In Intelligent Design, Christopher Buccafusco, Mark A. Lemley, and Jonathan S. Masur make a compelling case against the United States’ current full-cumulation approach to design protection, which allows designers to obtain protection for a qualifying design under copyright, trademark, and design patent law all at the same time and without any requirement of election.They also argue that design rights, including design patents, “have become too powerful” and suggest several “policy changes designed to bring design rights more in line with social welfare.”
But while Buccafusco, Lemley, and Masur express great concern about what design patents cover, they do not discuss how the current test for design patent infringement actually works.Under that test, which was set forth ten years ago by the U.S. Court of Appeals for the Federal Circuit in Egyptian Goddess, Inc. v. Swisa, Inc., design patents are given a much narrower scope than the cases discussed in Intelligent Design might seem to suggest. While this does not undermine the larger argument made by Buccafusco, Lemley, and Masur—namely, that we should not allow designers to gain the functional equivalent of a utility patent using the design patent system—it does affect the way we should evaluate some of their proposed policy solutions.
Under Egyptian Goddess, the presumptive scope of a design patent is narrow and may be further narrowed if there is close prior art. Therefore, at least two of the proposals in Intelligent Design—namely, functionality screening, invalidity based on “principal features,” and adding an independent invention defense—may not be either appropriate or necessary. Adopting such proposals may have the unintended consequence of expanding or unnecessarily confusing design patent doctrine, results clearly not intended by the authors of Intelligent Design. Instead, a policy lever not discussed by Buccafusco, Lemley, and Masur—namely, statutory subject matter—may be more helpful in addressing the very valid concerns raised in Intelligent Design.
This response will explain how courts analyze design patent infringement under the Federal Circuit’s Egyptian Goddess test. Then, it will discuss how the contemporary test for infringement might affect some of the solutions proposed in Intelligent Design. Finally, it discusses how the question of what constitutes proper statutory subject matter may be a more fruitful policy lever to explore in addressing these important issues of design patent law and policy.
The en banc Federal Circuit set forth the current test for design patent infringement in its 2008 decision in Egyptian Goddess, Inc. v. Swisa, Inc.Under Egyptian Goddess, a design patent is infringed when “an ordinary observer, familiar with the prior art, would be deceived into thinking that the accused design was the same as the patented design.” This is a test of visual similarity, not a test of trademark-like consumer confusion. And “the proper inquiry” is “whether the accused design has appropriated the claimed design as a whole.” As to how courts should conduct this inquiry, the Federal Circuit stated:
This suggests a two-part framework for analyzing claims of design patent infringement.
First, the claimed design and the accused design must be compared.If the designs don’t look the same, when considered in a vacuum, there is no infringement as a matter of law. We might think of this step as setting forth the “presumptive scope” of a design patent. Second, if the designs are “not plainly dissimilar,” the prior art may be used to narrow the presumptive scope of the patent.
The degree of visual similarity required by courts at Egyptian Goddess step one is quite high.For example, in Lin v. Belkin International, Inc., the plaintiff alleged that this USB cable infringed this design patent:
The court granted summary judgment of noninfringment.Based on its own visual review, the court found that the designs were so “sufficiently distinct and plainly dissimilar . . . that no reasonable jury could find the two designs to be substantially the same.”
In Wallace v. Ideavillage Products Corp., the plaintiff alleged that this spinning brush infringed this design patent:
The district court granted the defendant’s motion for summary judgment of noninfringement, concluding, upon visual review, that the designs were “sufficiently distinct” and that the plaintiff could not, “as a matter of law, prove that the designs appear substantially the same.”In an unpublished decision, the Federal Circuit affirmed.
In Performance Designed Products LLC v. Mad Catz, Inc., the plaintiff alleged that this video game controller infringed this design patent:
The judge dismissed the claim with prejudice under Federal Rule of Civil Procedure 12(b)(6), concluding that these designs were so “plainly dissimilar” under Egyptian Goddess that any attempt to amend the claim would be “futile.”
In all three of these cases, the courts decided that the accused designs did not look similar enough to infringe, as a matter of law, based on abstract visual comparison alone.They are good examples of how narrow the presumptive scope of a design patent is under Egyptian Goddess.
Under Egyptian Goddess,the presumptive scope of a design patent can be further narrowed using the prior art.This is done by visually comparing the claimed design, accused design, and any relevant prior art identified by the accused infringer. This comparison highlights differences that may not have been noticeable at step one.
This second step was intended to “cabin unduly broad assertions of design patent scope by ensuring that a design that merely embodies or is substantially similar to prior art designs is not found to infringe.”
Egyptian Goddessstep two also seems to have the practical effect of preventing the monopolization of most, if not virtually all, design elements that could be considered “functional” in the broad, trademark sense.If the particular shape or arrangement of a product feature or features “is essential to the use or purpose of the device or it affects the cost or quality of the device,” one would expect to find that shape or arrangement in the prior art. The facts in Richardson v. Stanley Works provide a helpful example of how this would work.
Here is Richardson’s claimed design:
Here is the accused design:
The only real similarity between these two designs is the arrangement of the various tool elements—namely, the fact that the hammer and the jaw/claw are on one end of the handle and the crow bar is on the other. This arrangement, however, was present in the prior art.Indeed, according to the district court:
According to the district court, this prior art “illustrate[d] the functional necessity of placing the hammer-head and jaw at one end of the handle and the crow-bar at the other end,” and thus required that general arrangement to be “factored out” as a part of claim construction.
But a straightforward application of Egyptian Goddess step two should produce the same result. If the arrangement of these design elements was in the prior art, then consideration of that prior art would serve to emphasize the visual differences between the claimed and the accused designs and prevent a finding of infringement based on that arrangement alone.This is particularly—though not exclusively—true where a particular product has evolved to the stage of establishing a type-form.
Buccafusco, Lemley, and Masur argue that “[w]hen prior art discloses a design that substantially anticipates the principal features of the claimant’s design . . . , the [U.S. Patent & Trademark Office] should deny the patent or the courts should invalidate it.”It’s not entirely clear what the authors mean by “substantially anticipates” or “principal features” but, whatever those phrases mean, they would seem to suggest a dramatic change to design patent law.
Under current design patent law, it is well-established that a design patent protects the visual appearance of the claimed design as a whole—not the appearance of its constituent elements.Because the same test applies to both infringement and anticipation, this principle mandates a holistic approach to invalidation under § 102.
Buccafusco, Lemley, and Masur don’t specifically say that their proposal would pertain to invalidation under § 102 in particular.But, if it did, it would seem to require either an abandonment of the infringement-anticipation symmetry principle or a dramatic increase in the scope of design patents. And the latter is clearly not what they want. But the fact remains that, if design patent infringement could be found whenever an accused product “substantially” duplicates “the principal features of the claimant’s design,” that would be a profound change from our current system—and not a change for the better. It would add confusion and uncertainty to a test that is, on the whole, working quite well. If the price of maintaining a very limited scope is an extremely high burden for anticipation, the tradeoff may well be worth it.
Buccafusco, Lemley, and Masur argue that the Supreme Court “should reintroduce an effective form of functionality screening to design patents.”They suggest that adopting the claim-construction approach approved in Richardson v. Stanley Works and rejecting the approach used in Sport Dimension, Inc. v. Coleman Co. “would be a good start.” However, it is not clear that making this change would actually change the result in most, if any, cases. Indeed, it appears that the Egyptian Goddess test, as currently understood and implemented, actually solves the problems that motivated this proposal—albeit in a different way.
Before diving into the reasons for this, it may be helpful to clarify a few points of vocabulary. Throughout their article, Buccafusco, Lemley, and Masur express concern about the protection of “functional aspects,” “functional elements,” and “functional features” of designs or products. It’s clear that, when they do so, they’re referring to a concept of “functionality” that is much broader than the sense that word “functionality” is used in the case law on design patent validity.
Buccafusco, Lemley, and Masur also appear to use the words “aspects,” “elements,” and “features” as synonyms for a wide range of product attributes. For the purposes of this discussion, however, it may be helpful to distinguish between these concepts. Therefore, this Part will use:
For an example, consider a teacup with a handle shaped like a dragon. The handle would be a feature of the teacup that has both functional and ornamental attributes because its shape is both decorative and allows a person to hold the cup.In a design patent that claims the shape of the entire cup, the shape of the handle is one element of that claimed design.
In Richardson, the Federal Circuit ruled that a pre-Egyptian Goddess line of claim-construction cases—specifically, the line of cases that talked about construing design patents to distinguish between “functional aspects” and “ornamental aspects” of a design—are still relevant even after Egyptian Goddess.But it’s not clear that this sort of “factoring out [of] the functional aspects” of a design actually changes the way any particular claim of infringement would (or should) be analyzed under the Egyptian Goddess framework. In other words, it’s not clear that doing (or not doing) this type of claim construction actually makes any difference to the ultimate question of infringement.
For example, even without any verbal claim construction, the infringement allegation in Richardson should never have survived Egyptian Goddess step one.The claimed and accused designs in Richardson are no more (and arguably much less) visually similar than designs that have been deemed “plainly dissimilar” without any express “factoring out” of any “functional aspects.” And even if Richardson’s infringement claim survived a straight-up application of Egyptian Goddess step one, a visual comparison with the prior art at Egyptian Goddess step two would have reached the same result—protection extending only to the way the claimed design looked, as opposed to how its commercial embodiment worked. With or without any claim-construction “functionality screening,” this case should have come out the exact same way, with a finding of noninfringement.
Similarly, it’s difficult to see how the claim construction method used in Coleman would have—or, at least, should have—affected the ultimate question of infringement. Coleman alleged that this Sport Dimension personal floatation device infringed this design patent:
This claim was implausible on its face.If the courts had ever reached the question of infringement, this claim should have failed at Egyptian Goddess step one.
But regardless of how a claim is (or is not) verbally construed, nothing in Coleman changes the fact that a claim of design patent infringement must still be evaluated under Egyptian Goddess. And the Egyptian Goddess test does not allow for a finding of infringement based on functional similarity alone.Therefore, Coleman did not actually give “the plaintiff the ability to block competitors who attempt to market flotation devices that [merely] perform the function in the same way.” And, as discussed above, Egyptian Goddess step two provides a check to ensure that duplication of one public-domain element alone will not be enough to establish a valid claim of infringement.
It’s also worth noting that the district court in Richardson did not actually eliminate any elements (i.e., visual portions) of the claimed design.The district court stated that the arrangement of the “four primary utilitarian” features—namely, “the handle, the hammer-head, the jaw, and the crow-bar”—was “dictated by the functional purpose of the tool and therefore . . . not protected.” But the court still construed the claim to cover the actual appearance of those features, specifically: “the standard shape of the hammer-head, the diamond-shaped flare of the crow-bar and the top of the jaw, the rounded neck, the orientation of the crowbar relative to the head of the tool, and the plain, undecorated handle.” Therefore, the phrase “factoring out” may not be the most helpful way to refer to what the court actually did.
Viewed in light of what the district court actually did (and, thus, what the Federal Circuit actually affirmed) Richardson is best viewed as reiterating, albeit in a long-winded way, the basic principle that design patents protect the way something looks—not the way it works.But of course, that’s always the test, regardless of how the claim is construed.
Indeed, it’s difficult to discern any significant difference in infringement outcomes between cases where the courts engage in Richardson-style claim construction and those that do not.Thus, as a practical matter, it appears that Richardson-style claim construction adds nothing but unnecessary time and cost to design patent litigation. If, as it appears, the Egyptian Goddess framework is already achieving the results sought by Buccafusco, Lemley, and Masur, the better course would be to stick with that framework and abandon the Richardson line of cases.
Buccafusco, Lemley, and Masur suggest that Congress could help “prevent abuse of design right . . . by incorporating an independent invention defense into design patent law.”In other words, they would require the plaintiff to prove copying in order to prevail on a claim for infringement, like we do in copyright law.
However, given the very high degree of visual similarity already required for design patent infringement under Egyptian Goddess, it’s not clear that this would actually narrow the scope of protection in a significant way—at least for most design patents. For design patents that claim all or most of a configuration design,the degree of required similarity is so high that the likelihood of infringement without actual copying is vanishingly low. This is especially true for designs that are actually “ornamental” in any ordinary sense of that word.
But under current law and practice, a design patent applicant does not have to claim all—or even most of—a particular configuration design. An applicant can claim basically any random fragment of a product’s shape as a separate “design.”For these types of design patents, there is a nonzero chance of inadvertent duplication. And, under the Federal Circuit’s current definition of “ornamentality,” these design patents can actually be drawn to functional features. This is a problem. But it’s not clear that creating an independent-creation defense would be the best way to solve this problem.
Adding an explicit requirement of copying might also have the unintended consequence of making design patent protection broader. Right now, a design patent is infringed when the accused infringer duplicates the entire claimed design.But adding an independent-creation defense would suggest that the “wrong” of design patent infringement is copying per se. That could lead courts to conclude that any and all copying should be considered infringement—even where the defendant has not copied the whole design or where the level of visual similarity is lower than the level currently required under Egyptian Goddess. Indeed, this is how courts have interpreted and applied the requirement of “copying” in copyright law. And courts would almost certainly look to copyright law in interpreting and applying any design patent independent-creation defense. Moreover, allowing the design patent owner to introduce evidence of copying may actually make juries morelikely to find infringement, regardless of the underlying degree of similarity. These issues should be considered in evaluating the merits of any potential independent creation defense.
It is true that we should not have a system where design patents “cover the design’s utility”or give “the plaintiff the ability to block competitors who attempt to market [products] that [merely] perform the [same] function in the same way.” But as noted above, there is one way that design patent applicants can successfully gain control over some functional features of a product—namely, by using partial claiming to limit the scope of their patent to just the functional part (or parts) of a design. This is neither what the system was designed to do nor is it a socially beneficial use of the design patent system.
One way to address this problem would be to revitalize the statutory requirement of ornamentality. This would be consistent with Buccafusco, Lemley, and Masur’s suggestion that we “try to shore up the doctrinal screens that prevent design-related rights from bleeding over into backdoor utility patents.”But there is another potential policy lever, not discussed in Intelligent Design, that deserves consideration—namely, statutory subject matter.
The Patent Act provides that “[w]hoever invents any new, original and ornamental design for an article of manufacture may obtain a patent therefor, subject to the conditions and requirements of this title.”Today, a design patent applicant can claim basically any random fragment of a product’s shape or surface ornamentation as a separate “design.” But that was not always the case. And, as noted above, a design patent protects the visual appearance of the claimed design as a whole. So the question of what can be claimed as a separate “whole” is really important. The use of the word “design” in the statute does not end the inquiry; after all, “[t]he word ‘design’ is mercurial; it has multiple meanings that have ebbed and flowed over time.” And it may be that, on the whole, the costs of allowing free-rein fragment claiming may exceed any perceived benefits of the system. Therefore, the question of what types of “designs” should be eligible for design patents is vitally important and one that deserves more scholarly attention.
Moreover, it’s not clear that the type (or types) of designs that are protected by the design patent system must be the same “designs” that are protected by the copyright system or the trademark system. To put it a different way, there’s no reason why a “design for an article of manufacture” has to be the same type of “design” that’s considered “the design of a useful article” for the purposes of copyright law or a “product design” for the purposes of trade dress law.Thus, statutory subject matter might also provide a way to address the problems inherent in our current full-cumulation regime. In any case, this is an area that deserves further research and scholarly discussion.
In Intelligent Design, Buccafusco, Lemley, and Masur make a significant contribution to the design law literature. This response has attempted to add to their contribution by demonstrating that the Federal Circuit’s Egyptian Goddess test provided a strong bulwark against those who would try to use the design patent system to get “backdoor utility patents.”It has also examined how the courts’ application of the Egyptian Goddess test might affect some of the policy proposals made in Intelligent Design. In the course of this evaluation, this response further identified another potentially fruitful policy lever for addressing some of the very valid concerns raised by Buccafusco, Lemley, and Masur—specifically, statutory subject matter. The question of what constitutes a protectable “design” underlies many of the issues discussed in Intelligent Design and throughout the design patent literature. The time has come to explore that issue head-on.
 See Christopher Buccafusco, Mark A. Lemley & Jonathan S. Masur, Intelligent Design, 68 Duke L.J. 75, 121–23 (2018); cf. Estelle Derclaye, Introduction, in The Copyright/Design Interface: Past, Present and Future 6 (Estelle Derclaye, ed. 2018) (defining a “full cumulation” system of copyright and design protection as one where both rights “can subsist if the protection requirements are fulfilled and the two laws apply in tandem whether it raises regime clashes and/or overprotection, or not. In other words, there are no mechanisms in the legislation to deal with these problems.”). Other scholars have also noted the problem of overlap in this area. E.g., Peter Lee & Madhavi Sunder, The Law of Look and Feel, 90 S. Cal. L. Rev. 529, 532 (2017) (noting that “the cumulative effect of overlapping exclusive rights is likely to lead to overprotection”).
 Buccafusco, Lemley & Masur, supra note 1, at 124.
 They do cite the case that set forth that test and note that it “do[es]not enable claimants to protect downstream ‘derivatives’ of their designs.” Id. at 103 n.141 (citing Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc)). But they don’t discuss how courts have applied the Egyptian Goddess test.
 See Egyptian Goddess, 543 F.3d at 674. Although, “[i]n Egyptian Goddess, the Federal Circuit incorporated some of the language from—and suggested that it was adopting—the test announced by the Supreme Court in Gorham Manufacturing Co. v. White,” the Goddess test is actually different. See Sarah Burstein, The “Article of Manufacture” in 1887, 32 Berkeley Tech. L.J. 1, 11 n.64 (2017) (discussing one of those differences) (referring to Gorham Mfg. Co. v. White, 81 U.S. (14 Wall.) 511 (1871)) [hereinafter Burstein, The “Article of Manufacture” in 1887]. Indeed, shortly after the decision came out, leading practitioners noted that the Goddess test was different. See Christopher V. Carani, The New “Extra-Ordinary” Observer Test for Design Patent Infringement—On A Crash Course with the Supreme Court’s Precedent in Gorham v. White, 8 J. Marshall Rev. Intell. Prop. L. 354, 380 (2009) (“The Federal Circuit’s holdings in Arminak and Egyptian Goddess improperly replace the ordinary observer test with an ‘extra-ordinary observer’ test.”); James Juo, Egyptian Goddess: Rebooting Design Patents and Resurrecting Whitman Saddle, 18 Fed. Circuit B.J. 429, 430 (2009) (characterizing Goddess as “reinterpret[ing] Gorham v. White”).
 Not all of their proposals would be affected by this analysis. For example, Buccafusco, Lemley, and Masur “suggest that the PTO increase application and maintenance fees for design patents and use the money for improved examination.” Buccafusco, Lemley & Masur, supra note 1, at 81 (footnote omitted). For the record, I agree. See Sarah Burstein, Costly Designs, 77 Ohio St. L.J. 107, 156–57 (2016) (arguing that the USPTO’s current costly screen for design patents “could be enhanced by making broader claims even more expensive” and providing details about how that might be accomplished); id. at 156 n.282 (suggesting that “the PTO could also impose maintenance fees for design patents”); id. at 157 (“Implementing any—or all—of these increases could also help strengthen the PTO’s substantive screen by raising funds to purchase, develop or commission the development of better image-search technology for design patent examiners.”).
 On the concept of “policy levers” in IP law, see, e.g., Dan L. Burk & Mark A. Lemley, Policy Levers in Patent Law, 89 Va. L. Rev. 1575, 1579 (2003); Pamela Samuelson & Suzanne Scotchmer, The Law and Economics of Reverse Engineering, 111 Yale L.J. 1575, 1581 (2002).
 See Egyptian Goddess, 543 F.3d at 678. The Federal Circuit has had exclusive jurisdiction over patent appeals, including design patent appeals, since 1982. See Federal Courts Improvement Act of 1982, 28 U.S.C. § 1295(a)(1) (1982) (providing the Federal Circuit with jurisdiction over all civil actions “arising under . . . any Acts of Congress relating to patents”).
 Egyptian Goddess, 543 F.3d at 672; see also Burstein, The “Article of Manufacture” in 1887, supra note 4, at 11 (“In this context, ‘the patented design’ means ‘the claimed design.’ Therefore, in analyzing infringement, the fact finder must compare the claimed portion of the design—i.e., whatever is shown in solid lines in the patent drawings—to the corresponding portion of the accused design.” (citing Egyptian Goddess, 543 F.3d at 662; Hutzler Mfg. Co. v. Bradshaw Int’l, Inc., No. 1:11-cv-07211, 2012 WL 3031150, at *9–10 (S.D.N.Y. July 24, 2012) (footnotes omitted)). “If a patentee is able to show that there is no substantial difference between the claimed design and the purported commercial embodiment, a comparison between that embodiment and the accused design is permissible” but such a comparison is never required. See High Point Design LLC v. Buyer’s Direct, Inc., 621 F. App’x 632, 642 (Fed. Cir. 2015) (citing L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117, 1125–26 (Fed. Cir. 1993))).
 See, e.g., Egyptian Goddess, 543 F.3d at 678 (noting that the designs must “appear” to be the same to support a finding of infringement); id. at 678 (emphasizing that the Gorham test focused on similarity of appearance). See also Unette Corp. v. Unit Pack Co., Inc., 785 F.2d 1026, 1029 (Fed. Cir. 1986) (“Likelihood of confusion as to the source of the goods is not a necessary or appropriate factor for determining infringement of a design patent.”). This is consistent with the Supreme Court’s decision in Gorham v. White. See Sarah Burstein, The Patented Design, 83 Tenn. L. Rev. 161, 177 (2015) (“[W]hen read in context, it is clear that Gorham’s test is one of visual similarity, not a test of actual deception or trademark-like likelihood of confusion.” (citing Gorham Mfg. Co. v. White, 81 U.S. (14 Wall.) 511 (1871))).
 Egyptian Goddess, 543 F.3d at 677 (citing Amini Innovation Corp. v. Anthony Cal., Inc., 439 F.3d 1365, 1370–71 (Fed. Cir. 2006)); KeyStone Retaining Wall Sys., Inc. v. Westrock, Inc., 997 F.2d 1444, 1450 (Fed. Cir. 1993); Braun Inc. v. Dynamics Corp. of Am., 975 F.2d 815, 820 (Fed. Cir. 1992). Thus, there is no “fragmented literal similarity” in design patent law. Cf. Olufunmilayo B. Arewa, From J.C. Bach to Hip Hop: Musical Borrowing, Copyright and Cultural Context, 84 N.C. L. Rev. 547, 570 (2006) (noting that, in copyright law, “[f]ragmented literal similarity is the test of substantial similarity in cases where only a portion of a work is copied, without copying of the work’s overall essence or structure”).
 Egyptian Goddess, 543 F.3d at 678.
 See Keurig, Inc. v. JBR, Inc., No. 1:11-cv-11941, 2013 WL 2304171, at *5 (D. Mass. May 24, 2013), aff’d, 558 F. App’x 1009 (Fed. Cir. 2014) (“Courts have interpreted this language as establishing ‘two levels to the infringement analysis: a level-one or “threshold” analysis to determine if comparison to the prior art is even necessary, and a second level analysis that accounts for prior art in less obvious cases.’” (quoting Wing Shing Prods. (BVI) Co. v. Sunbeam Prods., 665 F. Supp. 2d 357, 362 (S.D.N.Y.2009) and citing Great Neck Saw Mfrs., Inc. v. Star Asia U.S.A., LLC, 727 F. Supp. 2d 1038, 1052 (W.D. Wash. 2010))).
 See Egyptian Goddess, 543 F.3d at 678 (holding that “the comparison of the designs . . . must be conducted as part of the ordinary observer test”).
 See id. (“In some instances, the claimed design and the accused design will be sufficiently distinct that it will be clear without more that the patentee has not met its burden of proving the two designs would appear ‘substantially the same’ to the ordinary observer . . . .”).
 This is also the maximum scope of a design patent. See Ethicon Endo-Surgery, Inc. v. Covidien, Inc., 796 F.3d 1312, 1337 (Fed. Cir. 2015) (rejecting Ethicon’s attempt to use the prior art to expand the presumptive scope of its claim).
 See Egyptian Goddess, 543 F.3d at 678 (stating in those instances that “resolution of the question whether the ordinary observer would consider the two designs to be substantially the same will benefit from a comparison of the claimed and accused designs with the prior art . . .”). See also Ethicon, 796 F.3d at 1337 (“[C]omparing the claimed and accused designs with the prior art is beneficial only when the claimed and accused designs are not plainly dissimilar.” (citing Egyptian Goddess, 543 F.3d at 678)).
 hile a full discussion of how lower courts have understood and applied Egyptian Goddess is beyond the scope of this response, these examples representative of the author’s research to date.
 Lin v. Belkin Int’l, Inc., No. 8:16-cv-00628, 2017 WL 2903261, at *6 (C.D. Cal. May 12, 2017) (showing the accused product and the claimed design).
 See Wallace v. Ideavillage Prod. Corp., 640 F. App’x 970, 972 (Fed. Cir. 2016) (showing images of the accused design); Body Washing Brush, U.S. Patent No. D485,990 fig. 1.
 Wallace v. Ideavillage Prods. Corp., No. 2:06-cv-05673, 2014 WL 4637216, at *4 (D.N.J. Sept. 15, 2014).
 Wallace, 640 F. App’x at 975.
 See Performance Designed Prods. LLC v. Mad Catz, Inc., No. 3:16-cv-00629, 2016 WL 3552063 (S.D. Cal. June 29, 2016) (discussing plaintiff’s infringement claim); Asymmetrical Video Game Controller, U.S. Patent No. D624,078 fig. 1 (showing the front view of the claimed design); Memorandum of Points and Authorities in Support of Mad Catz, Inc.’s Motion to Dismiss Complaint for Patent Infringement Pursuant to Fed. R. Civ. P. 12(b)(6) at 4, Performance Designed Prods. LLC v. Mad Catz, Inc., No. 3:16-cv-00629 (S.D. Cal. May 4, 2016), ECF 14-1 (showing the front view of the accused product).
 Performance Designed Prods., 2016 WL 3552063, at *6–7.
 See id.; Wallace, 2014 WL 4637216, at *4 (“Indeed, a comparison supports a finding that these two designs are sufficiently distinct and Ms. Wallace cannot, as a matter of law, prove that the designs appear substantially the same.”); Lin v. Belkin Int’l, Inc., No. 8:16-cv-00628, 2017 WL 2903261, at *6 (C.D. Cal. May 12, 2017) (“Upon comparing the design of Belkin’s product with the Claimed Design, the Court finds the designs sufficiently distinct and plainly dissimilar such that no reasonable jury could find the two designs to be substantially the same.”). Notably, none of these courts expressly “factored out” any “functional aspects” prior to making these determinations. Cf. Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1293 (Fed. Cir. 2010) (“The district court here properly factored out the functional aspects of Richardson’s design as part of its claim construction”).
 That’s not to say this test has always been perfectly applied or that there have been no outliers. For example, Apple’s graphical user interface design patent claim should have never made it to a jury. See Samsung’s Submission in Response to the Court’s August 2, 2012 Order at Exhibit A, Apple Inc. v. Samsung Elecs. Co., Ltd., No. 5:11-cv-01846 (N.D. Cal. Aug. 3, 2012), ECF 1565-1 (showing the two designs side-by-side). But on the whole, the Egyptian Goddess framework has proved quite effective in appropriately constraining design patent scope.
 See supra note 16 and accompanying text.
 Although the patent owner retains the ultimate burden of persuasion on the issue of infringement, the accused infringer bears the burden of identifying any relevant prior art. See Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665, 678 (Fed. Cir. 2008) (en banc).
 See Egyptian Goddess, 543 F.3d at 678 (noting that “[w]here there are many examples of similar prior art designs, . . . differences between the claimed and accused designs that might not be noticeable in the abstract can become significant”).
 See id.
 The author is not aware of any cases where anyone has been held liable for infringement post-Goddess for merely practicing the prior art. And if there were a case where a piece of prior art looked the same as the accused design, one would expect that case to settle quickly in light of the infringement-anticipation symmetry principle. See Int’l Seaway Trading Corp. v. Walgreens Corp., 589 F.3d 1233, 1239 (Fed. Cir. 2009) (noting that the maxim “that which infringes, if later, would anticipate, if earlier” applies to design patents as well as utility patents (quoting Peters v. Active Mfg. Co., 129 U.S. 530, 537 (1889))).
 See generally Sarah Burstein, Commentary: Faux Amis in Design Law, 105 Trademark Rep. 1455 (2015) (demonstrating “that ‘functional’ and ‘ornamental’ do not mean the same things in design patent law that they do in trademark law”). And by “design elements,” I mean sub-parts of a claimed design. The benefits of the Egyptian Goddess framework are substantially lessened when it comes to fragment claiming. See infra Part III.
 See TrafFix Devices, Inc. v. Mktg. Displays, Inc., 532 U.S. 23, 33 (2001) (noting one way that a product-design feature can be “functional” under trademark law).
 This was not how Stanley Works actually framed its arguments to the court. See Motion for Partial Summary Judgment of Non-Infringement and Lack of Willfulness, Richardson v. Stanley Works, Inc., No. 2:08-cv-01040 (D. Ariz. Nov. 5, 2008), ECF 30 [hereinafter “Stanley Works Br.”]. So the discussion that follows does not reflect the precise approach used by either the district court or the Federal Circuit to analyze the case. See generally Richardson v. Stanley Works, Inc., 597 F.3d 1288 (Fed. Cir. 2010); Richardson v. Stanley Works, Inc., 610 F. Supp. 2d 1046 (D. Ariz. 2009). But the underlying facts are still instructive.
 Richardson, 597 F.3d at 1291.
 Id. at 1292. Normally, the factfinder should compare the claimed design to the actual accused product. But in Richardson, “Stanley successfully applied for and obtained U.S. Patent D562,101 . . . on the basic Fubar design. All five versions of the tool are built around that same basic Fubar design” and there seemed to be no serious dispute that this was a fair comparator. See id. at 1291.
 See Richardson, 610 F. Supp. 2d at 1050 (discussing the functional purpose of that arrangement).
 Id. (internal citation omitted).
 See id. In its discussion of its claim construction, the district court used the word “configuration” to refer to the arrangement of these parts. Id. (“The ‘167 patent does not protect the configuration of the handle, hammer-head, jaw, and crow-bar utilized in the Stepclaw.”). However, the word “configuration” is a term of art in design patent law that is most commonly used to refer to the shape of a physical object or part thereof. See, e.g., U.S. Dep’t of Commerce, Patent & Trademark Office, Manual of Patent Examining Procedure § 1502.01 (9th ed., rev. 08.2017, Jan. 2018) [hereinafter MPEP]; In re Schnell, 46 F.2d 203, 209 (C.C.P.A. 1931); Gorham Mfg. Co. v. White, 81 U.S. 511, 525 (1871). The district court’s atypical use of the word “configuration” in Richardson appears to have led some people to interpret it as requiring that the shape of any functional features must be eliminated from the scope of the design. See Brief of Amicus Curiae American Intellectual Property Law Association in Support of the Petition for Rehearing En Banc at 2, No. Richardson v. Stanley Works, Inc., No. 09-01354 (Fed. Cir. Apr. 22, 2010). But it is clear that when the court talked about “factoring out” anything, the court was referring to the arrangement—not the actual shapes—of these features. See Richardson, 610 F. Supp. 2d at 1050; see also infra notes 70–72 and accompanying text.
 This assumes, of course, that such further emphasis is needed. The visual differences between the claimed and accused designs are so stark that it’s not even clear a court would—or should—get to step two. See supra notes 36–37 and accompanying text. But if they did, step two should prevent the patentee from effectively gaining control over how its design works—as opposed to how it looks.
 See Anne J. Banks, What Is Design?: An Overview of Design in Context from Prehistory to 2000 A.D. 55 (2004) (“In the redesign of products, the invented form is gradually improved until the standard, or type form, is reached.”). This seems to have been the case in Richardson. See Richardson, 610 F. Supp. 2d at 1050 (“The number of other patented designs that use this [arrangement] and the absence of alternative designs strongly suggest that this [arrangement] is the best [arrangement] . . . .”). Of course, for truly pioneering products—i.e., ones that are the first in, and themselves create, a whole new product category—there may be little to no close prior art. This might cause some to be concerned that, under a straight-up application of the Egyptian Goddess test, the first producer could use design patent protection to lock up the best design. See Buccafusco, Lemley & Masur, supra note 1, at 102 (expressing concern that a design patentee could obtain protection for a configuration design in which each useful element is shaped in “the best way to achieve [its respective] function”). But in practice, it seems highly unlikely that any pioneering-product designer would actually hit on the ideal (or type-form) on the first try. And even if this risk were deemed high enough to shape legal rules around it, the better approach may be to work on reviving the statutory requirement of “ornamentality.”
 Buccafusco, Lemley & Masur, supra note 1, at 124–25.
 See supra note 10 and accompanying text. See also, e.g., Crocs, Inc. v. Int’l Trade Comm’n, 598 F.3d 1294, 1303 (Fed. Cir. 2010) (“The ordinary observer test applies to the patented design in its entirety, as it is claimed.” (citing Braun, Inc. v. Dynamics Corp. of Am., 975 F.2d 815, 820 (Fed. Cir. 1992))). So, for example, no matter how often Samsung suggested that Apple’s design patents covered “rounded corners,” that was not all they covered—a point not lost on the jury. Amended Verdict Form at 6, Apple Inc. v. Samsung Elecs. Co., No. 5:11-cv-01846 (N.D. Cal. Aug. 24, 2012), ECF 1931 [hereinafter “Verdict”] (finding, for example, that the Samsung Galaxy Ace infringed U.S. Patent No. D618,677 while the Samsung Fascinate did not, even though both phones has rounded corners). That is not to say that it would have been impossible for Apple to claim just the rounded corners; under current law, the claimed design can be something less than the design of an article as a whole. See infra note 82 and accompanying text. But it is to say that a claim for, for example, a complete phone screen would not also separately protect the shape of the corners of that screen.
 See Int’l Seaway Trading Corp. v. Walgreens Corp., 589 F.3d 1233, 1239 (Fed. Cir. 2009) (“[I]t has been well established for over a century that the same test must be used for both infringement and anticipation.”)
 35 U.S.C. § 102 (2018).
 This proposal may be related to their suggestions for changes to how courts evaluate obviousness under 35 U.S.C. § 103. For example, they suggest that the fact that a designer would know it was mechanically possible to “combine elements from prior designs” means that it should be considered visually obvious to do so. See Buccafusco, Lemley & Masur, supra, at 125. For a contrary view, see Sarah Burstein, Visual Invention, 16 Lewis & Clark L. Rev. 169, 200 (2012).
 It’s true that it’s extremely difficult to invalidate design patents under current Federal Circuit law. See Sarah Burstein, Is Design Patent Examination Too Lax?, 33 Berkeley Tech. L.J. (forthcoming) (manuscript at 7) (on file with Duke Law Journal Online). But, unless we’re going to abandon the symmetry principle, it may be worth keeping the rule to constrain design patent scope. See id. at I.
 Buccafusco, Lemley & Masur, supra note 1, at 124–25.
 Indeed, it’s not clear that the one factual scenario Buccafusco, Lemley and Masur used as an example really represents a problem with the current law or with one litigant’s litigation strategy. They argue that the “prior art discloses a design that substantially anticipates the principal features of . . . Apple’s patent on a rectangle with rounded corners.” Id. Presumably, they were referring to U.S. Patent No. D618,677, one of the design patents at issue in Apple v. Samsung. See Verdict, supra note 44, at 6. But they don’t specify which piece(s) of prior art they thought should be invalidating. And some (potentially important) pieces of prior art were excluded in Apple v. Samsung because Samsung failed to properly disclose them in discovery. See, e.g., Order Denying Samsung’s Motions for Relief from Magistrate Judge Orders at 7, Apple Inc. v. Samsung Elecs. Co., Ltd., No. 5:11-cv-01846 (N.D. Cal. Aug. 2, 2012), ECF 1545. Samsung did, however, share some of the excluded art with the press. See generally Declaration of John B. Quinn Submitted at the Request of the Court Regarding Samsung’s Disclosure of Public Information in Response to Press Inquiries, Apple Inc. v. Samsung Elecs. Co., Ltd., No. 5:11-cv-01846 (N.D. Cal. Aug. 1, 2012), ECF 1533. So, it’s possible that Buccafusco, Lemley, and Masur are aware of close pieces of prior art that were not part of the record at trial. Of course, none of this changes the fact that design patents are still incredibly difficult to invalidate under the current test. But if that prior art were in the record, it could have narrowed the presumptive scope of the claims. See supra Part B.
 Buccafusco, Lemley & Masur, supra note 1, at 126.
 Id. (citing Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1293–94 (Fed. Cir. 2010); Sport Dimension, Inc. v. Coleman Co., 820 F.3d 1316, 1320–23 (Fed. Cir. 2016)). See also id. at 138 n.319 (“As a matter of functionality screening, the opinions in Richardson and Apple seem to strike the balance correctly.”).
 Indeed, the author is hard-pressed to think of an example of a case where Richardson-type claim construction has (or should have) changed the result, compared to a straightforward application of the Egyptian Goddess test.
 See generally supra Part B.
 See, e.g., Buccafusco, Lemley & Masur, supra note 1, at 101–02. Not that there’s anything wrong with that. But it’s still worth noting that the Federal Circuit uses the word “functionality” in a narrower sense in the context of design patent validity. See Burstein, Faux Amis, supra note 33, at 1456–57. In recent years, it appears that courts have been using a more expansive notion of “functionality” in discussions of design patent infringement and claim construction. See Christopher V. Carani, Design Patent Functionality: A Sensible Solution, Landslide Mag., November-December 2014, at 20, 23 (distinguishing between what he calls “statutory functionality” and “claim-construction functionality”). See also, e.g., Ethicon Endo-Surgery, Inc. v. Covidien, Inc., 796 F.3d 1312, 1334 (Fed. Cir. 2015) (using different standards of functionality for validity and for claim construction).
 See, e.g., Buccafusco, Lemley & Masur, supra note 1, at 78 (“This grant of IP rights over partly functional elements is important. When IP law grants protection to useful or functional features of a product rather than merely aesthetic or ornamental ones, it can convey substantial market power. . . . If IP law allows claimants to gain some protection for functional aspects of a design, it should not do so easily or cheaply.”) (emphasis added); id. at 115–16 (again seeming to use “aspects,” “features,” and “elements” as synonyms in a discussion of Sport Dimension v. Coleman).
 These definitions do not necessarily match how these terms are used in all of the relevant case law. See, e.g., Amini Innovation Corp. v. Anthony Cal., Inc., 439 F.3d 1365, 1372 (Fed. Cir. 2006) (apparently using “aspects” and “elements” as synonyms). And there may be better ways to define them that might better elucidate this line of case law. This is just a first cut to facilitate the discussion here.
 By “ornamental,” I mean “ornamental” in any reasonable English sense of that word, not “ornamental” in the sense that term is currently used by the Federal Circuit. See infra note 82 and accompanying text.
 See Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1294 (Fed. Cir. 2010) (“[W]hen [a] design . . . contains ornamental aspects, it is entitled to a design patent whose scope is limited to those aspects alone and does not extend to any functional elements of the claimed article.”).
 See Richardson, 597 F.3d at 1293 (describing what the district court did in claim construction as “factor[ing] out the functional aspects of Richardson’s design”).
 Aside from perhaps making judges feel more comfortable with their own visual assessments, at least. But it appears that judicial comfort with this visual analysis may be increasing as more and more case law accumulates in this area.
 See supra notes 36–37 and accompanying illustrations.
 See supra Part A (discussing three such cases). See also Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1293 (Fed. Cir. 2010) (“The district court here properly factored out the functional aspects of Richardson’s design as part of its claim construction.”). Therefore, no verbal claim construction was necessary to the conclusion that the Richardson and Stanley Works designs are “plainly dissimilar.” But that’s how Stanley Works framed its argument. See Stanley Works Br., supra note 35. So here we are.
 See supra notes 38–41 and accompanying text. Again, Stanley Works did not frame the issue this way. See Stanley Works Br., supra note 35. See also Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1293 (Fed. Cir. 2010) (“The district court here properly factored out the functional aspects of Richardson’s design as part of its claim construction.”).
 See generally Richardson, 597 F.3d at 1296 (affirming the finding of noninfringement).
 See Sport Dimension, Inc. v. Coleman Co., 820 F.3d 1316, 1318 (Fed. Cir. 2016).
 Coleman appears to have been laboring under what I’ve referred to as “the concept fallacy” in design patent litigation—i.e., the mistaken belief that design patents protect general concepts, as opposed to just the claimed designs. See Sarah Burstein, Design Law, Tumblr (July 2, 2014), http://design-law.tumblr.com/post/90571053836/does-this-reflector-for-use-in-golf-infringe [https://perma.cc/8P7Y-KJWS] (describing the so called “concept fallacy” in design patent litigation). At this point, one may ask why rational litigants would spend time and money on claim construction on cases where it doesn’t (or at least shouldn’t) actually matter. There are at least three plausible reasons for this. First, at least for early cases like Richardson, the full implications of Egyptian Goddess may not have been clear at the time of filing. See Stanley Works Br., supra note 35 (filed about a month after Egyptian Goddess was decided). But even now, some litigants (and their attorneys) might not really understand how the Egyptian Goddess test works. Design patent litigation was not popular until recently and, even now, there aren’t any U.S. attorneys (at least as far as the author is aware) that dedicate all—or even most—of their practice to design patent litigation. Finally, it is possible that well-financed, sophisticated litigants view claim construction as an essentially risk-free second bite at the apple; even if it doesn’t go the way they want, they can still make their arguments under the Egyptian Goddess framework.
 In reality, the courts never got to the issue of infringement. This case ended up in the Federal Circuit after Coleman asked the court to file a stipulated judgment of noninfringement so that Coleman would be able to appeal the claim-construction ruling. See The Coleman Company Inc.’s Notice of Motion [Corrected] and Motion for Entry of Judgment and Designation as Final Under Fed. R. Civ. P. 54(b), Sport Dimension, Inc. v. The Coleman Company, Inc., No. 2:14-cv-00438 (C.D. Cal. Feb. 23, 2015), ECF 110. See also Sport Dimension, Inc. v. Coleman Co., Inc., No. 2:14-cv-00438, 2015 WL 12732711, at *1 (C.D. Cal. Mar. 18, 2015) (granting that motion). Thus, the question of infringement was not before the Federal Circuit. See Sport Dimension, Inc. v. Coleman Co., 820 F.3d 1316, 1318 (Fed. Cir. 2016). The Federal Circuit remanded the case but parties settled before any substantive decision on infringement was made. See Order on Dismissal with Prejudice, Sport Dimension, Inc. v. The Coleman Company, Inc., No. 2:14-cv-00438 (C.D. Cal. Oct. 14, 2016), ECF 169.
 See supra Part B. Indeed, as the Federal Circuit stressed, under the Gorham “ordinary observer” approach, “[i]dentity of appearance . . . or ‘sameness of effect upon the eye, is the main test of substantial identity of design.’” Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665, 670 (Fed. Cir. 2008) (quoting Gorham Mfg. Co. v. White, 81 U.S. 511, 527 (1871)) (emphasis added). Thus, mere functional similarity would not be enough.
 See Buccafusco, Lemley & Masur, supra note 1, at 80. Of course, any patent owner may allege facially-implausible claims and block competition through intimidation. But that can happen (and is likely to continue happening) regardless of whether courts require any type of express functionality screening as a matter of claim construction.
 See Richardson v. Stanley Works, Inc., 610 F. Supp. 2d 1046, 1050 (D. Ariz. 2009)). As noted above, see supra note 40, the district judge’s use of the word “configuration” instead of “arrangement” in this discussion seems to have caused some confusion.
 Richardson, 610 F. Supp. 2d at 1050.
 This is not to fault the district court judge, who took this language from a Federal Circuit decision. See Richardson, 610 F. Supp. 2d at 1051 (“The trial court is correct to factor out the functional aspects of various design elements, but that discounting of functional elements must not convert the overall infringement test to an element-by-element comparison” (quoting Amini Innovation Corp. v. Anthony Cal., Inc., 439 F.3d 1365, 1372 (Fed. Cir. 2006))). But if the Federal Circuit retains this filtration requirement, it may want to pick a different phrase to describe what it wants lower courts to do.
 See Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1294 (Fed. Cir. 2010) (“[W]hen the design also contains ornamental aspects, it is entitled to a design patent whose scope is limited to those aspects alone and does not extend to any functional elements of the claimed article.”) (citing L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117, 1123 (Fed. Cir. 1993). Cf. MPEP, supra note 39, § 1503.03(III) (“In general terms, a ‘utility patent’ protects the way an article is used and works (35 U.S.C. 101), while a ‘design patent’ protects the way an article looks (35 U.S.C. 171).”).
 Compare, e.g., the cases discussed supra Part A, with the cases discussed in the recent decision of Dyson, Inc. v. SharkNinja Operating LLC, No. 14-CV-779, 2018 WL 1906105, at *11–14 (N.D. Ill. Mar. 29, 2018). Although the Dyson court engaged in verbal claim construction to “discount” the “functional aspects” of the claimed designs, see id. at *5, it really didn’t need to. The illustrations prepared by the plaintiff’s counsel clearly showed that the claimed and accused designs were plainly dissimilar under Egyptian Goddess step one. See id. at *14 (“Perhaps the best illustration of the substantially different visual appearance of the handheld portion of the Dyson and Shark stick vacs is the overlay image submitted by Dyson’s infringement expert, Mauro, showing the Dyson on top of the Shark Rocket.”); id. at *17 (providing more comparative images).
 See Sarah Burstein, Design Patent Claim Construction: More from the Federal Circuit, Patently-O (Apr. 25, 2016), http://patentlyo.com/patent/2016/04/construction-federal-circuit.html [https://perma.cc/H3PS-RU3G] (“[I]t’s not clear what, exactly, the Federal Circuit hopes to gain by making district courts go through this [claim-construction] rigmarole. In this case, the accused product looks nothing like the claimed design . . . .”); Carani, supra note 55, at 27 (arguing that the “factoring out” approach has proven to be both unworkable and unnecessary).
 For more thoughts on the issue of functionality screening in claim construction, see Sarah Burstein, Apple, Samsung & Design Patent Claim Construction, Patently-O (Jan. 4, 2016), http://patentlyo.com/patent/2016/01/samsung-patent-construction.html [https://perma.cc/R7KH-F4EJ].
 Buccafusco, Lemley & Masur, supra note 1, at 81.
 See Buccafusco, Lemley & Masur, supra note 1, at 133 (“[W]e might consider introducing an independent invention defense to design patents like the one we have in copyright law.” (citing Robert E. Suggs, A Functional Approach to Copyright Policy, 83 U. Cin. L. Rev. 1293, 1302 (2015); Rebecca Tushnet, The Eye Alone Is the Judge: Images and Design Patents, 19 J. Intell. Prop. L. 409, 423 (2012))).
 For example, the design patents at issue in Richardson and Coleman.
 These types of designs are only a subset of the universe of designs that are currently considered “ornamental” by the Federal Circuit. See Burstein, Faux Amis, supra note 33, at 1457 (noting that, according to the Federal Circuit, “‘ornamental’ effectively means ‘not dictated by function’ and ‘not hidden during the entire lifetime of the completed product’”); id. at 1456–57 (“And, according to the Federal Circuit, ‘a design is “not dictated by function alone” when there are alternative designs or configurations available for the article of manufacture.’ In this analysis, ‘alternative design must simply provide “the same or similar functional capabilities.”’ So in design patent law, ‘functional’ essentially means ‘the only configuration that is fit for a particular purpose.’”) (footnotes omitted).
 That means that, even though Egyptian Goddess gives narrow scope to design patent claims, sophisticated applicants can still game the system to obtain broader coverage. This practice can be traced back to the U.S. Court of Customs and Patent Appeals’ decision in In re Zahn, 617 F.2d 261 (C.C.P.A. 1980). The Federal Circuit has adopted the holdings of the CCPA as its own precedent. South Corp. v. United States, 690 F.2d 1368, 1369 (Fed. Cir. 1982) (en banc).
 See supra note 82 and accompanying text.
 I suspect Buccafusco, Lemley, and Masur would agree.
 See infra Part III
 See supra note 10.
 See supra note 10 (discussing the concept of fragmented literal similarity). For example, in Apple v. Samsung, “the jury found—consistent with current law—that Samsung’s Galaxy Tab 10.1 did not infringe U.S. Patent No. Des. 504,889. . . . If Apple had a right to prevent others from copying the iPad2, the result would probably have been very different.” Sarah Burstein, Not (Necessarily) Narrower: Rethinking the Relative Scope of Copyright Protection for Designs, 3 IP Theory 114, 121 (2013). It’s true that the Samsung tablet looked like the iPad but the iPad was not a commercial embodiment of the D’889 patent. See id. at 121 n.50 (deeming Apple’s claim “dubious”). Indeed, Apple’s own exhibits in its UK case against Samsung emphasize the differences between the clunky, proto-iPad claimed in that patent and the accused Samsung tablets. See Samsung Electronics (UK) Limited v. Apple Inc.,  EWCA Civ 1339 ¶ 43.
 Cf. Shyamkrishna Balganesh, Irina D. Manta & Tess Wilkinson-Ryan, Judging Similarity, 100 Iowa L. Rev. 267, 271 (2014) (finding, in experiments with mock copyright juries, “that when provided with additional information about the simple fact of copying or the creative effort that went into the protected work, we saw an appreciable upward shift in subjects’ assessments of similarity between the works”). Of course, plaintiffs can already get in evidence of copying by adding a trade dress claim. But that just illustrates another problem with our full-cumulation regime.
 For more thoughts on why a requirement of copying and a defense of fair use might not be as narrowing as they might seem, see Burstein, Narrower, supra note 88, at 118–27.
 Buccafusco, Lemley & Masur, supra note 1, at 80 (footnotes omitted) (referring to Sport Dimension, Inc. v. Coleman Co., 820 F.3d 1316 (Fed. Cir. 2016)).
 For more on partial claiming, see Sarah Burstein, The “Article of Manufacture” Today, 31 Harv. J.L. & Tech. 781, 787–78 (2018).
 See Buccafusco, Lemley & Masur, supra note 1, at 124.
 Buccafusco, Lemley & Masur do, at one point, critique “[t]he current practice of ‘dotted line’ drawings” to disclaim portions of a design, which could possibly be read as a concern about statutory subject matter. See supra note 1, at 134. But their real concern seems to be the line drawings, not the disclaimers. See id. at 134 n.296 (“Patentees would still need some mechanism to disclaim portions of the photographed design, as they do today with dotted lines.”). In other words, it appears they are proposing that design patent applicants provide more visual details for whatever they claim—not necessarily that they be made to claim whole designs.
 35 U.S.C. § 171(a) (2018).
 This practice can be traced back to the U.S. Court of Customs and Patent Appeals’ decision in In re Zahn, 617 F.2d 261 (C.C.P.A. 1980). The Federal Circuit has adopted the holdings of the CCPA as its own precedent. South Corp. v. United States, 690 F.2d 1368, 1369 (Fed. Cir. 1982) (en banc).
 See Burstein, The “Article of Manufacture” Today, supra note 93 at 836–37 (noting that the very concept of what constitutes a protectable “design” has changed since Congress enacted the “total profits” remedy in 1887).
 See supra note 10 and accompanying text. See also, e.g., Crocs, Inc. v. Int’l Trade Comm’n, 598 F.3d 1294, 1303 (Fed. Cir. 2010) (“The ordinary observer test applies to the patented design in its entirety, as it is claimed.” (citing Braun, Inc. v. Dynamics Corp. of Am., 975 F.2d 815, 820 (Fed. Cir. 1992))).
 Burstein, The Patented Design, supra note 9, at 166.
 See generally 17 U.S.C. § 101 (2018) (providing protection for some designs as “[.p]ictorial, graphic, and sculptural works”); TrafFix Devices, Inc. v. Mktg. Displays, Inc., 532 U.S. 23, 28 (2001) (stating that “[i]t is well established that trade dress can be protected under federal law,” including trade dress consisting of “[t]he design . . . of a product”).
 See Buccafusco, Lemley & Masur, supra note 1, at 121–23 (discussing the problems caused by these overlaps). This may also be relevant to Buccafusco, Lemley, and Masur’s suggestion that “we might require designers to elect either copyright or design patent protection.” See id. at 81.
 See Buccafusco, Lemley & Masur, supra note 1, at 84–85 (“[S]ome innovators attempt to skirt the rigors of utility patent law by seeking protection through either the copyright or design patent regimes. These types of rights can amount to ‘backdoor utility patents.’”). See also supra Part II.