Jon Adams: Why We Play Role-Playing Games

1 Leave a comment on paragraph 1 0 Why do we play role-playing games? A number of answers to this question have been proposed. The most obvious answer is that we play role-playing games because they are fun. Creating a game that is fun is the first principle of game development (cf. Rollings and Morris 38). In a game like Diablo, this fun is the constant and immediate positive feedback of defeating mobs, exploring the map, and finding treasure. A somewhat less obvious answer is that games provide a challenge. This challenge offers the cognitive pleasure of learning a complex system with problem-solving structures (cf. Gee 3). In a game like Ark, for example, a player uses problem solving to learn how to survive in a world inhabited by dangerous creatures. There are perhaps more answers to why we play role‑playing games, but I want to propose one that is not at first obvious: we play role-playing games to prepare for the future. Specifically, we play games to prepare for a future dominated by algorithms. Some observers claim that this future has already arrived, that we already live “in the age of algorithms” (Domingos 1). Because algorithms are everywhere: they are in our computers, in our cell phones, in our cars, and even in our thermostats. Algorithms are changing our lives, often without us being fully aware of it. The problem is not to stop the proliferation of algorithms. That’s not possible. Rather, the problem is one of awareness, of acquiring a certain level of algorithmic literacy (cf. Rainie and Anderson). This is a problem that role-playing games are in a position to deal with because what a player in such a game does is confront algorithms. It is what role playing games are about.

2 Leave a comment on paragraph 2 0 Machines

3 Leave a comment on paragraph 3 0 Algorithms are used everywhere because they do things. They do things that help us in various ways, but they also do some things that are troubling. For example, if you have a passport photo, your “face has been turned into an algorithm” (Clare Garvie qtd. in Scola). This use of algorithms is troubling because we didn’t anticipated it and we can’t control it. I want to briefly point out three general ways in which algorithms are having an impact on our lives.

4 Leave a comment on paragraph 4 0 First, algorithms are beginning to replace us as workers. Robot-controlled algorithms are already replacing manual workers in many tasks. PricewaterhouseCoopers estimates that machines are likely to take over more than one-third of the jobs in America and Germany by the year 2030 (cf. Dennin). Such machines are not only poised to take over routine work, such as customer service and data processing, they are also threatening to take over such professions as medicine and law (cf. Susskind and Susskind). For example, Ross is a machine powered by IBM’s Watson technology that serves as legal researcher. It sifts through thousands of legal documents and delivers a ranked list of the most relevant ones. When it is asked a legal question, Ross replies with a few paragraphs summarizing the answer and a two-page explanatory memo. The results are indistinguishable from the work of a human lawyer, except that it is much faster (cf. Lohr, “A.I.”).

5 Leave a comment on paragraph 5 0 In another example, machines are widespread in banks and wealth management companies. BlackRock uses Aladdin, a machine for investment management; Goldman Sachs uses Kensho, a machine for financial research; and UBS and Deutsche Bank use Sqreem, a machine for financial crime protection. These machines are different from earlier rule-based models; they are the next generation that use deep learning techniques (cf. Crosman).

6 Leave a comment on paragraph 6 0 Second, algorithms are beginning to collect more and more data about us. We all know that algorithms track us on the Internet. Most of us are probably familiar with cookies and know that we can delete them. But there are other tracking algorithms that don’t use cookies. An algorithmic technique known as fingerprinting can identify unique features of our browser, give it an identifying number, and know when we return to the site (cf. Hill).

7 Leave a comment on paragraph 7 0 We can also be tracked even if our data is anonymous. There are laws against selling personally identifiable data in the US, such as a person’s name, Social Security number, and medical condition. But if the data is scrubbed of personal information, then it is considered “anonymized,” and it can be legally sold and publicly transferred. But there is so much publicly available information about us online that, combined with powerful algorithms, it is possible to re-identify a person’s “anonymized” data. This means scrubbed data can now be traced back to the individual user to whom it relates (cf. Lubarsky).

8 Leave a comment on paragraph 8 0 Third, algorithms are beginning to make decisions not only for us but also about us (cf. Lohr, Dataism). For example, when we apply for a job now, algorithms are used to review our resumés. Human resource managers rely on these algorithms so much that they let the algorithms screen out over 70% of the job applicants without a human ever reviewing them (cf. Mann and O’Neill).

9 Leave a comment on paragraph 9 0 In another example, a man in Wisconsin was sentenced to six years in prison in part based on the analysis of a machine called Compas. Using data about the defendant, Compas reported that he had a high risk of violence and a high risk of recidivism. The defendant challenged the judge’s sentencing because neither he nor the judge was able to examine the algorithm that produced the Compas report. The defendant’s challenge was denied because the algorithm is a trade secret (cf. Liptak). In 2015, a California Appeals Court upheld this decision that a trade secret is privileged evidence to prevent disclosure of an algorithm’s source code, even from the defense. This decision, “People v. Chubbs,” is now being cited across the country to deny defendants access to trade secret evidence (cf. Wexler).

10 Leave a comment on paragraph 10 0 I don’t think I’m overdramatizing the growing impact of algorithms. Others have suggested that algorithms are going to completely replace us (cf. Harari). At the moment, this seems a little extreme, but it is clear that algorithms are changing our lives and that we should at least be aware of their nature.

11 Leave a comment on paragraph 11 0 Games

12 Leave a comment on paragraph 12 0 With most algorithmic systems, the user requests information. There is interactivity between the user and such systems, but it is rather basic. The user sits outside the system and queries it, whether the system is the Internet or a doctor’s diagnostic machine. A role-playing game is also an algorithmic system, but the player doesn’t sit outside the system. The player enters the system, the game world, and explicitly confronts the algorithmic system. Lev Manovich pointed this out almost two decades ago: “as the player proceeds through the game, she gradually discovers the rules that operate in the universe constructed by this game. She learns its hidden logic—in short, its algorithms” (222). When we enter a game world, we learn to play the game, we learn about its algorithmic system, what the algorithms do and what they don’t do. We learn this by playing the game, by doing things in the game world, because algorithms themselves do things. They are procedural; they are processes for making decisions. When we play a role-playing game we do the same thing: we make decisions and we do things in the game world. But in order for us to do this successfully, we have to understand the game world’s algorithmic system. For example, the basics of any role-playing game consists of how to equip and move the player character, how to negotiate the map, and how to survive, either through combat or stealth. These are some of the simple mechanics of a game, and these mechanics are a system of algorithms.

13 Leave a comment on paragraph 13 0 Even a casual video game player learns a couple of general features of algorithms. For example, they learn that algorithms aren’t perfect, and that they sometimes have errors, exploits, and unintended consequences. Our player characters, for example, used to fall through the map rather frequently, and getting stuck on the map still happens now and then. An unintended consequence in Planetside still sticks in my mind. Planetside was a massively multiplayer online game that had friendly fire. You could kill your own team members and they could kill you. As a result, there was a player who liked to drive a heavy truck and run over other players on his own team. There was some modeling glitch that made it nearly impossible to dodge the truck, so that even a near miss was fatal. Many players complained bitterly in-game and on the forums about this maniac truck driver, but since the mechanics of the game allowed it, he continued to run over his teammates.

14 Leave a comment on paragraph 14 0 Another general feature of algorithms is that they form a closed systems. The player can do anything the game or algorithmic system allows, but what the game allows is limited. In Fallout 4, for example, the player can craft a house, but he can’t craft a bicycle, a car, or a boat. At first an open game world like Fallout 4 may seem vast, but that’s mainly because it is new and unexplored. A player can easily walk from one end of the map to the other, but he can’t walk beyond the map. Algorithmic systems are limited not by what they do but what they don’t and can’t do.

15 Leave a comment on paragraph 15 0 Perhaps the most important general feature of algorithms is that they create systems that are a world apart, a world that is not only independent of our world but completely indifferent to it. Algorithms demand that they be dealt with on their terms, not ours. They don’t adjust to us, so we have to adjust to them. In the classic role-playing game Morrowind many of the NPCs change their location and then stand in such a way that they block the movement of the player character. In the entrance hall of the mages’ guild in Vivec there are two particularly irritating mages that tend to position themselves so that they block the doorway leading to the interior of the guild. The point is that the player has to deal with the problem and  find a workaround. The mages, or the algorithm that controls their pathing, are completely indifferent to the player’s concern. When the player character bumps into one of the mages, it doesn’t move; instead he says, “What do you want, Outlander?”

16 Leave a comment on paragraph 16 0 Confrontation

17 Leave a comment on paragraph 17 0 Many features of algorithms appear in various types of video games, not just role-playing games. But in terms of confronting algorithms, role-playing games have an advantage because the player doesn’t play against other players but against the environment, that is, against the algorithmic system of the game world. The importance of this aspect of role-playing games becomes clear when we remember that, unlike most video games, a player at the beginning of a role-playing game doesn’t know all the rules that govern the game world. The player has to figure out what to do, how to play the game. The player has to “probe the game’s logic” (Johnson 42).

18 Leave a comment on paragraph 18 0 One of the most important features of algorithms that a player learns in confronting the game system is that the player controls the situation, because nothing happens until the player enters the sphere of a mob’s algorithmic system. For example, when the player meets the first boss in Dark Souls 3, the boss is on one knee with a sword in its side. The confrontation doesn’t begin until the player pulls the sword out. This first boss in Dark Souls 3 is considered a tutorial, a boss fight that teaches a new player the basics of boss fights in the game. What the player also observes is that the boss, as an algorithmic entity, has specific properties. For example, when a player character dies, respawns, and then returns for another attempt to defeat the boss, the boss has returned to a position in the center of the arena. This is a general feature of algorithms: they have an initial or default setting. And after a number of attempts to defeat the boss, the player will begin to notice a pattern, depending on whether he keeps his character close to the boss or away from it. In other words, the player learns that the boss, as an algorithmic entity, is predictable. And when the player finally defeats the boss, it despawns and the gate opens to the next zone. The boss, as an algorithmic entity, has reached its final state.

19 Leave a comment on paragraph 19 0 In his various attempts to defeat the first boss in Dark Souls 3, the player learns another lesson: the purpose of a game world is to retard or block a character’s progress. This suggests that the game has no intrinsic purpose independent of the player. The player’s motivation to play the game, his motivation to master the game world, gives him a certain control over the game world. When the player confronts the game world, it simply reacts. It is the player who has the task to change the game state, to invoke some algorithms and eliminate others. For example, the boss remains in its initial state until the player appears, then the boss enter its combat state. The algorithm, in other words, needs the player’s input. This is what empowers the player in a game world: the player gives algorithms purpose.

20 Leave a comment on paragraph 20 0 Character

21 Leave a comment on paragraph 21 0 Another important advantage that role-playing games have over other games is that the player uses an algorithmic entity to confront the algorithmic system of the game world, for the player character is an algorithmic entity. The player uses his character in two ways: in the strategy he uses to approach the game and in the tactics he uses to implement his strategy. A major aspect of a player’s strategy consists in how he configures his character. For example, in Dark Souls 3 the player may decide to favor magic and speed over strength and armor, so he ignores attributes that support strength and armor in favor of skills that support magic and speed. During the game, each time the character gains a level, the player has to decide which skills to increase to support the way he plays his character. And each time the player defeats an enemy that drops armor or a weapon, he has decide whether he wants his character to use it. While on the surface these are strategic decisions that the player makes about how to use his character, at a deeper level these decisions are about how to configure an algorithmic entity to confront the algorithmic system.

22 Leave a comment on paragraph 22 0 Based on his strategy, the player uses his tactics to confront the game world. For example, when facing a mob like a Ringed Knight in Dark Souls, the player keeps his character at range and burns the Knight down with magic. More difficult mobs and bosses require kiting and dodging, which is where speed becomes important, but the tactic is similar. In role-playing games, tactics are a combination of what the player decides to do and the character executing that decision. If the player decides to throw a fireball or roll away from an incoming attack, it is the character that does the actually throwing and rolling. This means that the player works intimately with his character, with his algorithmic entity, in confronting the game world.

23 Leave a comment on paragraph 23 0 Using a character in a role-playing game allows a player to learn a number of specific features of algorithms, features that define what algorithms do and don’t do:

  1. 24 Leave a comment on paragraph 24 0
  2. Game worlds count everything. As a player guides his character across the map of a game world, walking, running, turning, sometime jumping, the game world appears to be an analog system. But as soon as the character confronts an enemy, in a game like World of Warcraft, numbers begin to flash on the screen because the outcome of the encounter is measured in numbers. The outcome is measured in how much damage a character can absorb and how much she can produce, and at the same time, how much damage the enemy can absorb and how much it can produce. Algorithms are very good at numbers because binary is their native language. In fact everything is measured in numbers: the map of the game world is a grid and a character’s movements are measured on that grid. The only thing that is not measured in numbers is the player’s decisions about what to do. In encountering a sabretooth tiger in Ark, for example, the player can decide to fight or, because the sabretooth runs in packs, he can decide to flee.
  3. Game worlds are not realistic. Games like Everquest and Star Wars, with their fantasy and science fiction settings, aren’t intended to be realistic, but even in a game like Ark, which tries to be as realistic as possible, there are game conventions that override any sense of realism. For example, the player character and mobs in a role-playing game have a health bar. On the one hand, this is part of measuring how much damage a character or mob can absorb, but on the other hand, the visual symbol of the health bar is an algorithmic representation of “health.” What the player learns is that algorithms may imitate biological entities but they are made up of numbers not organic cells.
  4. Game worlds need to specify everything. Azeroth, the game world in World of Warcraft, is a fantasy world. Players can ride flying horses and tame spiders to fight for them. But even a fantasy world needs some consistency so as not to appear completely irrational. Yet in many dungeons in the World of Warcraft, a group’s tank hits the boss with her sword to maintain threat, but when the boss begins to perform a power move, the tank can avoid the attack by just walking through the boss. This inconsistency in collision detection is part of the game world’s non-rationality, but it also reveals the specificity of algorithms. This specificity defines the player’s sword as having collision detection but the character herself as not having it. For algorithms to function, every feature of every object needs to be specified.
  5. Game worlds have immaterial space. Space is a major feature of role-playing games, and it is one of a main reasons we talk about game worlds and why maps are important. Many of these maps in role-playing games are quite large, which in the past led to the problem of a character having to walk or run from one place to the next, often over and over again. To avoid the tediousness of walking everywhere, most games have a fast travel feature. A character can use a ring in World of Warcraft, a carriage in Skyrim or a bonfire in Dark Souls and move from one location in the game world to another location without traversing any of the locations in between. This travel is instantaneous. Algorithms are not bound to the spatial physics of our world.
  6. Game worlds are arbitrary. For example, in most games the player character has an inventory, but the inventory is often implemented differently in different games. In World of Warcraft, the character’s inventory is limited to the number of slots she has in her bags. In Fallout, the character has a seeming unlimited number of slots, but she is limited in the amount of weight she can carry before becoming encumbered. In Elex, the character has an unlimited number of slots and no weight limit. Notably, time in game worlds is also arbitrary, and a game world may or may not have a time structure. Everquest has a rapid day and night cycle. Dark Souls has no day and night cycle, but in some zones it is dark and in others it is not. World of Warcraft has a day and night cycle that has changed over the years since 2005.

25 Leave a comment on paragraph 25 0 Players understand these algorithmic features of role playing games. We see illustrations of this on the forums when players discuss the technical issues of a game. For example, in a discussion of how to play a tank in World of Warcraft, a player introduces the basic features of the threat algorithm.

26 Leave a comment on paragraph 26 0 Threat is a means of measuring the level of animosity a mob has towards a specific player. Each mob has a threat table, and every person who performs hostile actions towards that mob is put on that table.

27 Leave a comment on paragraph 27 0 There are two important actions which generate threat: dealing damage and healing. Other actions, such as casting a buff or debuff also generate threat, but in very small amounts which are not worth discussing.

28 Leave a comment on paragraph 28 0 Normally, threat is generated at a 1:1 ratio with damage done to the mob, and a 1:2 ratio with healing done. However, in order to facilitate tanking, tanks generate far more threat from their damage. (“Tanking Guide”)

29 Leave a comment on paragraph 29 0 As we can see from this description, threat is an algorithm that determines which character in a group a mob attacks. It is an algorithm that uses a measurement that is recorded in a table. This threat table is a data structure in the threat algorithm. If you are in a group in World of Warcraft you need to know this. And if you are the tank, then you need to know how to stay at the top of a mob’s threat table, that is, you need to understand how the threat algorithm works.

30 Leave a comment on paragraph 30 0 Conclusion

31 Leave a comment on paragraph 31 0 We are facing what Shawn DuBravac calls a “digital destiny”:

32 Leave a comment on paragraph 32 0 The point is that in 2025 electronic devices connected to the Internet and equipped with powerful sensors will be ubiquitous, surrounding us at all times, acquiring and analyzing data, not only to give us what we need the moment we need it, but to acquire more information on our true identity. (118)

33 Leave a comment on paragraph 33 0 The point that I have been trying to make is that a user who has played role-playing games will not be a passive victim of these ubiquitous algorithms. First, he will know that he is confronting an algorithmic system, and he will look for ways to use the system to his advantage, just as he did in role-playing games, for example, when he learned to kite mobs and use line-of-sight to avoid taking damage. We already have a name for this: it’s called gaming the system. Next, I think a user who has played role-playing games will look with some suspicion at the idea that algorithms gather data about his “true identity,” because such a user will be familiar with an identity formed by algorithms. Such an identity will look very much like a player character in a role playing game. Once a user see this, he won’t leave this algorithmic identity to the complete control of the algorithms. A user who has played role- playing games will want to configure his algorithmic character, to manipulate it to his own advantage because he will be familiar with configuring player characters. And he will know that an algorithmic system doesn’t really care what he does, as long as he appears to being playing the game.

34 Leave a comment on paragraph 34 0 Works Cited

35 Leave a comment on paragraph 35 0 Ark: Survival Evolved, Studio Wildcard, 2016.

36 Leave a comment on paragraph 36 0 Crosman, Penny. “Beyond Robo-Advisers: How AI Could Rewire Wealth Management.” American Banker, 5 Jan. 2017, https://www.americanbanker.com/news/beyond-robo-advisers-how-ai-could-rewire-wealth-management. Accessed 28 Oct 2018.

37 Leave a comment on paragraph 37 0 Dark Souls 3. FromSoftware. Bandai Namco Entertainment, 2016.

38 Leave a comment on paragraph 38 0 Dennin, James. “Here’s How Many US Jobs Could be Automated — and Why Robots Threaten American Workers Most.” Mic, 24 March 2017, https://mic.com/articles/172071/how-many-us-jobs-could-be-automated-and-why-robots-threaten-american-workers-most-mnuchin#.c5B0kEPFE. Accessed 28 Oct. 2018.

39 Leave a comment on paragraph 39 0 Diablo 3: Reaper of Souls, Blizzard Entertainment, 2014.

40 Leave a comment on paragraph 40 0 Domingos, Pedro. The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World. Basic Books, 2015.

41 Leave a comment on paragraph 41 0 DuBravac, Shawn. Digital Destiny: How the New Age of Data Will Transform the Way We Work, Live, and Communicate. Regnery, 2015.

42 Leave a comment on paragraph 42 0 Elex. Piranha Bytes. THQ Nordic, 2017.

43 Leave a comment on paragraph 43 0 Everquest 2. Sony Online Entertainment, 2004.

44 Leave a comment on paragraph 44 0 Fallout 4. Bethesda Softworks, 2014.

45 Leave a comment on paragraph 45 0 Gee, James. “Games and Learning: An Interview Overview.” Good Video Games + Good Learning, 2nd ed. Peter Lang, 2013.

46 Leave a comment on paragraph 46 0 Harari, Yuval Noah. Homo Deus: A Brief History of Tomorrow. Vintage, 2016.

47 Leave a comment on paragraph 47 0 Hill, Simon. “How Much Do Online Advertisers Really Know About You? We Asked an Expert.” Digital Trends, 27 June 2015, https://www.digitaltrends.com/computing/how-do-advertisers-track-you-online-we-found-out/. Accessed 28 Oct 2018.

48 Leave a comment on paragraph 48 0 Johnson, Steven. Everything Bad Is Good for You. Penguin, 2005.

49 Leave a comment on paragraph 49 0 Liptak, Adam. “Sent to Prison by a Software Program’s Secret Algorithms.” The New York Times, 1 May 2017, https://www.nytimes.com/2017/05/01/us/politics/sent-to-prison-by-a-software-programs-secret-algorithms.html. Accessed 28 Oct. 2018.

50 Leave a comment on paragraph 50 0 Lohr, Steve. “A.I. Is Doing Legal Work. But It Won’t Replace Lawyers, Yet.” The New York Times, 19 March 2017, https://www.nytimes.com/2017/03/19/technology/lawyers-artificial-intelligence.html. Accessed 28 Oct. 2018.

51 Leave a comment on paragraph 51 0 ———. Dataism: Inside the Big Data Revolution. Oneworld, 2015.

52 Leave a comment on paragraph 52 0 Lubarsky, Boris. “Re-Identification of ‘Anonymized’ Data.” The Georgetown Law Technology Review, April 2017, https://www.georgetownlawtechreview.org/re-identification-of-anonymized-data/GLTR-04-2017/. Accessed 28 Oct 2018.

53 Leave a comment on paragraph 53 0 Mann, Gideon, and Cathy O’Neil. “Hiring Algorithms Are Not Neutral.” Harvard Business Review, 9 Dec. 2016, https://hbr.org/2016/12/hiring-algorithms-are-not-neutral. Accessed 28 Oct. 2018.

54 Leave a comment on paragraph 54 0 Manovich, Lev. The Language of New Media. MIT Press, 2001.

55 Leave a comment on paragraph 55 0 Morrowind. Bethesda Softworks, 2002.

56 Leave a comment on paragraph 56 0 Planetside. Sony Online Entertainment, 2003.

57 Leave a comment on paragraph 57 0 Rollings, Andrew, and Dave Morris. Game Architecture and Design. New Riders, 2003.

58 Leave a comment on paragraph 58 0 Rainie, Lee, and Janna Anderson. “Code-Dependent: Pros and Cons of the Algorithm Age.” Pew Research Center, 8 Feb. 2017, http://www.pewinternet.org/2017/02/08/code-dependent-pros-and-cons-of-the-algorithm-age/. Accessed 28 Oct. 2018.

59 Leave a comment on paragraph 59 0 Scola, Nancy. “A Picture of You, in Federal Data.” Politico, 11 Oct. 2017, http://www.politico.com/agenda/story/2017/10/11/federal-data-individual-portrait-000540?lo=ap_d1. Accessed 28 Oct. 2018.

60 Leave a comment on paragraph 60 0 Skyrim. Bethesda Softworks, 2011.

61 Leave a comment on paragraph 61 0 Star Wars: The Old Republic. BioWare Austin. Electronic Arts, 2011.

62 Leave a comment on paragraph 62 0 Susskind, Richard, and Daniel Susskind. “Technology Will Replace Many Doctors, Lawyers, and Other Professionals.” Harvard Business Review, 11 Oct. 2016, https://hbr.org/2016/10/robots-will-replace-doctors-lawyers-and-other-professionals. Accessed 28 Oct. 2018.

63 Leave a comment on paragraph 63 0 “Tanking Guide Battle for Azeroth.” Icy Veins, 23 July 2016, https://www.icy-veins.com/wow/tanking-guide. Accessed 28 Oct. 2018.

64 Leave a comment on paragraph 64 0 Wexler, Rebecca. “When a Computer Program Keeps You in Jail.” The New York Times, 13 June 2017. https://www.nytimes.com/2017/06/13/opinion/how-computers-are-harming-criminal-justice.html. Accessed 28 Oct. 2018.

65 Leave a comment on paragraph 65 0 World of Warcraft. Blizzard Entertainment, 2004.

Source: https://opr.degruyter.com/playing-the-field-video-games-and-american-studies/jon-adams-why-we-play-role-playing-games/