STARCRAFT 2 IS BRUTALLY HONEST: LESSONS FOR THE ATTENTION ECONOMY

The title comes from Playing to Win (at life). For context:
Starcraft 2 is brutally honest. You open it up, log on to Battle.net, click Find Match, and you’re almost instantly paired up with someone deemed to be within your skill level. Then you play, you win or you lose, and eventually it’s over. You can study the graphs and the replays, you can watch professional streams and the Day[9] Daily and read all the forum threads you want–at pretty much no point does the game not reward you for doing an infinite amount of homework–but at the end of the day, you have to click that Find Match button again, play another game, and inevitably lose if you want to get better. As professional player Aleksey “White-Ra” Krupnik puts it, “More GG, more skill.”
There are plenty of games that are competitive. You can play Call of Duty online and get your balls e-stomped by lots of folks. The difference is that Starcraft 2 don’t fuck around. There are no teammates or lucky shots. There is no respawning. There are no unlockables or pay-to-win mechanics. The only difference between you and the guy who won is that the guy who won has trained harder and worked more so he was capable of outplaying you and sending you back to the Lose screen that helpfully reminds you that you’re ranked in the bottom 20th percentile in the world. It is cruel, almost.

It’s a fandom article, and it is passionate and sincere so I mean that with all due respect. I like that the author makes good on the frequent comparisons between Starcraft and chess, which which are easy to dismiss as superficial by someone not immersed in the community. But I want to give a bit more of an academic frame for describing just how “honest” Starcraft is. You don’t need to know anything about Starcraft to read this post, but it’ll help if you have a tolerance for digital ethnography.

Starcraft is a game that requires both skill and strategy, but the game is ultimately played by managing one’s attention, which is characteristic of the RTS genre. The most skilled players are the ones who can keep track and manage a lot of different things at once. The fundamentals of the game are mostly economic, so if you want to do well you need to pay attention to managing your resources and production; TLO calls it “a game of drones”. But skilled players who can better multitask and micro might defeat players with good economic macro but who can’t divide their attention in as many ways, and balancing macro and micro issues is ultimately a question of how to manage one’s attention. I’ve been blogging on the #attentioneconomy for a few months now, and the framework of Starcraft has helped my thinking about how attention management works. In this post I want to mention two specific ways in which Starcraft is “brutally honest”, and how those features reflect the economy of attention outside of the gaming context.

For what it’s worth, I’ve spent about as long playing SC2 as patrick, the author of the insertcredit post, and I played SC:BW back in the late 90s when I was in high school. I obviously play zerg, and I’m currently ranked plat on the ladder (SoMad#988). I was ranked diamond a few seasons ago but I’ve been afk lately. I have over 3000 career wins (and roughly as many losses), so I’ve played the game a bunch. I’m not good at the game by an means, but I’ll be speaking from experience.

1) There are no laws, there is only code

There are no rules in Starcraft. Unless you hack the program with a 3rd party mod, there’s literally nothing you can do in-game that is “unfair” or “against the rules”. You can’t break the law because there are no laws to break; anything you do within the game is considered fair. You can cheese, but everyone understands that cheesing is part of the game and dealing with it correctly is part developing your skills. There are referees in tournament play and sometimes refs issue decisive rulings, but those rulings almost always have to do with instances where the game has broken down through no fault of the players, either by lag or a disconnect or some other technical problem. The only way to cheat at Starcraft is to hack the technical infrastructure; otherwise, it’s all fair game. So there are security issues to manage hacking, but the issue of ensuring a fair match between the players themselves has almost no refereeing overhead. Like chess, but unlike just about every other sport, the game of Starcraft allows opponents to engage in full out, no-holds-barred competition. And in some deeply meaningful way, for patrick and thousands of other gamers, this competition is fair. Cruel, perhaps, but ultimately just.

This isn’t just a quirky feature of Starcraft, this is a lesson about competitive environments generally, and what fairness looks like for those concerned with the economy of attention. Neutral competitive environments are ones where:

it is very hard to break the rules
it is unambiguously clear when they are broken, and
there is very little interest from any of the participants in breaking them.
In the Starcraft community there is virtually no concern about any cheating or corruption at the tournament level, even though there may be hundreds of thousands of dollars on the line. This isn’t because of strict refereeing, and it certainly isn’t because the players are saints that can be trusted unconditionally with the responsibility. It’s because there aren’t any rules that anyone has any real reason to break in the first place. You don’t need an authority to police the game itself; the game ensures it’s own fair play.

We usually think about the law as an impediment to letting people do what they really want to do. We have a speed limit laws, presumably to keep people from driving as fast as they might otherwise want. When you build laws with the intention of restricting people’s behavior, it creates a situation where the people really want to break the law, and so we have to spend a lot of time and energy enforcing the law and ensuring they don’t. From the view of the attention economy, this set up is completely ridiculous. A law is an energy sink; we don’t want laws, they only get in the way.

Instead, there is only code. A code is different from a law. A code a common framework that unifies the system and provide a neutral playing field to ensure that competition between the participants is fair. Software is composed of code, not laws. Blizzard attempts to code for a balanced game, and then it is entirely in the player’s hands to use that playing field to their advantage. If something in the game is making the competition unfair, it would be a waste of time for tournaments to create a law requesting the players not to do it. Instead, Blizzard changes the code so that the unfair advantage is no longer an option to any players, thus (hopefully) restoring balance. Blizzard’s management of the code isn’t an impediment to competition, it is what makes fair competition possible at all.

Similarly, our other social organizations should be very reluctant to use “laws” as a means of ensuring fair competition. Not that we should abandon competition and force ourselves into indiscernability. Instead, we want to ensure that the code functions as a framework for fair competition, while giving competitors free reign to explore the codespace as they see fit. This is precisely how governance as coding works in the digital age, and Starcraft is a model of its success.

To return to the analogy with driving, we know that speed limit laws have virtually no effect on the speed of traffic. So why do we have the laws? Autonomous cars will eventually be able to adjust their speeds and respond to their environment with more accuracy and safety than human drivers. At that point the code itself does all the work of managing traffic and the law becomes superfluous. In fact, the data seems to indicate that the laws are already superfluous; automating the driving process would only reinforce their obsolescence. Traffic patterns are self-organized, they aren’t organized around the laws. That doesn’t mean traffic is chaotic or arbitrary, it means that well-trained drivers will drive well regardless of the “laws” that are meant to “govern” their behavior. Instead, drivers self-organize into traffic patterns that correspond to their training in the skill of driving, and to the code that organizes the infrastructure in which they act. The code in this case is the literal material infrastructure of the road: the street signs and dashes of paint that signal the stigmergic patterns of driving behavior in humans.

Of course, humans are incredibly intelligent and honed by evolution for large scale cooperation. If you train them well enough, they’ll organize without much code at all.

This leads to the second big lesson Starcraft has for the attention economy:

2) System management is managing a diverse collection of autonomous, intelligent agents

In Starcraft, all the units are autonomous and intelligent. You can direct them where to go, and to a certain extent you can tell them what to do when they get there. But otherwise, your units react on their own, according to their own internal logic. The fighting units will attack at anything in their range automatically, and they will run off attacking or avoiding things on their own if you don’t tell them to stay still. The harvesting units will continue to mine resources or repair units automatically without having to give them any explicit orders to do so. The intelligence of the individual units frees the player (a systems manager) to attend to other things within the game. While the player trusts the units to act in mostly predictable ways, their trust isn’t always well-placed: for instance, see Naniwa’s stalkers on Cloud Kingdom against MVP on GSL code S. If you weren’t watching it, just trust me ¯_(ツ)_/¯

Part of the player’s skill at the game is knowing how the units behave in certain contexts and being able to exploit the tiniest advantage those behaviors might give. Knowing how close you can stand to siege tanks, knowing how the units will rally or what their vision will let you see, these things aren’t just about knowing rules and giving units orders, but it is to some extent about working with the units as a collaborative effort. You aren’t their leader or their owner of your units, you are the system manager, and you are all in it together.

The intelligence of the units isn’t a trivial part of the game, and the community is explicitly aware of the consequences this has for gameplay. One of the major changes between SC:BW and SC2 was in making the individual units more intelligent; this change has been particularly in the case of the basic harvesters. In SC1 the harvesters were stupid and would mine minerals very inefficiently; good players spent a lot of time attending to their mineral lines to make sure they were mining as efficiently as possible. In SC2, workers manage their resource collection and cooperation in very efficient ways, allowing players to manage more complicated maneuvers and unit compositions. Having good macro in SC2 still requires attention, but that attention is usually not spent on worker management in the same way because they are largely smart enough to handle it on their own.

If you want to fuss about whether the drones in SC2 are “really” intelligent, I have things to say about that but I won’t say them here. I’ll just audibly sigh instead. That’s not to say the question of artificial intelligence isn’t relevant to SC2; in fact, there’s a flourishing community of AI researchers who work on Starcraft from a variety of angles. But for the moment, just take the claim at face value: units in SC2 are smart, and anticipating how they think and move is a large part of what a player’s skill amounts to. The units are smart in more or less the same way that your fingers are smart. A lot of the “knowledge” about how to articulate them is stored directly in the ligaments and bone structures and muscle fibers in your fingers themselves. The brain doesn’t issue orders directly so much as manage the high-level control over the operation of the whole hand system, and the brain trusts the hand to take care of most of the execution itself. This is how dynamical systems tend to work. Very little direct control from management, let the units do the jobs themselves. Again, Starcraft is a model of such systems, and is very instructive for thinking about digital organizations generally.

The implications of managing a collection of autonomous agents for the attention economy are mostly obvious, but I want to draw them out a bit for the issue of organizational identity. Consider just the first 30 second or so of the following typical introduction from a random Starcraft tournament.

This introduction is standard for the game. Notice the brief look into the player’s booth as they prepare; but once the game starts, the introduction of the players follows their position on the map and their manifestation as units in the game (“in this corner… Nestea!”). While the commentators might talk about “Nestea’s drones” in the sense of ownership, they are just as likely to talk about Nestea himself as being in different positions or taking different actions on the map.

In other words, the units that Nestea manages in the game are his embodiment on the map. Digital ethnographers talk about “avatars” as one’s virtual body, generating the forms of embodied presence familiar from virtual environments like SecondLife or Minecraft. Starcraft radicalizes the idea: one’s “avatar” in this game is composed of discrete functional units all behaving autonomously, and this generates a “body” more complex than other avatars by an order of magnitude. Dennett says I am wherever I have control, and Starcraft gives a model for how this works in digital environments when one’s body is composed, as Dennett says, of “lots of tiny robots”. Nestea the player exists in that virtual environment where his units are, and that existence does not always function as a coherent, unified whole. Instead, his identity is distributed across many activities and locations all around the map, exerting high level organizational control of a diverse set of autonomous agents. This flies in the face of Enlightenment individualist assumptions about the unity of identity, but is paradigmatic of organizational intelligence in the digital age.

One very immediate application of these distributed identities is the development of distinct and identifiable “styles” of game play, some of which might be so obvious that trained observers can identify the player by just looking at their in-game behavior. Many Starcraft enthusiasts would have no problem distinguishing between, say, a game by MC from a game by Parting, even though both play protoss using roughly the same units. Still, these pros have radically different styles, which manifest themselves in different methods of gameplay, and if you know what you are looking for, these differences are easy to spot. Let me say that again if you missed it. Players, working entirely by managing a system of autonomous, discrete units, can nevertheless let their own personalities shine through making them uniquely identifiable to others in the community. On the Enlightenment model this might be surprising, but it is par for the course in the digital age. Developing these styles is part of the active and vibrant Starcraft metagame, and keeping up with the metagame is an important part of practicing and developing one’s skills. Putting the point simply, the code provides a neutral framework for the development of culture, and Starcraft is an excellent model for watching these spontaneous digital cultures thrive.

_______

I can say a lot more to elaborate on the more philosophical issues alluded to in this post, but hopefully the two big issues discussed here give some sense of how Starcraft might help us think the attention economy in the digital age. There’s much more to say about the role of SC2 in the economies of attention; I personally find the ecosystems of streamers to be absolutely fascinating. I’m particularly interested in the way the Starcraft community is managing issues of both crowdsourced and corporate funding, and the issues of gender and identity surrounding the development of the e-sports scene. But in the spirit of patrick’s original inspiration, I’ve stuck with in game discussions in this post, and hopefully I’ve shown there is plenty of interest to discuss there as well.