!

You lie! Sure, I also thought that I was coding opponent behaviour as a simple state machine, but my team leader, my development house project manager, the publishing house project manager, and the publisher's marketing people all insisted that our game had "Ground breaking neural net based adaptive AI!". Some of these people owned suits, for god's sake. What do people like you and I know about AI design and implementation compared to experts like that? ;-)

Re:Riiiiiight.

[Arsewiper]

As most people can't use a bus timetable your lookups may be considerd advanced AI. Artificial stupidity is much closer to the real thing.

Re:Riiiiiight.

[ksquire]

On the other hand, games like Black & White and The Sims are a bit more advanced. With games like Thief or name-your-favorite-flying simulator, there's a lot more grey area between simulation and games than you're suggesting.

It is not only for public policy

[Edmund Blackadder]

Capitalism 2, for example is an amasingly intricate bussiness simulation.

I cannot say whether it is accurate or not, since i have never been the CEO of a multimillion corporation but some bussiness schools say it is quite useful.

other good PC games

[I Want GNU!]

Civilizations 3 is another great learning game. It gave me valuable life experience for the next time I have a 6000 year life span and an empire to build. When this occurs, I would like to be the Persians, because their special unit the Immortals will help me conquer competing civilizations. I also have learned that when I conquer foreign cities and they are unhappy, if I simply make them entertainers and they starve for a few turns, the city size decreases and they become happy again.

Re:other good PC games

[Bodrius]

It also taught the developers that using computer cheats to increase the difficulty level is much easier than coding any AI, and reviewers will still praise you for substantially improving the AI of your computer game.

Re:other good PC games

[NonSequor]

They still have to code *some* AI. All of the cheating in the world won't help if the computer just does nothing every turn.

Um..

[The Evil Troll King]

Perhaps public policy makers should stay away from gaming technology. Look what it's done to us. Imagine the headline:

Orrin Hatch changes name to "DethGod", vows to "V3T0 J00R A$$3$"

On second thought, that might be kind of cool....

Steve

Re:Um..

[Have Blue]

It would certainly make watching C-SPAN a lot more interesting.

Great idea - game theory is very insightful

[Seth Finkelstein]

It's a great idea. It's long known that game theory such as The Prisoner's Dilemma can yield a lot of social insight. As that page details:

This classic problem of game theory sheds light on many of the problems that have plagued ethical and political philosophers throughout history. It addresses that class of situations in which there is a fundamental conflict between what is a rational choice for an individual member of a group and for the group as a whole. It helps us understand how such dilemmas can be resolved for the greater good.

Putting these ideas into computer games can make the topic less abstract, more immediate and clear.

Sig: What Happened To The Censorware Project (censorware.org)

Re:Great idea - game theory is very insightful

[Arrian]

Game theory is a really interesting and helpful at analyzong real world problems, but has little or nothing to do with games as we know them. It has more to do with why someone always gets picked last at kickball than how to play SimFarm.

Personally, I think that though using a game style interface to simulations may make them more accessible or easy to understand, I doubt that "the establishment" will take anything seriously that is related to a passtime, at least for quite a while.

Re:Great idea - game theory is very insightful

[sheetsda]

Game theory is a really interesting and helpful at analyzong real world problems, but has little or nothing to do with games as we know them.

I'm not sure I agree with that. I think the AI in most present day games could benefit greatly by adding a bit of game theory. Take the Prisoners Dilemma, give an AI a worst possible alternative of dying (or whatever is appropriate for the game), and have it try to minimize that possibility (provided its primary motive is survival). I think you'd see some great improvements in simulating the way humans play games. Most of the games I've played recently have monsters that are somewhere between fearless and stupid, which doesn't make for a very interesting game. It tends to be: alert monster(s) to presence to make them move, lure monster into trap; they all fall for the same tactic. Now show me a monster who throws down some fire to make me take cover when he sees that BFG9000 in my hand then runs to get the pack of bigger monsters, who then try to set up an ambush, or run even further to alert still other monsters, and I'll be interested. Half-Life had some stuff like this in its human enemies, taken to its extreme I think you could simulate real humans and other inteligence very well.

Re:Great idea - game theory is very insightful

[Arrian]

Yeeeaaaahhhhh, sorta, but what I remember of the theory I from college, it doesn't apply to the flowing, loosely defined world of FPS's very well.

What I remember of game theory is that it is concerned with making decisions between set options to come to an acceptable outcome, and other people, their decisions, and their acceptable outcomes may or may not be taken into account. I know this is very vague and simplistic, but I don't have access to any books at the moment for a more concise and possibly accurate defination.

I see it more directly applicable to strategy games like chess. If I move my rook here, how likely is he to take it with his night allowing me to take his bishop with a pawn, and is it worth it at this point in the game. Brute force simulation may achieve the same or better effects against average Chess players on a PC, but more complicated strategy games (say an Axis and Allies implimentation) may have too many variables to simply simulate out all options to the Nth move and choose the best one.

In the fast paced flowing situation of an FPS, I doubt this type of calculation is feasable. Also, I don't know of anything on game theory that would suggest coming up with new ideas, it is about choosing between different options.

Where I do see an application for game theory is in the design of the AI: in giving it options and what the criteria to base the choices between those options are. As a monster, how important is it to me to, A: stay alive; B: Kill the player; C: prevent the player from accomplishing a certain goal; and D: how well do I work with other monsters? The developer them develops sets of tactics and determines how they relate to the monsters' motivations. For values of A between 0 and 50, group 1 tactics (j,k,l) are preferred, group 2 tactics (m,n,p) are neutral, and group 3 tactics (q,r,s) are right out. For values of A betrween 51 and 75, group 1 tactics are neutral, group 2 are preferred, and group 3 are neutral... etc. for all variables. Then, when an engagement is about to happen, the AI can say this monster has scores of A=52, B=37, C=89, and D=92, so tactics x,r, and j are preferred, and l,m,s and v are acceptable, but j,s, and v only apply underwater. Therefore we will give x and r each a 40% chance, and give l and m each a 10% chance, generate a random number and pick a winner. For the rest of the fight, we just follow the rules of the chosen tactic and get blown to hell by Gorden Freeman anyway. So yeah, game theory probably can be used in the context of an FPS, but in a stricter sense it is probably better applied in the design of the AI/bots in the game, and most likely, it is applied in the design of all the games we run into. More accurately maybe, it describes what the designer is doing when he designs his AI. Whether a better understanding of the theory would lead to better AI's, I really don't know.

Re:Great idea - game theory is very insightful

[Sylver Dragon]

What might help this situation is if you could allow the stated monster to re-evaluate its situation during a fight. As the old addage goes, "the best laid battle plan rarely survives contact with the enemy". This is often the case with the normal FPS type monster. The monster picks a tactic, usually something along the lines of all out attack, or suppressive fire, but once the player realizes what the monster is doing the player adjusts their style of play to fit that situation, usually to devistating effect. Also, there is a bad habit of making the monsters use some really stupid tactics, I don't know how many times I have applied the old "side-slide out, fire a few rounds, and duck back" routine. This is a good routine, but normally a good player will pick up on it and either time me, or throw some sort of area-effect weapon. Also, I've noticed that there seems to be a lack of group tactics, simple rolling would be nice, not to mention the concept of cover. The point of all this is that, not only does the computer need to be given a good set of tactics to choose from, it needs the ability to re-evaluate its choosen tactics during combat. Say, every few seconds, and whenever the monster, or one of its cohorts takes a hit. I realize that this may be logistically difficult to impliment, I am not a programer, so I don't know what it would take, but it strikes me that this would be a good thing to pursue.

Re:Great idea - game theory is very insightful

[Arrian]

I agree to an extent. The monsters in FPS's could be improved by allowing them to take more factors into account in their tactics. For example, pop out from cover for quick pot shots as long as the opponent stays in one area, but move to a different position if he gets hit or almost hit twice in a row, or stand in the open and blast away when the opponent is carrying a knife, but when he switches to a BFG, get under cover and attempt an ambush.

I'm not a developer, but I think to do this, you need to get the monster to follow a set of rules. The monster acts in a certain manner until something happens, then it re-evaluates the situation and begins to behave in an appropriate manner.

For example, if a monster begins an engagement by popping out of cover, blasting the opponent, and diving back under cover, several things may happen. The tactic might be successful and he can keep blasting until the player is dead. He pops out and is immediately hit by the player. He pops out and the player is nowhere to be seen. He pops out and he sees the player running away or ducking behind cover of his own. He pops out and the player is nose to nose with him. These are the events the developer needs to make to bot recognize and adapt to. Furthermore, there are likely to be qualifiers to make the bot behave differently in response to the same event: The bot's health, how much damage the opponent has taken, what weapons are available to the bot, what weapons the opponent is using, what type of terrain is the bot in, what is the bot's mission (kill the player, defend a location, find the player, delay him, etc...) After the developer determines these, he makes rules for the bot to follow; when in x situation and y happens, with a,b, and c modifiers do this, or with e,f, andg g modifiers do that.

In theory I think this method could produce very effective bots, but there would be a direct correlation between the time the developer spent coming up with possible situations and respoonses to the quality of the bot. There are probably quite a few problem with a bot made in this way. First, it potentionally requires a
computationally intense decision process every time anything happens. Next, it requires the developer to spend a lot of time just coming up with the large number of rules necessary to provide a decent amount of tactical flexibility, let alone implement them. Also, many of the monster would need to be tweaked or individualized even further because, even if they are tough to beat, monsters that act the same way all the time still get repetetive. Also, I don't think the moajority of the monsters in a game should be all that good. Imagine trying to work your way through Half-Life if ALL the monsters were as good as the average human? It might not be so important now, but this would require a large amount of memory and storage for tables of rules. Most importantly, again, is that the monsters will require a lot of work on the part of the developer to get a large enough set of rules to first make monsters adaptable to different situations, but to give them multiple reactions to each different situation. After all, if the emonster reacts the same way each time it gets into a certain situation, we get back to the original problem of humans being able to adapt to something and bots not.

I think this method may, theoretically at least, be able to produce bots that can react to situations almost as well as average players at least, given enough groundwork is done by the developer. I have no idea how it would be implimented, or even if it could be, though. If I were a programmer, it would be fun to try, though. And, adding teamwork to the bots would probably increase the complexity exponentially.

Re:Great idea - game theory is very insightful

[Sylver Dragon]

I wonder if it would be possible to set up some sort of needs based intellegence. Insted of the rule based intellegence. For example, assign interger values to certian events. Say, Dying is -1000, Killing the player is +1000. Then add in other minor values to allow for further descision making. Being covered is say +10*%body covered. Objective failure could be rated on a scale from -1 to -1000. Each of the players possible weapons could constitute variables on the cover number, say, knife is *0, Handgun is *1, shoutgun is *2, BFG is *50. Having a team-mate is +100. And so on. Basically reduce the behavior of the montser down to a simple interger equation. Then set the rules for the monster to occasionally check this equation for a higher possible number. One might also want to set certain events to force a recalculation (e.g. Taking damage, player changeing weapon, etc.) Also, to add variability into monsters you could give them a personality, attributes such as courage, loyalty, and common sense. These could then modify the weight for the various variables. So insted of a masive rule table, it becomes a variable table, which is then run through a standard equation, I would stick to all interger mathematics to help with speed issues. This might also make a nice way to work with monsters in a level editor. You simply set the attributes of the monsters and let the math do the rest. But again, like you said, the amount of time a developer spends working on the initial setup is going to have a huge effect.

abandonware

[I Want GNU!]

Does anybody else find it funny that on the case study links page, they give a link to download SimHealth. It leads to a defunct abandonware site but I think it makes a statement that a group like this considers abandonware ok to use. Just a bit of SimFood for thought...

Re:abandonware

[Anonymous Coward]

heres a working abandonware link for simhealt: ftp://ftp.anet.net.tr/windows/oyun/simhealth.zip

Serious Simulation

[C. Mattix]

This company is across the hall from mine. They do serious simulations. It is actually pretty cool stuff.

From their webpage:

Synthetic Environments for Analysis and Simulation (SEAS) is a business and an economic war-gaming environment developed at Purdue University in close association with the Department of Defense. SEAS, LLC replicates the "real world" in its most crucial dimensions including competition, regulation, decision variables, and interaction dynamics. It consists of inter-linked goods, stocks, bonds, labor, and currency markets. In these markets two types of agents interact: Live: People acting as buyers, sellers, regulators, and intermediaries.

Virtual: Artificially intelligent software agents mimicking human consumers in a narrow domain.
These agents allow the environment to achieve both depth, through human agents, and breadth, through virtual agents, in representation of the economy.

They are at: http://www.seasllc.com

Re:Serious Simulation

[Starcub]

Yeah yeah, but I bet they couldn't predict that the Patriots would win the Superbowl!

Re:Serious Simulation

[Gorak]

It consists of inter-linked goods, stocks, bonds, labor, and currency markets. In these markets two types of agents interact: Live: People acting as buyers, sellers, regulators, and intermediaries.

Well, they seem to be either slashdotted or not responding, but either way ... I bet they don't take black market factors into account. Look at the War on Some Drugs -- $65 billion dollars per year, and that's only in the US domestic market. Factor that in, and then see what happens when you eliminate the spending on the war, and the prices come a-tumbling down!

Re:Serious Simulation

[C. Mattix]

Actually, I believe they do factor in Black Market. It is just another set of agents, though ones that don't follow the same rules as the rest of the 'virtural society.'

Re:Serious Simulation

[dynoman7]

Dark alley on south side of Inter-linked City...

Shadow Agent: Psst..wanna buy an Inter-linked watch?
Tourist Agent: Ummm...That doesn't look like a real Inter-linked watch. How do I know it's_

FBI User in training, gun drawn (slightly pixelated): FREEZE, DIRT BAG!! PUT DOWN THE FAKE INTER-LINKED WATCH!!!

...

It's already happening !

[Jesse Duke]

"government agencies and non-governmental organizations (NGOs) can utilize game-based techniques, technologies, and approaches to produce innovative simulations, models, and game-based learning products that enhance public policy decisions."

John the IRS director [playing SimCity]: There, if just raise the tax a little, I should get more money to build a road [5 seconds later] Of and fuck, it's not fast enough ! [cranking up the tax rate to 90%]... Now that's better !

Margaret the Secretary [knocking on the door]: Mr. Director, President Bush wants to know by how much you estimate he will be able to increase governmental spendings.

John : Err... well, I haven't really had time to ... [Glancing at the screen, smiling] Actually, tell him he can double it easy does it !

Margaret : Very well Mr. Director.

John [looking at the screen again]: Aah *CRAP*, not again, all my nice residential areas are turning into shanty towns again. wtf? This game really sucks ...

Note idea not exactly new - "Diplomacy"

[Seth Finkelstein]

But note, this idea regarding gaming is not exactly new.

The Game of Diplomacy has a history long pre-dating computer games.

Sig: What Happened To The Censorware Project (censorware.org)

Re:Note idea not exactly new - "Diplomacy"

[Starcub]

"Movement at Diplomacy is controlled by written orders, which are exposed and executed simultaneously by the seven players. Each player writes his orders in secret"

Interesting. This reminds me of a Star Trek based board game I played in College many years ago. It was also a turn based combat simulation game where you had a certain number of points to allocate to various ship resources. Forgot what it was called though...

Re:Note idea not exactly new - "Diplomacy"

[Soulfader]

StarFleet Battles. Resurrected for the PC under the name StarFleet Command.

[OT] The Woodrow Wilson Center?

[Anomolous Cow Herd]

I'm sorry, but, although I am an Iranian-American, I know more than my fair share of American history. And I wouldn't trust anything that comes out of an institution with that vile individual's name on it, much less associate myself with it.

Woodrow Wilson, for those of you reading this from outside the U.S., was one of the most markedly racist presidents of the 20th century. Not only did he openly support the ideology and actions of the Ku Klux Klan, but the first ever screening of "The Birth Of A Nation," a blockbuster film of the era, was held at the White House. The Birth Of A Nation was essentially white supremicist revisionist history that portrayed the Reconstruction-era South as run by black tyrants who conspired to opress the white man and rape the white women. It was a true piece of filth, even considering the times.

As a non-white, non-Anglo-Saxon, and a Muslim, I cannot possibly condone anything that would spout forth from an office that bears this man's name. How many coloreds had to die at the hands of an angry white mob because of a president's explicit approval?

Microsoft Flight Simulator

[jsse]

has been an important educational tool for some islamic people for quite some time.

The first thing that popped into my mind...

[BitwizeGHC]

... was a sort of MMORPG that simulated a legislative body of some kind, with players acting out the roles of legislators, and receiving feedback from the game on how their actions affect the "populace". Kind of like Boys' State but IN TEH CYB4RSPAECE.

Re:The first thing that popped into my mind...

[Anonymous Coward]

This is actually the first decent idea I've read. trying to model the world is doomed from the start because no one understands the world.

It's far more realisitic to build a MMORPG to trial government policy.

Microsoft Age of Empires

[guttentag]

This is not news... Microsoft caught on to this years ago.

They released a "Fun" version to the public that is set in 10,000 AD and called it Age of Empires. Bill Gates has his own personal version set in the present day... it's called "Windows." Why do you think the government is so determined to get their hands on the source code? They want to play too.

Brainwashing...

[the_skywise]

I can tell you exactly how these games are going to work out:

A Peta game would be like this:
What do you want to eat for supper?
a> Beef.
b> Chicken.
c> A nice wholesome salad with walnut dressing.

a or b> YOU DIE! YOU'RE EVIL! Play again?

A Republican environmental policy game:
Do you lower the emissions controls for the coal power plants, knowing that it's still a level of environmental protection?
a> yes
b> no

b> USAMA BIN LADEN TAKES OVER THE WORLD THROUGH OIL CONTRACTS. YOU DIE! Play again?
(If the democrats offer the game and you press A> YOU'VE NOT DONE ENOUGH TO SAVE THE WORLD! YOU DIE FROM LACK OF OXYGEN! Play again?)

But go ahead, play your games and get "informed"... that's much better than actually hounding your senator about taking Enron/MPAA/RIAA/Health industry/Microsoft's money and letting them dictate law...

Yes!

[Odinson]

I can finally put my Civilization conquests on my resume!

Re:Yes!

[Monkelectric]

exactly! god *wont even take my application* for assistant deity, but based on my villager-killing experience in B&W, satan has offered me a long term position :)

SimDictator

[KingJawa]

The Sim packages would be good for political insight -- if you believe that dictorial decisions make for the best political system. Could you imagine a mayor who unilaterally bulldozes industrial zones for a park? Or if the city would brown-out if the Mayor didn't like nuclear power? Or if, like me, you wanted a stadium near your house?

Re:SimDictator

[kiwipeso]

If you want a stadium near your home, try living in Wellington, New Zealand. We have a Stadium by the Railway Station and Harbour.
http://www.WestpacTrustStadium.co.nz/
T his stadium is in the Thorndon suburb, which is populated with victorian era houses for the wealthy and embassies.

Another point to note is that wellington is currently about to bulldoze a former industrial zone to make a central city park, even though the mayor campaigned against it.

Robocode

[jsse]

It's good to see a larger deployment of games in educational ground.

I've been challenging my students to design a Java robot to beat the best robot in Robocode. Even though I failed get the ed board approving my idea of using game for teaching, the students really enjoyed learning Java thru gaming, regardless of the fact that no bonus mark would be giving to winners.

Now this paper can be a very strong support during next time I face the dinosaur ed board. :D

Simulations Not Always Helpful

[Logic Bomb]

As usual, the comic strip Doonesbury is way ahead of the curve. Check out a week's worth of strips starting on April 12, 1982 . Obviously, computer simulations of social phenomena can be more or less productive.

Re:Simulations Not Always Helpful

[sharkey]

Ahhh...War Games.

How about a nice game of chess?
No, let's play Global Thermonuclear War.

Hell, I'd piss on a sparkplug if I thought it'd help!

Re:Simulations Not Always Helpful

[Anonymous Coward]

If anything that showed how helpful simulations could be.

But it's a pretty cool comic none the less...

Interesting possibilities...Becoming "official"...

[RyanFenton]

At least according to most of the CSPAN I've ever watched or listened to while programming/gaming, as far as sources of technical information, there isn't much emphasis put on actually being *qualified* for someone trusted to give information to polititians, so much as someone who can seem "official".

Currently, most "official" information on computer-related matters getting to polititians comes from interested parties with lobbyists and the like. Occasionally, resourceful polititians will contact professors and others when a debating point is in question - but for the most part, it is convenient to just talk to the same people who are there and seem qualified and eager to speak right away.

Regardless of campaign finance, this will always be the situation*. Now, if game designers and other people closer to the programming angle of things get to show the effects of laws - they gain credibility in the eyes of polititians, right or wrong. Their simulations could possibly show them simplified answers to questions quicker than even the paid lobbyist can explain.

Ethically, one would need to show every point in the logic of any given simulation where the results could be flawed, or have margins for error, or where complications are ignored - much unlike what a random lobbyist would likely show. While most polititians don't like to be called "technocrats", they also seem to like feeling they have depth to the information they are presenting.

If more programmers could spend some time helping polititians, perhaps there wouldn't be so much a distance between the groups as there are now.

:^)

Ryan Fenton

* Though I am fervently in favor of campaign finance reform

I have a policy

[Renraku]

Teamkillers should be banned.

Game Theory

[Nergul]

Well, perhaps this is a good Idea. In fact, Games can easily help teach Economics to policy makers.

Economics at the start is actually quite simple. The more people want something, they more they will pay, the more it costs due to taxes, the less they buy, etc. AIs can easlily manage to do this even if they are only Table Lookups. But, policy makers might have a use for something like this.

Take a basis as something like one of the Corparation Games that is mentioned in the article. Take this, modify it a bit, and you might have a bit of a simulation. Such a simulation could be used to help the people to understand the basis behind thier desisions. Then, the explanations of Tax Cuts and thier math might not be:

"If you have billions of dollars, and spend billions of dollars in Tax Cuts, you still have billions of Dollars."

Perhaps the Canidates for President of the US could use such a game to make thier Budget plans. You just create a current state of the economy in the game, and then demonstrate the effect of the policy by way of a script file that could be given to the people. Then, we could understand thier budget plans.

Then again, policy making does have more to do with Fundraising dinners then Economics(or Lesiure Suite Larry in Clinton's case), so I guess that is out.

We already play games.

[Zapdos]

Or didnt you know that Enron had a code of ethics Manual.

We need more political sims

[Y-Crate]

When was the last time a real political sim came out? Go on, think about it.

Well, it's been a while. That prettied-up version of Civilization 2 (aka Civilization 3) sitting on your hard drive is not a political sim. It's been years since one has appeared. And the trend has been towards dumbing-down the political aspects of non-political games to the lowest-level possible. Just compare the relatively complex diplomacy of Sid Mier's Alpha Centauri to the Fisher Price foreign-relations of its successor Civ 3.

Political sims are valuable tools. They are a wonderful way to learn about the way governments work and the complexities of foreign relations that will not even be explored my most textbooks.

But they don't sell as much as SimCity 3000, Civilization 3 or Quake 3.

So, they aren't made anymore. Even the old mainstays of semi-intelligent gaming have fallen victim to the effort to make them more "accessilbe" (read: severly-dumbed down) to sell as many copies as possible.

I know it is a bit of a cliche, but back when graphics were far less detailed and 3D engines were unheard of, for a title to be considered unique, it had to rely on depth of gameplay, not color pallate.

Not that I have anything against bright, 3D games. I enjoy playing them everyday, however I think it is terrible that such enjoyable and valuable software has vanished and can only be found at Abandonware sites.

People are always skeptical...

[AugstWest]

...especially when the easiest humor lies down the skeptical path.

But I've learned a lot about the way life works from SimCity, and I've learned quite a bit about business from Sim games in general, especially Monopoly Tycoon.

It won't teach you everything, but you could very easily teach Economics 101 with a computer sim, if comeone actually put the thought and effort into it.

Re:People are always skeptical...

[Anonymous Coward]

SimCity is a common modeling tool in higher level economics classes.

Then you have learned nothing

[Anonymous Coward]

It makes much more sense to burn your conquered cities to the ground and bring your own people in to settle the site. That way, the city will not decide to join its former empire, even if that empire now consists of one city on an island in the South Pacific. If the Germans had caught onto this in WW2, they never would have had to deal with the pesky French Resistance, or Joan of Arc killing their spearmen with musketeers, thereby initiating a French Golden Age.

Re:Then you have learned nothing

[Anonymous Coward]

It makes much more sense to burn your conquered cities to the ground and bring your own people in to settle the site. That way, the city will not decide to join its former empire, even if that empire now consists of one city on an island in the South Pacific. If the Germans had caught onto this in WW2, they never would have had to deal with the pesky French Resistance, or Joan of Arc killing their spearmen with musketeers, thereby initiating a French Golden Age.

Not so. It only takes a few turns to starve them and as long as you don't take cities that are in the middle of their cities, it won't rejoin, and after starving people of your culture will be born. I haven't had any cities leave me yet. Plus, settlers take away from your normal cities and are costly and time consuming. Faster just to take it.

Re:Then you have learned nothing

[Anonymous Coward]

Has anyone ever told you that you lack a sense of humour?

Re:Then you have learned nothing

[NonSequor]

It would seem that you have been reading Machiavelli lately.

Why bother with sims?

[BillShatner]

I propose they do away with the simulation ideas, build an army of robots and control them remotely through a modified Quake 3 interface, release it as a mod, and take over the world.

I will call it The Allan Parsons Project.

Libertarian Society (Training Wheels)?

[resistant]

This is actually the technological approach I've been studying carefully for setting up a virtual libertarian society in which instead of being a distant pipe dream, it's an "actual society" (simulated, but society is a state of mind anyway, and this would be a way to see the other state of mind as very possible and desirable).

It's a damned big job, though. I'm hestitant to even try for real without interest from other parties. There's just too many practical details. Everyone else seems to have been studying "variations on the status quo" (meaning massive corporatism, statism, etc.).

If you've been thinking about the same sort of effort, drop me a line at "engineer at-sign meme hyphen engineer dot com". :)

It is accurate

[Anonymous Coward]

I bought a silica mine and two dairy famrm and now I have the bottled milk market cornered in New York.

Tokyo here I come!

Just imagine!

[sharkey]

If our public officials used Leisure Suit Larry as a training tool? Or have the Kennedys already done so?

Perhaps the Ashcroft used 3D Monopoly to help in crafting the Microsoft-DOJ antitrust settlement.

Oh Yeah, Very Popular

[poena.dare]

Woodrow Wilson and Balance of Power: The 1990 Edition, two things that got me through college and prepared me to be the l33t Doom player I am!

Not a bad idea

[Galvatron]

As you probably know, game theory has long been a mainstay of econmics, the most simple and well known example being the Prisoner's Dilemma. Computers allow us to create and understand far more complicated games, so well crafted games can certainly further our understanding of how people act in certain situations, under certain constraints.

This actually reminds me of an article I read a while back, before SimCity 2000, about Maxis doing contract work for some oil company to create SimOilRefinery, to help the company plan out their refineries.

Ender's Game

[bilbobuggins]

Anyone ever read Ender's Game by Orson Scott Card?

The MOVES Institute War Game Lab

[Professor]

The MOVES Institute, part of the Naval Postgraduate School, has a complete videogame R&D laboratory. Check out our web site at http://movesinstitute.org Look under Projects on that page and see what we are doing. See you all at E3!

Phft.

[Anonymous Coward]

Yeah, I tried that, didn't work. Leisure Suit Larry sure didn't help my pickin up of the ladies, but on the upside, this slick suit does repel most stains.

SimHealth, real research, goofy name...

[tcyun]

Markle is a philanthropic organization that did some work with Maxis quite a few years ago to develope SimHealth. The purpose of the project (as I recall) was to show policymakers the complexity of the environment in which their decisions would be executed. From their website (towards the bottom of the page):

Markle worked with Maxis and Thinking Tools in 1993 to produce SimHealth, a computer-based simulation of health care policy in the United States.

I agree that the individual games and the specific examples might seem strange... but think of how strange the concept of a flight simulator (for a real pilot, not for your PC) seemed 25 years ago. Researchers have been spending a great deal of energy attempting to simulate the interactions of a complex world, with a great deal of success. It will only be a matter of time before we have believable (and probabalistically accurate) simulations of some real life situations. (Note that believable is different than predictive, I am attempting to separate the possible outcomes in a simulation separate from what actually happens.)

It's got potential

[Erwos]

I honestly see this sort of thing having potential. Computer games, when intelligently made, can teach you about something. For instance, Railroad Tycoon II's educational value in supply and demand economics is extremely useful - it would probably be easier to tell people to play it for a week rather than waste their time in economics class. When you think of a game that teaches something, don't think of the crap edutainment games you bought for your kid when they were 5. Think of SimCity and the like, where you learn by getting better at the game. The irony is that I was thinking about doubling up my CompSci degree with government and politics - maybe that's a good idea after all? -Erwos

Damn campers

[Anonymous Coward]

Does this mean the camping bastard politicos will finally get what they deserve? I know whenever I stand in one place for too long, somebody usually puts a bullet in my head. Heh.

Already have it

[Anonymous Coward]

We already have a game that pushes NGO's agendas. It's called State of Emergency.

The military has been doing this for many years

[edhall]

A significant amount of the money that the Defense Advanced Research Projects Agency (the folks who paid for the ARPANET) spent on Artificial Intelligence in the '70's and '80's was to projects that purported to model military decision-making. Later in this period the Defense office of Net Assessment got into the act, funding some efforts to combine models of military forces with command-and-control models and rule-based decision systems. At least one of these efforts endeavored to create a system where humans and computer agents interacted in a wargame involving all these elements: the RAND Strategic Assessment System, developed in the mid 1980's. RAND was only one of many in using modeling and simulation for military policy analysis; MITRE and SAIC were among the others.

At the other end of the spectrum (and probably of more interest to gamers) was the SIMNET effort, a distributed battlefield simulation system started in the early 1990's which employed 3-D graphics, multichannel sound, and a large collection of (military) hardware models, often involving several geographically separated computational nodes and human players. This is actually of far less interest from a policy perspective (unless your idea of "policy" involves low-level military tactics) but is a lot closer to what most people think of as computer gaming.

(Yes, I worked on projects like these back oh so many years ago... I left that world about the time it was starting to focus on terrorism and "light-intensity conflict." Little did I know how prescient some of those scenarios were in light of recent events in Afghanistan.)

-Ed

Maxis actually did this in '93

[colmore]

A number of people have posted that this reminds them about Maxis' computer games (Simlife was the best, btw, they need to make a sequel)

Maxis actually *did* a Sim for the government. SimHealth was developed for the government, and later issued as a (very unsucessful) public game.

There was also a Wired article about the military using Doom and Quake for VR training a long while back.

Re:Maxis actually did this in '93

[Anonymous Coward]

Actually SimHealth was done for the Markle Foundation, not the Government. The Markle Foundation is a very cool group that deals with the media's impact on society and public policy. They funded SimHealth with the idea that games were a new form of media that could provide the public with a new way of becoming informed on an issue - in this case it was to show people the effects of various health care policies that were being developed by Clinton and congress.

The Government had nothing to do with SimHealth.

New Titles

[Perdo]

Sim Monopoly (Immediately bought and axed by Microsoft)
Sim Corporate Lapdog (Bush's favorite)
Sim Golf (never mind, they get enough of that.)
Sim Pedophile (For catholic priests)
Sim Vote Fixing (the Supremes played this one)
Sim Yes man (For aspiring corporate lapdogs)
Sim Republican (Sim soft money included)
Sim Democrat (Morals not included)
Sim IRS accountant (God mode: 100% tax rate)
Sim FCC (Sell stuff the public owns back to them)
Sim NSA (Record every conversation and admit nothing)
Sim FBI (Blunder everything except cover-ups)
Sim ATF (This is just a regular first person shooter)
Sim NASA (Go to moon for 15 billion then squander 15 billion a year not going to the moon.)
Sim DOD (Procure junk, Get hired by junk peddlers.)
Sim DOJ (Income1,000,000 !=goto jail, Try others arbitrarily.)
Sim Congress (See how little you can sell out the country for.)
Sim FTC (reinforce monopolies with weak penalties)
Sim Slashdot (Hire secretaries to mod comments you don't like)
Sim THE PEOPLE (we look like goatse after the poloticians are done)

Re:New Titles

[Magius_AR]

Sim Golf already exists :) Magius_AR

DOOM

[maxpublic]

Doom taught me everything I need to know about getting ahead in the modern work place. Accept the fact that your coworkers are zombies out to shoot you in the back the first chance they get, and that problems with your boss are best dealt with using a chainsaw, and life becomes much, much simpler....

Max

Actual Application

[MadFarmAnimalz]

There's been a lot of admittedly amusing derisory commentary in the posts here concerning this idea, but the idea itself I think has potential if approached properly.

By properly, I mean without neglect for the fun factor, and without neglect for the incidentally gained knowledge and problem-solving skills potential.

I am in management at a fairly large manufacturing firm, and one of the things we suffer from is sub-standard engineer caliber. This is a by-product of the low-grade education systems in this country (in the 3rd world, btw) as is the lack of any real quantifiable work ethic. Sorry to be blunt, but those are the facts I face daily.

I envision a game simulating the production process from raw materials to finished goods, perhaps with a tutorial intro, where the player comes up against all sorts of defects and has to manipulate the production line accorsingly. Your score is based on the traditional standards of defect rates, rework rates, etc. and the scores are stored online, publicly viewable. Given that we have 12 production lines, there could even be a multi-player option.

It's the gaming competitive drive here that will drive people to play, and if the game is properly simulated they _will_ learn.

Let's put this another way, a more significant way perhaps; if someone were to code this game I just described, we would buy it without so much as blinking. I am the decision-maker there, I know what I'm talking about.

Food for thought, I hope.

Re:Actual Application

[Paul Fernhout]

The OSCOMAK concept (Open Source Community on Manufacturing Knowledge) is intended to be useful in that direction. Essentially, the idea is to create a shared database of manufacturing recipes (which include wear and failure probabilities) and use those to make manufacturing webs. I have an example of using such recipes in a simple simulation on the web site. The project really hasn't gotten off the ground though (mostly for not having the time to pursue it much).

http://www.kurtz-fernhout.com/oscomak

right ...

[cascadingstylesheet]

I'm sure creating an imaginary world where your beliefs are actually the rules would help you make your point. But how would it help make good policy?

For example, I'm sure in SimSocialSecurity, you would have a little government account with your name on it, waiting for you. Of course in the real world, it would still be a just tax and an entitlement program, with the same name, and no actual account for you. But you'd believe the game, because it would be more vivid even than a politician's speech.

Re:right ...

[cascadingstylesheet]

Whoops, make that "just a tax" not "a just tax". See, computers are distorting politics already :)

Buckminster Fullers' Game is online

[josh drvsh]

Want to play a game with real time data to try to solve the worlds' energy & political issues?
Overview:
http://www.worldgame.org/
Game:

http://www.osearth.com/

i had a policy once

[malachai_321]

i had a policy once. it reminded me of my joystick. my big, juicy, hot, red, joystick oh well, it broke the other day :(

Sim Air Pollution

[ndlxs]

I work at a governmental air pollution agency; I have long thought that what the decision makers needs at my agency is Sim Air Pollution, along the lines of the industrial simulations that Maxis did 10 years ago. What they really want to do is to game; they want to lower the emissions, throw in a few scenarios like energy shortages, and see how that affects smog.

It tooks me quite a while to get the nerve to actually suggest this approach; eventually I found a few people who really like the approach. Of course, what we do is considerably more complicated; we have models for everything from emissions from cars, to photochemical models, etc.

Games a great model for accessibility to all users. You just can't beat their approach to user interfaces.

Increasing educational simulation dollars?

[Paul Fernhout]

My wife and I have tried to have a company to make educational simulations http://www.kurtz-fernhout.com such as our free (GPL) garden simulator http://www.gardenwithinsight.com/. After close to ten years of trying, we have found it so far to be unprofitable, for many reasons. Some are personal issues like we're both designers and coders more than marketers and networkers, and we don't have PhDs (in part from wanting to do more simulations vs. doing math proofs or experiments). In general, the educational software market collapsed around 1996, which also included a big consolidation of major players.

The general issues applicable to any such effort to make educational simulations include:
* money spent on education is mainly spent on teachers and buildings,
* successful companies in education spend over 90% of funds on advertising, packaging and sales, not content development,
* parents spending money on software tend to buy that which immediately promises increased test scores not increased insight,
* a lot of flashy science suddenly appears very incomplete and sketchy when you try to actually build a simulation based on it (one of the big values of simulation IMHO),
* the real world is very complex, and simple simulations may teach the wrong things,
* the public code available to use in simulations (from government labs) is often poorly written and takes a major investment to make useable even when the science may be good,
* it is difficult to meet a good educational design goal of having open ended models a community can comment on and improve without essentially becoming open source or free software, and yet funding agencies generally expect any grant proposal to include a plan to produce a revenue stream and so to be completely proprietary and thus a grant proposal that says code will be written and given away doesn't do as well as one that says code will be written and kept proprietary (silly, but mainly true, got the letter from NSF to prove it),
* historically, people who did computer simulation work couldn't get PhDs (since most PhD programs seek to produce experimentalists or mathematicians, again we both have the scars to prove it) and it is difficult to get educational support or appear credible without one, and
* all the standard issues of innovative education being to an extent subversive (like getting people to think for themselves and ask questions) and that colliding with a funding system with mainly other priorities.

There are some obvious exceptions like Maxis (which got in at the right time and focuses on games and consulting), however when you consider the potential for computers and the billions spent on education, the number of comprehensive educational simulations which are used in practice is small. There are a lot of labors of love out there, but whether most people can support themselves on such is a different question. (We get by on mostly unrelated consulting.) These sorts of simulations can take a lot of time to do well (our free garden simulator took six person-years and is only a shadow of what we wanted to do). To do it right, you have to make something that is both a robust program (on a variety of platforms) and is also good educational science. There are few people who can do both, and so in general it takes a large team and much expense. The dollars just aren't there so far for lots of really good simulations, in large part reflecting an entrenched world view in academia (which staffs government funding agencies) that prizes experiment and mathematical proofs over simulation. (The military has historically been an exception to this.) All are needed and useful when done well, and they can all work together in synergistic ways. Maybe this report will help change things for the better. In general, things are improving in terms of academia accepting simulations (driven for cost reasons more than anything) and people who are in academia right now have less trouble doing simulation focused research.

As a caveat, one thing we have discovered is that generally the people who write a comprehensive educational simulation learn more about a subject then those who use the simulation. So having students construct a simulation may be a more useful educational experience than just using one.

Be careful what you learn

[mwood]

What was that story? It was set during an interstellar cold war of some sort, and described the impact of a new game that had begun popping up all over the Earth. It turns out that They have crafted a game which is teaching Our children to lose. Hmmm....

Simulations are frauds

[Untimely Ripp'd]

I'm surprised at how few of these comments address at all the central problem of simulations: We don't know the answers to important questions about the detailed operations of the systems we're simulating.

Inevitably, these simulations will incorporate the ideological biases of whoever is funding the simulations. This is already a policy-making problem: politicians receive support for their agendae from models created by ideological think-tanks.

I do find it odd that most of the few comments that grasped this used leftist ideology as their example. Most such models are employed by people who assume as given the ideas of folk like Milton Friedman. Thus, hilarious results like "raising the minimum wage will cost X hundred thousand jobs".

The study of the economic behaviour of individuals falls under the academic auspices of sociology rather than economics. Economists, in the main, have contempt for sociologists, and so their models usually base everything on a comical expectation of human behaviour that begins with this critical (and ridiculous) assumption: free people in free markets act rationally to achieve their own satisfaction.

Unfortunately, no element of this assumption bears any relation to our reality:

1. We are not free, and we never can be. Physical reality and the demands of our neighbors will always constrain us, even if we espouse a philosophy in which we owe nothing to anyone.

2. Markets are not free, and they never can be. The balance and counterbalance of individual and institutional power and interests ensure that all markets at all times are under conditions in which neither producers nor consumers resemble the "ideal" agents in the model of a free market.

3. People are not rational, and they never can be. We are complicated animals full of competing impulses and very complex reasoning processes. Even if we were given perfect information, we would be unable to make perfect decisions; but we are given imperfect information, which is then distorted by powerful forces in ways sure to be NOT in our personal interest.

I suspect it would be hard to build a simulation that starts with the above premise and somehow leads to a world in which sugar by the shipload is combined with just enough starch to make it easily chewable, then colorfully packaged (with millions of dollars spent developing the color schemes and the inevitable cartoon characters), labeled a "cereal", presented as a part of a healthy breakfast, and finally purchased by adults to feed to their children (whom they adore).

However, what if we required that Captain Crunch be labeled as "candy" instead of "cereal", and that it include health warnings along the lines of: "Feeding this product to your children will rot their teeth and make them fat, and quite probably affect their ability to learn well in a school setting. This increases the chance that when they grow up they will be sickly, foolish, and poor, and that their own children will do even worse." In the real world, anyone proposing this would be denounced as some sort of communist. Since only a kook would even consider it, you're unlikely to find any simulator that would incorporate such a policy into its model. And, even if you did, the effects would be guesswork, because nobody knows how consumers would respond to such a message.

Woodrow Wilson? The father of WWII?

[SysKoll]

Are we talking about the Woodrow Wilson here? The guy was a walking disaster. He teamed with Frenchman Georges Clemenceau, another disastrous politician (seems that the French are producing bad pols on an industrial scale). To give you a small idea of the shortsightedness and sectarism of Clemenceau, consider this: He opposed Pasteur and his discoveries for years on the sound scientific basis that Pasteur was a Catholic and Clemenceau hated all Christians. Now that's an open mind.

So Wilson and Clemenceau, in cahoot with the Brits, managed to win WWI at a staggering cost in human lives, and then proceeded to wreck the Austro-Hungrian Empire -- the only thing that was preventing utter chaos in the Balkans -- and saddle Germany with impossibly high war damage tributes. This, as we know, cleared up the path for Uncle Adolph and his NSDAP, as well as the various Major Unpleasantness that followed up to this day.

Woodrow Wilson shares History's limelight as the Co-father of WWII and the Wrecker of the Balkans. Great job.

One should be worried about the works of a fundation that thinks Wilson was a great guy. Was are they going to simulate with this game? How to piss up as many dangerous people as possible and survive the ensuing nuclear winter?

-- SysKoll

Hmmm...

[Cruciform]

Funny this should be announced right after those ex-Dynamix guys released PornStar3D.

Maybe the government wants to invest in buying them a better engine :)

LORD OF THE KNOBS2

[Anonymous Coward]

ONE RING TO WANK THEM ALL
one ring to wank them all

ONE RING TO CUM THEM
one ring to cum them

ONE RING TO BRING THEM ALL
one ring to bring them all

AND IN THE DARKNESS RAPE THEM.
and in the darkness rape them.

"Oil Imperium"

[Anonymous Coward]

No thanks, I'm fine playing Oil Imperium on my Atari ST. If they release it as a game, it will never succeed.

Games & Education

[ksquire]

There's actually a long history of using games to support learning -- going back to things like Diplomacy, or Risk as a lot of people here have mentioned. And, of course, the military use games & simulations to support learning extensively. My favorite recent example may be the BridgeBuilder game. Every medium, from the novel to television has been used to support learning, and I don't think that digital games will be any different, although I think that as a young medium, digital games are only starting to develop to the point where they can really be used to support learning in robust ways.

I think a lot of the problems people have is that too often, proponents of educational digital games aren't really clear about what it means "to learn" or "to teach." We know that cross-words and other classic game types can be used for drill and practice. And, we know that people can develop skills from hi - fidelity simulations. And, a game like Carmen Sandiego did a nice job at getting kids to do learn some basic facts -- by using a context to get kids to look up information and then reinforcing it through game structures. Supply and demand is pretty easily done through games, as are other things like population dynamics. In general, I think that letting kids experiment with systems that are defined by known rules is something we can do. Basically, you're taking a system simulation -- and then wrapping it around a narrative context to let users' develop goals, but then constraining their actions through a game mechanic, such as limited resources, health, access to space etc.

I find the Civilization comments particularly interesting, as my dissertation is looking at how playing Civ3 affects players' and students' understanding of history. I agree that there are several excellent learning opportunities in the game -- particularly around geography -- but there are several other unanswered challenges as well. A game like SimCity or Civilization has all kinds of opportunities for players to learn misconceptions about history -- like that the Pyramids allow free government changes, or whatever the rule is. Having used SimCity in both schools and after-school settings, I've found that the whole thing is pretty complex. Most students realize pretty quickly that cranking up the taxes for a few months isn't really realistic -- or they start asking why that doesn't happen more often, creating a teachable moment for a teacher. Any time you talk about a game "teaching" something, you're opening up a can of worms because learning is a much more active process of interpreting experience and constructing understandings than it is passively receiving the values or biases of a game.

In fact, most educational research shows that when looking at games as an instructional strategy, the reflection, debriefing, and extension activities surrounding gameplay are at least as important as the gameplay itself. You can imagine the difference between playing Civilization just for fun, versus playing the game and comparing it to historical timelines, or deconstructing its simulation biases. Generally speaking, games can get factual knowledge relatively easy -- and I think playing Civilization on a realistic map provides interesting geographical lessons. However, I agree with the skeptics here that we shouldn't assume that people are necessarily developing valuable academic skills through playing the game. But, the same can be said for lectures and problem sets, as well.

For anyone that's curious, I'll be taking up this issue on an E3 panel this year with Doug Church (Looking Glass), Will Wright, Brenda Laurel, Henry Jenkins, and a few others.

On a related note, I'm working on the games-to-teach project at MIT, where we've been developing prototypes of what next-generation educational software might look like.

To avoid flames...

[ksquire]

Before I get flamed, let me also say that I realize the irony of saying "Any time you talk about a game "teaching" something, you're opening up a can of worms because learning is a much more active process of interpreting experience and constructing understandings than it is passively receiving the values or biases of a game. " and working on a project called the games-to-teach project.