simulations miscellany, 3 July 2012

Yes, it’s time for another edition of PAXsims’s regular simulations miscellany, offering a selection of recent simulation-related items from around the net. If you have suggestions for future blogposts, please send them on!

* * *

At the gaming website Kotaku, Michael Peck discusses “Why It’s So Hard to Make a Game Out of the 21st Century.” His answer? Because we’re just guessing:

These are not trick questions. They are merely unanswerable, or at least the answers don’t appear until after the fact. Lots of people in the Pentagon, think-tanks and universities get oodles of taxpayer money to devise forecasts, mathematical models and even make games to predict what will happen. Their answers may be better informed than yours and mine—perhaps they have access to classified intelligence data—but this doesn’t necessarily mean that their answers are more accurate than yours or mine. The pros often blow it in spectacular fashion (practically none of the experts on the Soviet Union predicted its abrupt collapse). This is not to say that Civ is a better predictor of the future than a mammoth $300 million Pentagon simulation likeWarsim. What it means is that predicting the future, whether you’re a game designer or a talking head on TV, is to guess. The problem is that often these guesses are cloaked as expert opinion, or game marketing copy that boasts of impressive research. They’re still guesses.

I partly agree. Political experts have a pretty poor record at prediction, although there is evidence that intelligence analysts tend to be right somewhat more often—not so much because of access to classified material (which helps much less than you might think when it comes to long-range political-military prediction), but rather because of the analytical methodologies that they tend to use. It is particularly hard to predict something that is twenty years out. When you do, moreover, this is best framed as a thought exercise intended to promote discussion and original thinking (as with the US National Intelligence Council’s Global Trends 2025 and Global Governance 2025 projects), rather than a hard-and-fast prediction of the future to be.

That being said, these are much more than guesses. Moreover—and I think, more importantly—game methodologies allow you to actually explore the effects of a broad range of future outcomes by changing only a few variables at a time. Think that cyberwarfare will become increasingly common, as capabilities outgrow and proliferate faster than countermeasures? Then try a simulation with that assumption. Thing that cyberwarfare will favour technologically advanced states? Then try that. Want to know the impact of global warming or energy shortages? Try keeping most other variables constant, but alter those particular sets of variables. Done well (with a “sage” rather than “seer” approach to the process), the exercise can be quite useful.

* * *

A Facebook discussion of Michel’s piece launched by Adam Elkus led me to come across an interesting discussion by Kelsey Atherton on how games model covert action.

When covert action shows up in strategy games, on the other hand, controlling agents is all about picking objectives and using the people available to best execute that.  For example, in the Total War series, agents are either spies or assassins, whose presence only becomes known to opponents if they have failed their mission, or if they have been observed by enemy spies. If they fail a mission against a high-enough value target, they are likely to be executed, but all the calculations behind that are beyond player control. Agents are useful for much the same way we imagine them now: information on enemy developments & deployments to better plan ones own military moves, targeted killings on individuals otherwise beyond the reach or purview of conventional forces.  That said, there are limitations on how useful this is as a form of modeling. Agents here operate within strictly coded boundaries, and so they cannot, say, spread false information amongst the enemy leadership (so no Dudley Bradstreet’s here), or engage in any other behavior that breaks the established rules.

It is well worth a read for those interested both the in the political-military content of contemporary digital games, and for those interested in simulation modelling more generally. Adam’s comments on the same issue are also worth reading.

* * *

Among other things, the Training & Simulation Journal has recent short reports on how hard it is to replicate human behaviour and lessons from the recent Unified Quest wargame.

* * *

Training and simulation needs often get seen through the prism of the US military (by far the largest single consumer in the world of digital simulation, training, and education material) or NATO allies. In a presentation to the recent Simulation & Training Africa 2012 conference, Major General Luvuyo Nobanda (Chief Director: Force Preparation, South African Army) offers the perspective of a developing country of more limited means.

* * *

Meanwhile back in the US,  the Department of Navy has added the computer game Starcraft to its military officer training curriculum in an effort to provide realistic leadership simulation. Enemies will no doubt soon discover the fearsome power of the Zerg rush…

(OK, no it hasn’t really, but it’s another funny piece of satirical military reporting by the folks at Duffel Blog.)

* * *

My own annual Brynania classroom civil war simulation was recently profiled in the Spring-Summer 2012 issue of the McGill News alumni magazine. There are also a couple of earlier videos on Brynania that give a real flavour of what 120 bright students can get up to that week, one by the folks at TV McGill and the other by McGill’s Headway TV series. Oh, and there’s also this year’s UNHCR team asking a couple of improv coffee shop musicians to sing about their fictional jobs…