A few days ago, there was an interesting ruling in the Triple Town / Yeti Town game cloning case, a.k.a. Spry Fox, LLC v. Lolapps, Inc. Triple Town and Yeti Town are both casual puzzle apps where tile elements are assembled and evolve. Spry Fox had sought to license the game to the defendants, but the defendants pursued a cloning strategy instead. The litigation presents roughly the same sorts of legal issues as the current EA/Zynga dispute and the recent Tetris clone decision.
This particular case has been talked up pretty extensively in the blogosphere over the last year. For instance, James Grimmelmann had some thoughts about the early stages of the litigation and game cloning generally:
"if Triple Town flops on the iPhone because Yeti Town eats its lunch, at some point Dave and his colleagues won't be able to afford to spend their time writing games any more."
Eric Goldman weighed in yesterday on this new ruling, connecting it to the EA litigation:
"The Triple Town ruling suggests that Zynga probably can’t score a quick win."
And even Marty Schwimmer chimed in with some thoughts about an Alaskan verison of Gone With the Wind.
At the heart of these cloning disputes is a simple question: how closely can you copy a video game? In other words, how does copyright law apply to video games? The cloning controversy is an old one, but there are many critics (and some defenders) of the common industry practice of copying game mechanics while essentially swapping in new art assets to make the clone not an exact copy. But does that strategy always work as a matter of copyright law?
The big problem for copyright protection here is that copyright does not protect games as systems -- something that Bruce Boyden explains (excellently) at length in this paper. Section 102(b) of the copyright statute codifies the so-called "idea/expression" dichotomy and expressly forbids copyright protection for ideas and systems. But as applied to video games, where is the line between expression and system?
This case doesn't answer that question, but it contributes to the discussion. The procedural posture in Triple Town is a ruling on a motion to dismiss, which puts a higher burden on the movant -- the defendant here. The legal question is whether Spry Fox (the plaintiff) has alleged facts that, if proven at trial, could result in a favorable ruling. So this opinion is no decision on the merits -- this is just about whether Spry Fox can make out its case in theory. However, in a typical copyright case like this one, most of the facts are pretty obvious, so I think it's fair to say that this ruling does a little bit more than the typical ruling on a motion to dismiss.
So it is worth paying attention when the court says interesting things about the nature of video games:
"...written description of the visual aspects of the games is not as effective as the side-by-side 'screen shots' that Spry Fox provided in its complaint, and that those screen shots are in turn less effective as a comparison between the two games than actually playing them. Neither Spry Fox nor 6Waves suggested that the court should compare Triple Town and Yeti Town by actually playing them."
In the past, many courts have analogized video games to films and other conventional media in order to perform an infringement analysis. But am I right to read this as the court saying that procedural similarity was not something briefed by the parties so it's waived (perhaps)? The court makes a similar point later:
"It is as difficult to compare two video games by looking at a few screen shots and reading written descriptions of game play as it is to compare two movies by looking at posters and reading excerpts of screenplays"
So my impression is that the judge would welcome a comparison of game play in order to do a copyright infringement analysis, which is pretty interesting. Not many courts have opened that door.
"At this stage of the litigation, where the court has only the complaint, its description of Triple Town, and the accompanying screen shot images, the court concludes that the idea underlying Triple Town is that of a hierarchical matching game, one in which players create objects that are higher in the hierarchy by matching three objects that are lower in the hierarchy. Frustrating the player’s efforts are antagonist objects; aiding the player are objects that destroy unwanted or ill-placed objects. Spry Fox’s copyright gives it no monopoly over this idea. 6Waves (or anyone else) is free to create a video game based on the same idea...."
So there can be no copyright for game design at this level of game mechanics, because protecting that would amount to protecting an idea/system, which is barred by 102(b). Pretty interesting. But then, in what some have called a "switcheroo," the court says:
"A video game, much like a screenplay expressed in a film, also has elements of plot, theme, dialogue, mood, setting, pace, and character. Spry Fox took the idea underlying Triple Town and expressed it with its own characters, its own setting, and more. These objective elements of expression are within the scope of Spry Fox’s copyright."
So... to the extent that a game idea is embodied in "objective elements of expression" that resemble "plot, theme, dialogue, mood, setting, pace, and character," those things are not an "idea" and may be within the scope of copyright. But how do we distinguish between "objective elements of expression" and the "idea" in Triple Town? The idea part of Triple Town seems to be set forth above, but what are the "objective elements" of Triple Town? The court doesn't spell this out in the abstract so much as it demonstrates its understanding by applying it to the facts:
"Spry Fox’s allegations are more than adequate to illustrate plausibly the objectively similar expression embodied in Yeti Town. The object hierarchy is similar. Progressing from grass to bush to tree to hut is similar to progressing from sapling to tree to tent to cabin. Perhaps more importantly, the object hierarchy coupled with the depiction of the field of play comprise a setting and theme that is similar to Triple Town’s. A snowfield is not so different from a meadow, bears and yetis are both wild creatures, and the construction of a “plain” is not plausibly similar to the construction of a “patch,” at least as the two games depict those terms. Whether 6Waves’s choice of language in its dialog boxes is similar enough to Spry Fox’s is a closer question, but it is a least plausibly similar. There are apparent differences between games (for example, yetis are not bears and “bots” are not campfires), but a court must focus on what is similar, not what is different, when comparing two works."
Did you follow that analysis? The court has said that "hierarchical matching" is an idea above. But here the court seems to place the particular "object hierarchy" within the scope of copyright. Bears are not yetis -- but Yeti Town resembles Triple Town in that it mirrors a "wild creature" object within the hierarchy. Snowfields are not meadows, but again, this is just swapping out one object for a similar object that can fit in the same slot in the hierarchy. And the end result creates a similar "setting and theme." I think I get what the court is doing here, but Eric says it "isn't entirely clear" and Marty concurs about the ambiguity.
But in my opinion, if you borrow some basic concepts from object-oriented programming, it isn't that hard to understand. E.g., meadows and snowfields are part of a terrain class, yetis and bears are a mob class, etc. Class similarity within a game structure, can count for something in copyright law. Just as narratives can deploy certain characters (the sidekick, the snitch, the brute) in certain patterns, so video games can be formally mapped as particular sorts of object relationships with a skin of narrative.
This analysis, I think, resonates with Jesper Juul's notion in Half Real that games are merger of rules and fiction. It also ties in with Ian Bogost's forthcoming Game-O-Matic -- a piece of software that takes certain relationships embodied in a formal game structure and allows the user to insert the objects that the game "verbs" animate.
Again, it's important to note that this decision contributes to a conversation about game copyright -- it does not settle the issue. But it nudges copyright a little bit further toward protecting game mechanics. Even if you don't follow what the court is doing, I think that, based the court's analysis, if this were EA v. Zynga (instead of Spry Fox v. Lolapps), EA would win.
But the more important question is: What should the rule be here? Game design as an art deserve some form of protection, surely. Bruce is right that 102(b) is tilted against the protection of games, but what is really behind the idea/expression dichotomy? The way I explain the idea/expression line to my students is to say that the copyright monopoly should end when copyright threatens private control over things that are truly important. Facts, for instance, must be available to all. Important ideas (discoveries, theories, laws of nature) must also be available to all. Basic plot structures and genres must be available to all, because they serve as platforms for new creativity. So the idea/expression line is really about, imho, keeping certain things outside the sphere of private ownership as a matter of optimal policy.
I admit that I am not happy about companies that clone innovative game mechanics. It does not seem fair or likely to produce a better marketplace in game titles. When a game is essentially cloned, there's a danger that the production of genuinely creative games will go unrewarded and that companies making quick clones will be more profitable than those that spend money on new ideas. That does not seem like a fair or healthy way to set up the games industry.
But there are serious dangers in letting copyright be the solution to this unfairness. If the protection of game mechanics expands too far, the cure could be worse than the disease. Game companies could end up owning specific video game genres for over a century due to the crazy duration of contemporary copyright law. The creation of new independent games could be chilled by fear that the mechanics too closely resembled those in a game by EA, for instance. That result would be as bad-- or worse-- than the clones.
Hopefully the courts can find some kind of happy balance, shutting down the most egregious clones without creating monopolies in new genres. But I'm a bit worried, given the trends we've seen in other realms of intellectual property law. Awarding copyright protection in plot-like game mechanics in a way that maximizes the prospect of future creativity will be a tough thing to do. Courts pushing out the boundaries of copyright in games should do so very carefully.
Hmm, I think that what the court is saying could be expressed by analogy.
Let's say that the clone involved dice on a checkerboard. The hierarchy was expressed by the number of pips that were face up. Get three threes together and they collapse to a four, and so on. The obstacles were represented by a poker chip. Same game... classic "dressing" versus "salad" example. But the expression is very different. No plant life. No animals.
In my reading of what the court is saying, the expression is substantially different in my hypothetical. The issue they are focusing on is that the *dressing* is quite similar in the two games in the suit.
As another hypothetical... let's say that someone cloned the game, with black and white pandas. They color-shifted the art, but everything was still bushes and houses. There the dressing is even MORE similar.
Now we have a spectrum... the dice version is almost certainly not copyright infringement. The one that is color shifted pandas almost certainly is. The question is whether any of the ones in between are.
I don't see how this is different from copyright cases in other media, which have also hinged on HOW similar things are.
Posted Sep 28, 2012 4:52:58 PM | link
Yeah, I think you're right, but it seems to me that what the court is doing is a little different that what you see in other media. If you read Marty Schwimmer's post, he seems to be saying that Gone With the Wind set in Alaska would be a totally different film, which I read as essentially: "What is this court saying? A snowfield is not a meadow and yetis are not bears." So, if that's true, I think it must be relevant that the yetis and bears peform a wandering/blocking function and the meadows and snowfields perform a blank title function, and there are certain similar steps in the chain of evolution as the scoring pieces move to higher value. So, to a certain extent, the function of the object in both games dicates the form. (E.g., yetis and bears are animals because they move and prevent placement -- a meadow stays still and works as a tile, but it doesn't move so it couldn't be used to skin the bear/yeti function in the game.)
Perhaps it is analogous to other media in that you can do something similar in a narrative analysis by, e.g., pointing out that certain characters appear at specific places in narrative structures to perform certain functions. E.g. sidekick rescues hero as means of introducing sidekick and sidekick is held captive later in the plot -- so stories have certain structural slots where elements can be fit. The progression up the hierarchy in Triple Town is kind of like that too -- it's a temporal progression from grass to bush to tree. But some parts of this -- e.g. bear/yeti vs. meadow/snow seem like a particular structural/functional pairing that you wouldn't see, at least in that form, in a standard narrative.
I admit, though, that if these games were reduced to bare formal elements -- poker chips and numbers, e.g. -- the court's copyright analysis wouldn't make sense. It couldn't say "bears are like poker chips." So I agree with you that there's still much about this that reads like a conventional copyright comparison. So I should be careful, I guess, not to suggest that the opinion would find completely formalistic game mechanics (bare rules) subject to copyright.
Posted Sep 28, 2012 5:44:09 PM | link
Moving more away from the exact case here, and how the judge is handling it, and more towards answering your closing fears and questions:
I agree that mechanics are ideas and should not be copyrightable. However, is there room for the idea of copyrighting (or in some way having IP over) particular implementations of mechanics? I am not thinking here about "dressing" in Raph's terms - perhaps what I mean his closer to his definition of content as statistical variation on challenges.
As an example, you could describe the mechanics of chess as "players take it in turns to move pieces across a grid, aiming to capture their opponent's pieces". The thing that makes chess uniquely chess, amongst other turn-based capture games, is the different kinds of pieces and the ways in which they can move and capture, as well as the starting positions and the rules around checkmate (and more: castling, etc.).
Chequers uses the same core mechanic, but fleshes it out with very different content. It is not a "copycat" game. Whereas if I made a game that was exactly like chess, except I renamed and reskinned the pieces, and changed it so that pawns could move backwards, I would have changed the balance of chess, but the differences in design would be small, and it should probably be an IP violation - if chess weren't in the public domain!
Similarly if someone looks at Diablo 3 and says "that's a great skill progression system for an RPG" and then copies it wholesale, that could be IP infringement. Whereas if they take the basic idea (equip N abilities at a time whenever you want from the ones you have unlocked so far, each of which has M different sub-abilities that can be set) and choose a value for N and M, and create their own, different, abilities then it would not. Copyright similarity principles could be used in court to decide if the abilities that the new game has are too similar to the ones in D3.
So under this, Blizzard put effort into coming up with the core mechanic, and that can't be protected. However, the much greater effort of populating that mechanic with content and abilities that are fun to play and balanced is protected.
This was just me thinking aloud. Has someone else has already written about this sort of idea? What are the problems with it? Could this be a starting point for something workable?
(Obviously similarities in dressing would still be copyrightable in the usual way - if developer A steals the plot and graphics wholesale from developer B's game that's copyright infringement).
Posted Sep 29, 2012 5:30:22 AM | link
I'm not sure I understand your idea. Are you saying that if we took chess (assuming it was copyrighted) and replaced the particular pieces with bears or fruit or elements from the periodic table, that would be infringing? So the mechanics would get a "thin" protection, but it wouldn't extend to the "core" mechanics?
If I'm getting that right, I think that's interesting, and maybe how this will evolve over time. For me, the conceptual problem is that I don't know how to draw a line between a "core" mechanic and protected particular embodiment of that core mechanic.
In a way, this calls to mind patent law, where you *do* have a protected core invention, spelled out explicitly in the drafted claims. Any use that falls within the claims is infringing, but any use that falls outside the claims is okay. What you're describing (I think) is the reverse. Anything that duplicates the core mechanic is fine, but particular embodiments of that core mechanics might be protected.
If I'm understanding you right, that's appealing, conceptually, to me, but again, drawing the line would be heard.
Posted Sep 29, 2012 8:55:26 AM | link
Greglas: Yes, you have understood me. I realise that the complexity in my suggestion is in drawing that line - that is why I was saying my post was just a suggestion of an area to think about.
One of the reasons it occurred to me is that I have heard professional game designers repeatedly say that coming up with an idea is the easy bit - it's refining it into a polished, balanced, fun system that is the real effort. So protecting that effort, however you define it, seems to be a useful aim.
Posted Sep 29, 2012 2:24:22 PM | link
There are two reasons for the blatant copying taking place. At first the two reasons might sound paradoxical, so bear with me.
1) Too much (stupid) money. 57% of all investment in games has been in mobile games. If you look at the appstore, you'll see a lot of cloning (e.g. Tiny Co's Monsters is a copy of Dragonvale). Same thing with Facebook.
2) (Relatively) cheap to develop a mobile/social game. Even companies not funded to the hilt can get away with copying a proven game by slightly changing the theme.
No one has bothered copying Ultima or Civilization in the 90s. You have to ask what changed.
Posted Sep 29, 2012 2:28:03 PM | link
The key to this discussion is here:
"Spry Fox had sought to license the game to the defendants, but the defendants pursued a cloning strategy instead."
If you're a publisher who's been approached by a studio, you shouldn't be allowed to decline to publish their game and then make a blatant copy of that same game.
Now, there is room to argue that this sort of restriction doesn't belong in copyright law, but instead should be enforced by contract. If this is the case, then Spry Fox should have insisted on language to this effect in their contract with lolapps before they entered into negotiations over Triple Town.
If they didn't, and if they lose the case because of this fact, then the industry needs to just chalk this debacle up as a learning experience, adapt and move on.
Regardless of the outcome, though, informed consumers should not be ok with what lolapps did, and I hope that they'll choose to play Spry Foxes' Triple Town instead.
Posted Sep 30, 2012 10:30:19 AM | link
Greg, thanks for the shout-out! To answer Riftstalker's question, I think what's changed is that we are, again, in a boom market. The boom this time is in smartphone gaming apps, but previously you saw similar disputes arise in the early '80s involving arcade games. In a 15-month period in 1981-82, there were 8 separate cases decided involving allegedly infringing arcade games, including 3 appeals, and two International Trade Commission proceedings to bar importation of certain arcade games. What gave rise to all this litigation was a combination of the sudden emergence of a lucrative market, combined with uncertainty about the underlying rights. Not only were video game copyrights uncertain, but the protectability of computer code itself was not fully established. So there were a lot of blatant knock-offs, and courts steered around the problems in determining what, exactly, was protected in the game play by classifying video games as audiovisual works, the same as cartoons or movies. The knockoff cases fade out after August 1982.
On the opinion itself, I'll have to find time to write up a full post, but the "switcheroo" you identify Greg seems to me to be flat-out contradictory. The court bows to precedent that says that game structure is not copyrightable, but then holds that game structure is analogous to the plot of a novel and thus copyrightable. I think this is a mistake for the reasons I argue in my article, but when it comes to video games I think courts have been apt to conflate the rules and structure of the game with nonliteral protected aspects of other works, such as movie or novel plots. Games may have plots, but games are not plots.
Posted Sep 30, 2012 1:38:45 PM | link
Thanks, Bruce -- yes, just because I think I understand what the court is saying doesn't mean I think it fits with precedent or the copyright statute. But contradictions like this are how things change sometimes -- but it helps a lot if the Supreme Court performs the contradiction. :-)
I don't know if either of these parties has the energy to pursue an appeal, but it will be interesting to see if the district court in the EA case is influenced by this.
Matt -- Triple Town is really much better. I played them both before writing this and I admit that I've come back to Triple Town more than once since... :-)
Posted Oct 1, 2012 4:09:55 PM | link
I'd like to see what the 9th Circuit does with the issue, although a final decision here is probably a year away. I like the Allen v. Academic Games League opinion, with its somewhat delphic "Games are meant to be played" pronouncement (i.e., playing a board game in public is not a public performance), but I'd be skeptical a panel would straightforwardly apply the game rules cases to videogames. Even apart from the tricky question of what constitutes videogame "rules," and whether game rules are only noncopyrightable because they are instructions to humans (I argue no, but I suspect the 9th Circuit would just stop at "videogames are audiovisual works").
Posted Oct 2, 2012 4:44:42 PM | link
I think it's probably a mistake to read too much law into the settlement here.
6waves is a company in trouble. They might well have defended this effectively, but it was easier and cheaper for them to settle, giving up a game that hasn't been a big hit for them, and getting out of a situation that makes them look like assholes.
Mind you, my sympathies are entirely with SpryFox, almost regardless of what the law is; but this isn't really a precedent. It's a company with financial troubles shedding one of them.
Posted Oct 16, 2012 12:47:59 AM | link
Well, the opinion on the motion to dismiss means something, but it's pretty weak precedent.
But you're right re the settlement -- and it's usually very hard to read anything into a settlement if the terms are not disclosed, which they usually are not. Most cases settle, and it's hard to tell who "won" given that it could be for a token sum or a substantial amount of money. Cases often tend to settle after rulings like this, though, where the court tips its hand to where it may be heading with a ruling. Easier to save the costs of litigation and reach an agreement based on the signals in the opinion.
Posted Oct 16, 2012 9:36:17 AM | link
When comparing the Triple Town and Tetris analyses, I noticed that the two judges seem to contradict each other, or at least offer quite different perspectives, on the difficult question of the protectibility of core game mechanics that you discussed. For one specific comparison, look at how they both deal with the question of copying a game's playfield dimensions.
Here's the Tetris opinion on Xio's use of Tetris Company's 20x10 playfield:
"Xio copied a field that was the exact same dimensions as Tetris. Even assuming it is a rule to have a field higher than it is wide, which the Court does not necessarily find, it is not a rule to have the playfield be exactly 20 units by 10 units. Xio was free to program a puzzle game with the playing field designed “in an almost unlimited number of ways” as admitted by Xio’s expert. Xio was not limited to those precise dimensions and was free to take the general idea of having a long game board and express it in its own unique way. For example, it could have had a field three times as high as it is wide or 15 units high by 8 units wide, without copying the exact game dimensions and infringing the look and feel of Tetris’s expression. Thus, I find this to be protectible expression that Xio infringed."
Now, here's the Triple Town order's analysis of Lolapps's use of Spry Fox's 6x6 playfield:
"Although the court need not decide the issue in this motion, it appears that some elements of Triple Town are not protectable because they are functional. Much as copyright does not protect ideas, it does not protect the “functional process[es]” that are “indispensable to the idea” inherent in a game. For example, Spry Fox’s choice of a six-by-six game grid is not likely an expressive choice. A grid that is too small would make the game trivial; a grid that is too large would make it pointless. There is perhaps a range of functionally appropriate choices for the dimensions of the game grid; perhaps a seven-by-seven grid, or a six-by-seven grid, would serve the game’s purposes just as well. But it would extend copyright protection beyond its proper scope to afford protection to a functionally-dictated choice like this one."
Both judges make sure to review the video game case law, but the ambiguity in where exactly to draw the idea/expression line when analyzing game mechanics, or which of the rules in a game are to be considered 'functional process[es]' that are 'indispensable to the idea', allowed the judges to form completely different arguments and conclusions from very similar circumstances. Both judges even specifically consider the possibility of differently-sized playfields to support their argument. However, the Tetris judge decides that it's self-evident that changing the game board's dimensions wouldn't affect the game's 'function', while the Triple Town judge decides that it would (or at least that it might).
Maybe one way of understanding this ambiguity is uncertainty in the precise 'function' a video game (as entertainment-producing device) dictates, and whether game mechanics form any part of this function. One may take a utilitarian perspective (as did the Tetris judge), where a game is understood to 'function' only insofar as it creates an audiovisual work as its output. In this view, any particular game mechanics would 'function' as well as any other, disregarding any qualitative differences in the resulting gameplay, and thus extending copyright protection to 'core' game mechanics. Or one may take a more performative perspective (as did the Triple Town judge), where the function or purpose of a game also incorporates the possibilities for the player's own action and engagement. In this view, the specific affordances and constraints produced by the game mechanics would also serve a function in the overall 'cybertextual' system, and would thus not be covered by copyright (or trade dress, as in the Tetris case).
Whether this latter, performative perspective on game copyright could be linked back to any supporting case law distinct from the standard idea-expression ambiguity, I'm not sure. As I understood it, this is something like the history Bruce Boyden traced out in his "Games and Other Uncopyrightable Systems" paper (which was cited by Xio in its defense brief), but the judge's opinion ignored this claim: "Xio argues that some independent principle protects game rules rather than it being the natural extension of the idea-expression dichotomy applied to games. I find no case law or support for such a premise."
In general though, the lack of any analytical method other than 'common sense' being applied in the US courts to resolve this ambiguity makes the the question of copyright (and trade dress) infringement in computer games quite unpredictable (arbitrary, even).
Posted Nov 7, 2012 10:12:13 PM | link
Great observations -- thanks. I hadn't zeroed in on the grid comparisons between those two cases and it's an excellent point of entry for the question of functionality.
What's interesting to me is that from a game analysis perspective, there really is a difference (I think) between the grid dimensions in Tetris and in Triple Town. 20X10 in Tetris works in a way that is somewhat different than 6x6 in Tripletown. It's sort of analogous to copying the use of one ball for a game versus copying the use of 57 balls in a game. The latter choice seems more arbitrary, hence more subject to protection as an expressive choice of the game creator. By comparison if Tetris were a 22X10 grid, where the blocks fell one more step, I don't think it would change the game play so much, but if TT were 6x8, it would change the game substantially.
Also, it's interesting that the TT court is ultimately leaning in favor of protecting game play (in my opinion) and, for that reason, as use say, realizes that the grid size *functions* in terms of its influence on player performance. But despite finding that functionality, the opinion seems to be reaching for ways to view game mechanics as protected to the extent they can mirror narrative progression. So to the extent the TT court takes away the game field size as functional, its recognition of the role of that element in structuring performance leads to a greater solicitude for that form of creativity.
To the extent the Tetris court focuses on the expressive choices in the pre-player text as controlling the infringement analysis (the non-performative text), this favors protection, but protection of Tetris as a template primarily, not as a system. And that's more consistent with what Bruce says in this paper.
Re "lack of an analytical method" -- that's true, but it's pretty common with respect to the law and new technologies. Especially with substantial similarity analysis, courts do make things up when they encounter new sorts of problems. Over time, this creates common law precedent, which ultimately gains some stability and works to order investments. The video game case law is "only" 30 years old and I don't think the problem is quite solved at this point. So for the moment, the line does seem arbitrary, but as these and other opinions trickle down, we might get some more definite lines to offer guidance to the Zyngas and 6waves of the world. However, it seems that with copyright law, there is always going to be indeterminacy about the exact scope of the right to prevent similar works.
Posted Nov 8, 2012 5:55:37 AM | link