Last time, we talked about copyright and video games we discussed how copyright protects the literal computer code of the gaming programs, as well as what you see on the screen. Certainly, copycatting in the video game industry often has been rampant. This article will get a bit more legally technical and go back and explore in great detail *how* courts determine if two computer programs are actually similar enough that one program used the other one without permission and therefore infringed its copyright.
Disclaimer: The following information does NOT constitute legal advice and is only for general educational purposes. Each situation is different and specific legal issues usually require additional research and investigation, so do not rely on this article to address a particular legal issue; use this as a starting point to gain a general understanding. This article, although educational in purpose and substance, nevertheless, might be deemed attorney advertising, and prior results do not guarantee future success.
4. Idea v. Expression Dichotomy in Copyright Law
As I said last article, if you take, line for line, the literal exact copy of one computer program to another, without permission, yes, that is usually copyright infringement. This applies both to the source code (the language the programmer uses) and the object code (the ones and zeroes into which the source code is “complied”). We also discussed how copying the “sequence” of an audio-visual display might qualify for copyright infringement (for instance, you developed your own code, but your game looks just like a pre-existing one, such as Ms. Pac Man). What about copying part of a computer program? Or if you don’t exactly copy it word for word, but copy it’s … structure - the “logic” of the way it is arranged?
It is a basic truism of copyright law that you cannot copyright ideas - that’s the reason everyone has a murder mystery story, because the idea of someone offing their spouse for the insurance policy, cannot be copyrighted. However, the expression of the idea - the characters, acting, plot twists, locations, etc. as a whole, can be expressed differently, and those differences could be copyrighted. This is what’s called the "Idea-Expression dichotomy."
So, what’s this have to do with computers? Well, the classic example is accounting ledgers, or today’s classic spreadsheets. There isn’t a whole lot of wiggle room on how they are able to look. A bunch of columns and rows, right? So you can’t copyright that. Seems a fairly simple concept, right? This sort of thing can flow down even to the computer program. I mean, how many ways are there to write code, in the same computer language, to generate a program that creates a spreadsheet? For sure, Excel looks and feels a bit different than Googlesheet, and at its higher levels, Excel many even have additional functionality (or vice-versa, I don’t know which is actually a more powerful tool). But is the programming to create them so different? Perhaps, perhaps not.
If there is no unique way to express it, that might lean against copyrightability, being more of an idea. Then again, if there is a means to “express” the programming very differently - more efficiently, more elegantly, with more functionality - that unique manner of programming could very well be copyrightable. One court actually stated that the actual coding of a computer program could be the easiest part of the entire process in developing it. The court said that, by far, it is harder to take a general idea (I need a spreadsheet) and organize the various routines, subroutines, and interdependent parts, then actually coding all of it, once it is laid out (although the court did admit “debugging” was time intensive and difficult).
5. Substantial Similarity
Ok, now that I know if my program borrows to much of either your language or even maybe your logic, I *might* be guilty of copyright infringement. How do you go about proving it?
Copyright 101 generally says, you must show (i) you own the copyright on the allegedly infringed materials and (ii) that the other party actually copied it (as opposed, to say, developing it on their own).
Further, as it is rarely possible to prove copying through direct evidence, copying may be proved inferentially by showing that the defendant had (a) access to the allegedly infringed copyrighted work and (b) that the allegedly infringing work is substantially similar to the copyrighted work. This is known as the “Substantial Similarity” standard.
5a. The Extrinsic v. Intrinsic Tests of Substantial Similarity
From 1946 to 1986, one of the leading copyright cases in judging “Substantial Similarity” was Arnstein v. Porter (1946), which created a two step test, and originally was used in the context of musical compositions. This two-step test had an (1) Extrinsic Test and an (2) Intrinsic Test. If under *both* tests, the alleged copying work was judged “too similar” to the alleged original, then copyright infringement occurred (provided other elements of copyright infringement are met). If only one of the two tests are met, then there is no copyright infringement.
Step 1 (“Extrinsic Test of Substantial Similarity”): A finder of fact, with expert assistance if necessary, weighs the alleged original vs. the alleged copy, to determine if both are substantially similar, from a technical, or objective point of view. If not, there is no copyright infringement. Why the need for experts? One court used an allegedly infringed sculpture as an example, and stated the analysis might turn on specific criteria which can be listed and analyzed, such as the type of artwork involved, the materials used, the subject matter, and the setting for the subject. Can you imagine an “ordinary” person trying to distinguish between Creama Beige and Creama Marfill marble, and weighing their legal Substantial Similarity?
Step 2 (“Intrinsic Test” a.k.a. “Ordinary Observer Test of Substantial Similarity”): If there was an objectively technical Substantial Similarity, then the fact finder (for instance the jury, but could be a judge), must decide - without the aid of expert testimony - but with the perspective of the "lay observer," whether the copying was "illicit," or "an unlawful appropriation" of the copyrighted work.
(Extrinsic) Ideas v. (Intrinsic) Expression: Put another way, one court, in the case of Sid & Marty Krofft Television v. McDonald's Corp (9th Cir. 1977), described the Extrinsic Test as “expertly” and “objectively” comparing the Substantial Similarity of *ideas* only, whereas the Intrinsic Test, *subjectively* compares the *expression* of those ideas. Copyright “cannot be limited literally to the text, else a plagiarist would escape by immaterial variations.” Again, in order for there to be infringement, the experts, objectively must believe there is a Substantial Similarity of ideas, and also the ordinary person must subjectively believe there to be a Substantial Similarity in the way those ideas are expressed.
No “Bright-Line” Rule: Just remember though, while you can theoretically copy ideas, you cannot copy their expression. Obviously, you can’t get to the “expression” part, if you didn’t even copy the idea, though. That said, courts repeatedly have said there is “no bright-line” rule; rather, every case must be judged according to its own specific facts and “feel” and that’s the reason the subjective test (the “Intrinsic Test”) is so important.
A Common Sense Example of All This: In the Sid & Marty Krofft case, various children’s characters from PufNStuf were allegedly copied by McDonalds. For instance, PufNStuf wears what can only be described as a yellow and green dragon suit with a blue cummerbund from which hangs a medal which says `mayor', whereas `Mayor McCheese' wears a version of pink formal dress - `tails' - with knicker trousers. He has a typical diplomat's sash on which is written `mayor', the `M' consisting of the McDonald's trademark of an `M' made of golden arches.
It should be pointed out that McDonalds specifically targeted PufNStuff artists and designers in designing the McDonalds character. Thus, McDonalds had access to the copyrighted designs and made only subtle changes to the general idea and expression; in other words, the average child was not likely to notice or care about the difference in shape or color of the “diplomat sash” vs. the “cummerbund” - that’s the power of *subjectivity* of the Intrinsic Test - to look beyond superficial differences, to determine if the whole was an infringement.
5b. Structure - Sequence - Organization (SSO) Test of Substantial Similarity
About forty years later, courts struggling to apply this Substantial Similarity standard in the copyright context to computer programs, realized this “Extrinsic vs. Intrinsic” test no longer worked very smoothly. In 1986, a new watershed standard began to develop, in the case of Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., a case in the United States Third Circuit Court of Appeals.
The Third Circuit, by the way, is one of twelve intermediate federal appellate “Circuit Courts,” each of which cover a wide geographic area of states (Circuits #1 - #11), with the exception of the D.C. Circuit (that just covers the D.C. area), and there is also the Federal Circuit which handles certain appeals based on subject matter, such as patent claims. If you don’t know how this works, basically you bring a case in one of the 94 lowest federal courts (called “United States district courts”) and if someone loses, they appeal it to the local Circuit Court (for instance, the United States Court of Appeals, Third Circuit).
If a litigant needs to appeal the case again *theoretically* they can petition the United States Supreme Court to hear the case. However, the U.S. Supreme Court only rules on about 80 cases per year (out of the 8,000 seeking a final appeal, or about one percent). So, as a practical matter, overwhelmingly, the local U.S. Circuit Courts really make the important law in this country. The only “rub” is that technically they only affect the geographic area in which they oversee, so the “important law” can be drastically different in New York vs. California or anywhere in between - even on exactly the same set of facts (!!). This is the reason some people like to sue in certain states and not others - if they have a choice.
The facts in Whelan are a bit convoluted. Basically, Party A, wanted a scheduling program for Party A’s business and Party A hired Parties B-1 and B-2 to develop it (let’s call the underlying code, Code #1). Parties A, B-1 and B-2 all claimed they had at least partial if not full rights to Code #1. Eventually, Party A, started a new company focused on selling Code#1, and also Code #2, which Party A theoretically independently developed with many unique elements, but which also was based on much of Code #1.
Meanwhile Party B-2 started their own business and was using Code #1. Party A sued Party B-2 to stop using Code#1. Party B countersued saying they (Party B) owned the rights not only to Code #1, but also to Code #2 since it was a “derivative work.” However, no one alleged the actual code was copied literally - what they alleged was that the overall “structure” of the program - the way in which the routines, and subroutines were arranged conceptually, was copied.
While analyzing the copyright law for this case, the court said the existing two-part test (Extrinsic-Intrinsic) was not terribly useful. “The ordinary observer” test (the second, “Intrinsic” part) which was developed in cases involving novels, plays, and paintings, and which does *not* permit expert testimony, is of doubtful value in cases involving computer programs on account of the complexity of the programs and their unfamiliarity to most members of the public.” So, what with to replace it?
The court came up with a new test, called the "Structure - Sequence - Organization" test of Substantial Similiarity. A very complicated, thorough, and extremely thoughtful decision boiled down to a paragraph is thus: it is not necessary for an infringer to actually copy any lines of code per se; rather they could copy the structure, sequence, and/or organization of computer program. Further, the infringer need not copy all of the logic or structure of the program; rather, the “court must make a qualitative, not quantitative, judgment about the character of the work as a whole and the importance of the substantially similar portions of the work.”
Under this test, in essence, if you are a computer programmer and programmed Game A for Company A, and then you move to Company B, and with your skills, knowledge and expertise acquired at Company A, you then program Game B for Company B, which is very different from Game A, *but* which uses much of the structure, sequence or organization of Game A’s programming, theoretically, you might be guilty of copyright infringement even if both games look different or rely on different coding.
Why? Because remembering above (1) you had access to the underlying code (the stuff you programmed at Company A) and (2) the two programs are “Substantially Similar.” How do we know they are Substantially Similar? We compare the “Structure, Sequence, and Organization” of the routines and subroutines. Obviously, the more similar they “look and feel” the more chance you will have being found of copyright infringement.
5c. Abstraction-Filtration-Comparison test (AFC)
So, remember how I said before, that the United States Circuit Courts, technically, only control the geographic areas they serve? That, in other words, the Second Circuit, could do things differently and rule a different way, than the Third Circuit, even with the same issues, or even the same set of facts? Well, that’s kind of what happened.
A few years after the Whelan Associates case, and its Sequence-Structure-Organization test of Substantial Similarity, the Second Circuit came out with its own test. In 1992, in the case of Computer Associates Intern Inc. v. Altai, developed something called the Abstraction-Filter-Comparison test. The theory behind this test is that, as above, the closer something is to an idea, rather than the expression of the idea, the less you can copyright it.
So you can’t copyright the idea of a spreadsheet or of a scheduling program. Further, if realistically, there is only one “good” way to code such a program, you can’t copyright that either, because essentially then you are copyrighting the idea. The further away you get from that idea, the more ways there are to do the coding for instance, the more creative it is, the more chance it is copyrightable.
So, you would analyze each portion of the of the program, it’s routines, its logic, its code, and figure out what level of “abstraction” it is; the more detailed, original, and less it is “required” to be there, the more it resembles something closer to the *expression* of the idea, rather than the idea itself. Those things that look more like an idea or are inseparable from the idea because there is only “one way to do it” you “filter out” as not deserving copyright protection.
Once you do *all* that … well, then finally, you “compare” what’s left and measure it’s Substantial Similarity. At this point, the court's substantial similarity inquiry focuses on whether the defendant copied any aspect of this protected expression, as well as an assessment of the copied portion's relative importance with respect to the plaintiff's overall program.
If you copy someone’s program generally, not literally, you still might be liable for copyright infringement, especially if you had access to their underlying processes or coding (which could include hiring away your competitor’s programmers).
The courts will look far beyond literal copying, i.e. more than word for word. Over time, the courts have used several tests (1) Extrinsic & Intrinsic or (2) Structure - Sequence - Organization or most recently (3) Abstraction - Filtration - Comparison, to compare computer programs to determine if the allegedly infringing work is “Substantially Similar” to the original, in which case the offender likely will liable for copyright infringement.
Oscar Wilde supposedly said, “imitation is the sincerest form of flattery” but if you wish to remain out of court, be very careful of what you copy and how.
Hyperlinks Enabled Try Them!
 https://en.wikipedia.org/wiki/Ms._Pac-Man (my person best high score 180,000 on the “fast” version).
 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F. 2d 1222, 1230-31(3rd Cir. 1986).
 Reyher, v. Children’s Television Workshop, 533 F.2d 87, 90 (2nd Cir. 1976) cert denied, 429 U.S. 980 (1976).
 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F. 2d 1222, 1231-32 (3rd Cir. 1986).
 154 F.2d 464, 468-69 (2nd Cir. 1946).
 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F. 2d 1222, 1232 (3rd Cir. 1986).
 562 F.2d 1157
 Nichols v. Universal Pictures Corp., 45 F.2d 119, 121 (2nd Cir. 1930).
 Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F. 2d 487, 489 (2nd Cir. 1960).
 Sid and Marty Kroft Productions, Inc. v. McDonalds Corp, 562 F.2d 1157, 1166-67 (1977).
 797 F. 2d 1222(3rd Cir. 1986).
 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F. 2d 1222(3rd Cir. 1986).
 Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F. 2d 1222, 1245-46 (3rd Cir. 1986)
 982 F. 2d 693 (2nd Cir. 1992).
 Computer Associates Intern., Inc. v. Altai, Inc., 982 F. 2d 693, 707-708 (2nd Cir. 1992) (this is often referred to as the “Merger Doctrine” in copyright law.
 Computer Associates Intern., Inc. v. Altai, Inc., 982 F. 2d 693, 706, 710 (2nd Cir. 1992).