Jun 28 2016



Well we’ve seen some pretty amazing democratic results in the past few days. The consensus seems to be that democracy is a great thing so I’d like to take a hard look at that in the face of a result that doesn’t seem so great.

First let’s understand that the purpose of democracy is not to ensure the best possible results. No one who implemented the process thought it was going to solve all problems and generate optimal choices for their organization. What they needed was a way to secure legitimacy — to be able to act with the support and investment of the affected populations. Corporations, for example, are very rarely democratic. Sure the board may vote but they aren’t representing the company’s population — the labour pool — but rather shareholders or other external interests. Corporations don’t need to be democratic because they have a totally different mechanism for generating legitimacy: the labour contract. They pay you, you do what they say, or you find another employer. It’s actually a pretty good mechanism in its intended context and is arguably better at decision making than a democracy. Also almost equally not in the case of publicly held companies since it admits to only one genuine objective: constant growth. All other “mission statements” are myths the company tells itself to seem like part of a more complex, more nuanced environment. But nonetheless, the system has almost perfect legitimacy — you can’t argue that the company has no right to tell you to do your job (as agreed in that contract).

So where does legitimacy come from when we’re talking about governments?

There was a time when that answer was pretty simple: divine right. Given an all-powerful deity or deities, if the population believed that a particular individual was divinely selected to fulfill the role of leader then, well, that was that. Perfect legitimacy. This became muddy fast. Questions about the nature of the deity, the existence of the divine, and of course whether or not the leader was actually divinely decreed all undermine the method. It’s fine while the population is too poor to care (cf. Maslow) or insufficiently educated to ask the right questions, but as soon as any significant element of the population (and your worst fear here is probably the priesthood itself) is sophisticated enough to question the roots of that premise — divine right — legitimacy is hard to hold on to.

And so enter democracy in all its forms. Whether or not the method creates good decisions, it unquestionably has legitimacy. Witness the Brexit vote: everyone is wondering whether that result was a good idea but no one is wondering whether it was legitimately arrived at. Legitimacy is unexamined. Democracy is so powerful a legitimizer that we don’t even think to ask whether the vote of all citizens should be a legitimate way to make decisions. And that’s risky. That’s a question that should be front and centre: should a government care what the majority of its citizens want today? Are they equipped to answer the question at hand? We know they will reliably tell you what they want (well, the buyer’s remorse evident after the Brexit vote might suggest even that is not necessarily true) but is the population at large equipped to tell you what they need?

They are, I think, but the question has to be framed very carefully. “Should we leave the European Union?”, for example, is a hopelessly complex question buried in some very simple text. It doesn’t imply anything about what it would actually mean to leave the EU. It doesn’t give us any information about what will replace the things that are lost. It doesn’t really give use any information at all, since it turns out it’s possible to not be in the EU and yet have a very similar treaty in place — being out of the EU can mean having all the privileges and obligations of a member except a vote. It seems unlikely that Leave voters were hoping for that result. This leaves the population to do its own research, which it won’t.

Sure we can say they should. But they won’t. We can’t wish the population would behave a certain way but rather we have to design systems that accommodate how they will behave. This is User Interface Design 101 — when users use your thing wrong, the correct response is never “they should read the manual”. They won’t. You have to cope with that, not wish you had a different (and probably fictional) population.

And so the many forms of representative democracy exist. They solve (partially, imperfectly) that problem by putting supposed experts in the role of handling the details. This works pretty well. Sure we have trends where our representatives deliberately ignore their obligation and vote with their hearts instead of doing the research, but at least when that happens we have individuals we can take to task for it. That’s their job and when they fail we can do something about it. If we want to. Failing to research a topic before voting is, for them, a violation of the public trust. I have strong feelings about how that should be handled. It should certainly be handled with vastly more force than we apply today since, let me repeat, that is their job. When they fail at this minimal description of their obligation, their vote is┬áthe same as a public vote except that they get to vote for thousands or millions of people on their behalf. Without consulting them.

So we have through trial and error found a way to make legitimate decisions on behalf of a population. Not great decisions, but at least legitimate ones. And we have a further refinement that allows us to ensure that the direct voters have an obligation to consider the question seriously and in all its complexity. They research. They debate. They consider. Things we have no time to do in any detail. And for most of us, no interest.

So the Brexit vote, by going to a deeply flawed method of decision making, couching a horrendously complicated question with a million sub-questions as a single simple question that the public is free to read any number of ways, was a totally legitimate decision based on zero information. That is about the most irresponsible thing I can imagine a government doing in good faith: guarantee legitimacy to an uninformed decision.

Sep 2 2011

Nasty, brutish, and not very short game design tools

Bear with me here because I have a head cold and it may affect my brain, so the narrative in this post may wander a little. The idea, however, occurred to me while healthy so let’s suppose it’s sound enough.

Okay, so I’m reading Hobbes’ huge work of political philosophy, Leviathan. I’ve read it before, occasionally under duress. Hobbes is a very conservative philosopher — he’d be at home in any decent quality assurance team. He has no problem at al saying you should start your argument from first principles and then actually doing it. So Leviathan, which is a many hundreds of thousands of word argument against the French Revolution and maybe specifically against some thing pamphlet by Thomas Paine (who I vastly prefer reading most days), starts from first principles.

First he rails against terms that have no meaning, picking a few pointed examples so as to also poke his contemporaries (like Paine, certainly). He says, for example, if you stick two words together that are meaningless in reference to each other, such as “Free Will”, then they are together meaningless. They are an absurdity. They are mere noises. This is good stuff — you don’t have to necessarily agree with his demonstrations to agree that using meaningless terms is dodgy philosophical work.

Then he starts defining the terms he’s going to use in his book. He starts as low as he can go and then builds up a vocabulary for talking about why people do things. What motivates them, what is good in them, what is bad in them, and so forth. In fact he does this for six chapters.

So this is six chapters packed with carefully defined words about human capabilities and motivations and interactions. And better, they are almost all defined with their opposites. Good and Evill are defined. Attraction and Aversion. Terpitude and Pulchritude. And so on. This is an enormous effort and fascinating to read.

Insert the sound of an old vinyl record scratching as I bang the tone arm off it.

Games are machines.

That’s why we use words like “mechanics” (well, you all do — I prefer “mechanism” because “mechanic” sounds like a guy who fixes my car (and I don’t own a car)). When we make games, we make all kinds of little pieces (or steal them) and fit them together so when you pull one, this other one spins, and tightens another, that causes a player to twitch. Some of the machines are fitted loosely and need constant attention from the operators. Some are fitted very tightly indeed, and admit no fingers except at the designed interfaces. These create different play experiences.

At the interfaces are a lot of guages and dials and lights. We have skills, attributes, hit points, aspects, stats (whatever that means), armour class, base attack bonus, beliefs, and on and on. This is the dashboard — the character sheet — and all those things are indicators and controls. If one changes, it better mean something to the operator (otherwise it’s clutter on an essential instrument!) If the player tweaks the value of one, something better change in the outputs of the machine.

Now when crafting this dashboard a designer often runs into one prosaic, painful, trivial-seeming, and yet essential problem.

You may now imagine the music has resumed.

You need to name all these things because they will mostly need to be words on a character sheet. You will need a lot of words that are about peoples’ motivations and abilities. It would be ideal if you had their opposites handy as well, because it very often happens that you want to label extreme ends of a range or that you want people to choose between two conflicting options. It would be great if the definition was handy and clear as well, because it’s always a thrill when you not only find the right word, but you also find the exact perfect right word.

If only we game designers had a good list of words like this.