Slashdot Mirror
Court Addresses Legality of Shrinkwrap Licenses
NullProg writes "This article here comments on a legal case where a shrink-wrap license may be binding. This a scary precedent for any developer who has added a feature to their software
already present in a competitors version."
Because the GPL expands upon the standard rights you have with any copyrighted work that you come accross (which by default are very few). The "viral" nature of the GPL only comes into effect when you attempt to redistribute the work. By default, you can't redistribute a copyrighted work at all (except for dwindling fair use exceptions), so even the part of the GPL that bothers everyone is actually a relaxation of restrictions on your activities.
Typical EULAs attempt to restrict your rights to an even smaller set than your default rights. Not only can you not redistribute the software (same as standard copyright) -- but they might, for example, try to prevent you from publishing benchmark results on the software. Standard copyright law does not prohibit you from benchmarking; that's where shrinkwrap attempts to come in with an additional (unsigned) "contract".
I'm gonna have to stop contributing to Open Source projects then.
No, really.
If I could be sued for duplicating functionality which is present in open source products, just because I've seen the code, then I'll just have to stop looking at GPL'd code so I don't put myself into that position... For example at work I implement graphics pipelines (image manipulation). Since this is strongly related to, say, what Gimp does, this means I can't even look at Gimp source code because it'll put me at risk for some idiotic copyright infringement claim.
Sorry, Gimp project. I can't afford a lawsuit. I suppose I won't be able to contribute. I'm glad I've never looked at the Gimp code!
It would appear that this case is very similar to the case where Lotus sued Paperback Software claiming that Paperback Software had breached their copyright by copying their UI. I think in that case there may have been more blatant copying but it seems to come down to a similar issue (it is hard to tell excatly how much was copied from reading the article). There is a bit about the case here and here.
Sorry, no. Section 0 of the GPL clearly states:
The GPL assumes that since you have legally obtained a copy of the program -- recorded onto a hard disk, CD-ROM, book, or other piece of your property -- that you already have the right to use it. In doing so, you're simply legally using a piece of your private property, an action to which copyright traditionally does not address.
Well-established, constitutionally upheld, internationally valid, largely uncontroversial Berne-Convention-class copyright only affects you when you copy, publicly perform, redistribute, etc. a covered work. It is only the experimental, alpha-test-quality, constitutionally untested, and controversial DMCA-class laws which attempt to extend copyright from the right to copy and publish to the right to enslave and deprive the user.
So it could be claimed that he learned how to write his code from looking at the GPL code.
You can't copyright an idea. United States copyright law, 17 USC 102, states that "[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work".
Will I retire or break 10K?
I'm sorry, but you really shouldn't base your future code contributions on a badly worded sentence on /..
Copyright law, the only thing which the GPL covers, applies only to a specific work. Example: A painting of a sunset can be copyrighted, whilst "paintings of sunsets in general" cannot. It is not illegal to study a particular sunset painting, then create a sunset painting of your own.
What this means for you and your (likely to remain ) hypothetical GIMP-contributing aspirations is that unless you actually copy the code that implements the features in the Gimp that you like, you have nothing to worry about. "A filter that does X" is not copyrightable. Even conceptual details such as "array of function pointers accessed with menu item id" aren't copyrightable. Thus the GPL doesn't apply to them.
Unless you either can't help but copy the GIMP code (or are too lazy to do otherwise) you have nothing to worry about.
The enemies of Democracy are
Following is the text snippet from Bowers v. Baystate covering the shrinkwrap license: ...
... are governed exclusively by this title." 17 U.S.C. 301(a) (2000). The First Circuit does not interpret this language to require preemption as long as "a state cause of action requires an extra element, beyond mere copying, preparation of derivative works, performance, distribution or display." Data Gen. Corp. v. Grumman Sys. Support Corp., 36 F.3d 1147, 1164, 32 USPQ2d 1385, 1397 (1st Cir.1994) (quotingGates Rubber Co. v. Bando Chem. Indus., 9 F.3d 823, 847, 28 USPQ2d 1503, 1520 (10th Cir.1993)); see also Computer Assoc. Int'l, Inc. v. Altai, Inc., 982 F.2d 693, 716 (2d Cir.1992) ("But if an 'extra element' is 'required instead of or in addition to the acts of reproduction, performance, distribution or display, in order to constitute a state-created cause of action, then the right does not lie "within the general scope of copyright," and there is no preemption.' ") (quoting 1 Nimmer on Copyright 1.01[B] at 1- 15). Nevertheless, "[n]ot every 'extra element' of a state law claim will establish a qualitative variance between the rights protected by federal copyright law and those protected by state law." Id.
... simply because a plaintiff must prove that copying was not only unauthorized but also commercially immoral." Id.
/foldoc.cgi?reverse+engineering (last visited Jul. 17, 2002). Thus, the contract in this case broadly prohibits any "reverse engineering" of the subject matter covered by the shrink-wrap agreement.
... remedies."), aff'd in relevant part, 36 F.3d 1147 (1st Cir.1994). In this case, the breach of contract damages arose from the same copying and included the same lost sales that form the basis for the copyright damages. The district court, therefore, did not abuse its discretion by omitting from the final damage award the duplicative copyright damages. Because this court affirms the district court's omission of the copyright damages, this court need not reach the merits of Mr. Bowers' copyright infringement claim. ...
A
[12] Baystate contends that the Copyright Act preempts the prohibition of reverse engineering embodied in Mr. Bowers' shrink-wrap license agreements. Swayed by this argument, the district court considered Mr. Bowers' contract and copyright claims coextensive. The district court instructed the jury that "reverse engineering violates the license agreement only if Baystate's product that resulted from reverse engineering infringes Bowers' copyright because it copies protectable expression." Mr. Bowers lodged a timely objection to this instruction. This court holds that, under First Circuit law, the Copyright Act does not preempt or narrow the scope of Mr. Bowers' contract claim.
[13][14] Courts respect freedom of contract and do not lightly set aside freely-entered agreements. Beacon Hill Civic Ass'n v. Ristorante Toscano, 422 Mass. 318, 662 N.E.2d 1015, 1017 (Mass.1996). Nevertheless, at times, federal regulation may preempt private contract. Cf. Nebbia v. New York, 291 U.S. 502, 523, 54 S.Ct. 505, 78 L.Ed. 940 (1934) ("Equally fundamental with the private right is [the right] of the public to regulate [the private right] in the common interest."). The Copyright Act provides that "all legal or equitable rights that are equivalent to any of the exclusive rights within the general scope of copyright
In Data General, Data General alleged that Grumman misappropriated its trade secret software. 36 F.3d at 1155. Grumman obtained that software from Data General's customers and former employees who were bound by confidentiality agreements to refrain from disclosing the software. Id. at 1154-55. In defense, Grumman argued that the Copyright Act preempted Data General's trade secret claim. Id. at 1158, 1165. The First Circuit held that the Copyright Act did not preempt the state law trade secret claim. Id. at 1165. Beyond mere copying, that state law claim required proof of a trade secret and breach of a duty of confidentiality. Id. These additional elements of proof, according to the First Circuit, made the trade secret claim qualitatively different from a copyright claim. Id. In contrast, the First Circuit noted that claims might be preempted whose extra elements are illusory, being "mere label[s] attached to the same odious business conduct." Id. at 1165 (quoting Mayer v. Josiah Wedgwood & Sons, Ltd., 601 F.Supp. 1523, 1535, 225 USPQ 776, 784 (S.D.N.Y.1985)). For example, the First Circuit observed that "a state law misappropriation claim will not escape preemption
The First Circuit has not addressed expressly whether the Copyright Act preempts a state law contract claim that restrains copying. This court perceives, however, that Data General's rationale would lead to a judgment that the Copyright Act does not preempt the state contract action in this case. Indeed, most courts to examine this issue have found that the Copyright Act does not preempt contractual constraints on copyrighted articles. See, e.g., ProCD, Inc. v. Zeidenberg, 86 F.3d 1447, 39 USPQ2d 1161 (7th Cir.1996) (holding that a shrink-wrap license was not preempted by federal copyright law); Wrench LLC v. Taco Bell Corp., 256 F.3d 446, 457, 59 USPQ2d 1434, 1441-42 (6th Cir.2001) (holding a state law contract claim not preempted by federal copyright law); Nat'l Car Rental Sys., Inc. v. Computer Assocs. Int'l, Inc., 991 F.2d 426, 433, 26 USPQ2d 1370, 1376 (8th Cir.1993); Taquino v. Teledyne Monarch Rubber, 893 F.2d 1488, 1501 (5th Cir.1990); Acorn Structures v. Swantz, 846 F.2d 923, 926, 6 USPQ2d 1810, 1812 (4th Cir.1988); but see Lipscher v. LRP Publs., Inc., 266 F.3d 1305, 1312, 60 USPQ2d 1468, 1473 (11th Cir.2001).
In ProCD, for example, the court found that the mutual assent and consideration required by a contract claim render that claim qualitatively different from copyright infringement. 86 F.3d at 1454. Consistent with Data General's reliance on a contract element, the court in ProCD reasoned: "A copyright is a right against the world. Contracts, by contrast, generally affect only their parties; strangers may do as they please, so contracts do not create 'exclusive rights.' " Id. This court believes that the First Circuit would follow the reasoning of ProCD and the majority of other courts to consider this issue. This court, therefore, holds that the Copyright Act does not preempt Mr. Bowers' contract claims.
[15][16] This court now considers the scope of Mr. Bowers' contract protection. Without objection to the choice of law, the district court applied Massachusetts contract law. Accordingly, contract terms receive "the sense and meaning of the words which the parties have used; and if clear and free from ambiguity the words are to be taken and understood in their natural, usual and ordinary sense." Farber v. Mutual Life Ins. Co., 250 Mass. 250, 253, 145 N.E. 535 (Mass.1924); see also Kelly v. Marx, 428 Mass. 877, 881, 705 N.E.2d 1114 (Mass.1999) ("The proper course is to enforce contracts according to their plain meaning and not to undertake to be wiser than the parties.") (quoting Guerin v. Stacy, 175 Mass. 595, 597, 56 N.E. 892 (1900) (Holmes, C.J.)).
In this case, the contract unambiguously prohibits "reverse engineering." That term means ordinarily "to study or analyze (a device, as a microchip for computers) in order to learn details of design, construction, and operation, perhaps to produce a copy or an improved version." Random House Unabridged Dictionary (1993); see also The Free On Line Dictionary of Computing (2001), at http://wombat.doc.ic.ac.uk/foldoc
[17] The record amply supports the jury's finding of a breach of that agreement. As discussed above, the district court erred in instructing the jury that copyright law limited the scope of Mr. Bowers' contract protection. Notwithstanding that error, this court may affirm the jury's breach of contract verdict if substantial record evidence would permit a reasonable jury to find in favor of Mr. Bowers based on a correct understanding of the law. Larch v. Mansfield Mun. Elec. Dep't, 272 F.3d 63, 69 (1st Cir.2001). The shrink-wrap agreements in this case are far broader than the protection afforded by copyright law. Even setting aside copyright violations, the record supports a finding of breach of the agreement between the parties. In view of the breadth of Mr. Bowers' contracts, this court perceives that substantial evidence supports the jury's breach of contract verdict relating to both the DOS and Windows versions of Draft-Pak.
The record indicates, for example, that Baystate scheduled two weeks in Draft- Pak's development schedule to analyze the Designer's Toolkit. Indeed, Robert Bean, Baystate's president and CEO, testified that Baystate generally analyzed competitor's products to duplicate their functionality.
The record also contains evidence of extensive and unusual similarities between Geodraft and the accused Draft-Pak--further evidence of reverse engineering. James Spencer, head of mechanical engineering and integration at the Space and Naval Warfare Systems Center, testified that he examined the relevant software programs to determine "the overall structure of the operating program" such as "how the operating programs actually executed the task of walking a user through creating a [GD & T] symbol." Mr. Spencer concluded: "In the process of taking the [ANSI Y14.5M] standard and breaking it down into its component parts to actually create a step-by-step process for a user using the software, both Geodraft and Draft-Pak [for DOS] use almost the identical process of breaking down that task into its individual pieces, and it's organized essentially identically." This evidence supports the jury's verdict of a contract breach based on reverse engineering.
Mr. Ford also testified that he had compared Geodraft and Draft-Pak. When asked to describe the Draft-Pak interface, Mr. Ford responded: "It looked like I was looking at my own program [i.e., Geodraft]." Both Mr. Spencer and Mr. Ford explained in detail similarities between Geodraft and the accused Draft- Pak. Those similarities included the interrelationships between program screens, the manner in which parameter selection causes program branching, and the manner in which the GD & T symbols are drawn.
Both witnesses also testified that those similarities extended beyond structure and design to include many idiosyncratic design choices and inadvertent design flaws. For example, both Geodraft and Draft-Pak offer "straightness tolerance" menu choices of "flat" and "cylindric," unusual in view of the use by ANSI Y14.5M of the terms "linear" and "circular," respectively. As another example, neither program requires the user to provide "angularity tolerance" secondary datum to create a feature control frame--a technical oversight that causes creation of an incomplete symbol. In sum, Mr. Spencer testified: "Based on my summary analysis of how the programs function, their errors from the standard and their similar nomenclatures reflecting nonstandard items, I would say that the Draft-Pak [for DOS] is a derivative copy of a Geodraft product."
Mr. Ford and others also demonstrated to the jury the operation of Geodraft and both the DOS and Windows versions of the accused Draft-Pak. Those software demonstrations undoubtedly conveyed information to the jury that the paper record on appeal cannot easily replicate. This court, therefore, is especially reluctant to substitute its judgment for that of the jury on the sufficiency and interpretation of that evidence. In any event, the record fully supports the jury's verdict that Baystate breached its contract with Mr. Bowers.
[18] Baystate does not contest the contract damages amount on appeal. Thus, this court sustains the district court's award of contract damages. Mr. Bowers, however, argues that the district court abused its discretion by dropping copyright damages from the combined damage award. To the contrary, this court perceives no abuse of discretion.
The shrink-wrap license agreement prohibited, inter alia, all reverse engineering of Mr. Bowers' software, protection encompassing but more extensive than copyright protection, which prohibits only certain copying. Mr. Bowers' copyright and contract claims both rest on Baystate's copying of Mr. Bowers' software. Following the district court's instructions, the jury considered and awarded damages on each separately. This was entirely appropriate. The law is clear that the jury may award separate damages for each claim, "leaving it to the judge to make appropriate adjustments to avoid double recovery." Britton v. Maloney, 196 F.3d 24, 32 (1st Cir.1999) (citing Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 451 n. 3, 113 S.Ct. 884, 122 L.Ed.2d 247 (1993)); see also Data Gen. Corp. v. Grumman Sys. Support Corp., 825 F.Supp. 340, 346 (D.Mass.1993) ("So long as a plaintiff is not twice compensated for a single injury, a judgment may be comprised of elements drawn from separate
The court followed the ProCD case, which directly conflicts with Vault v Quaid on the matter of whether state contract law shrinkwrap terms are "preempted" by federal copyright law.
Both ProCD and this court found that they were not because an "extra element" exists. Both courts got it wrong because they ignored 17 USC 117, which states that the "owner" of software can install it on a machine. In other words, installation is a first sale right by black letter law.
In particular, the negotiation that occurs during the contract of sale fully comprehends installation and use. The shrinkwrap offers nothing new to the purchaser, and therefore there is no "extra element".
Nimmer, the foremost authority on Copyright wrote a very long law review article specifically debunking ProCD's analysis on preemption. This court did a crappy analysis that wasn't even at the level of ProCD.
Actually, the way a proper clean-room project works is that one group of programmers analyzes the hell out of the target code and produce a product specification that is one heck of a lot more detailed than a "list of features". Basically, they produce a document that completely describes how the product will work.
This design specification is then turned over to a second group of coders who have never seen the target codebase. In a really good clean-room project this second group probably doesn't even know that they're working in a clean-room. They create a completely new codebase from scratch to fulfill the requirements of the design document. Since no one working on the new codebase have ever seen the codebase being reverse-engineered, there is NO issue of copyright infringement. The clean-room defense is even more airtight if the coders don't know they're working in a clean-room.
If both the target codebase and the end product are really well-written, I would not be surprised to see that large parts of the two codebases come out almost identical. Good coders will implement the same functionality in ways that are very similar, if not identical.
utter rubbish