From “Substantially the Same” to “Substantially Similar” and “Plainly Dissimilar”: How the Design Patent Infringement Standard Drifted from Gorham

By Robert G. Oake, Jr. Updated September 28, 2026.

Case and statutory citations in this article link to the documents in ProSearch. A free Design Patent Pro account is all that is needed to open them.

Download this article as a PDF (54 pages)

Principal, Oake Law Office, PLLC. The author served as lead counsel for Egyptian Goddess in Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc), and filed amicus curiae briefs in support of petitions for rehearing en banc in North Star Technology International Ltd. v. Latham Pool Products, Inc., No. 23-2138 (Fed. Cir. 2025), and Range of Motion Products, LLC v. Armaid Co., No. 23-2427 (Fed. Cir. 2026). An earlier version of this Article was posted on SSRN in August 2026. This version was revised in September 2026 to address the Federal Circuit’s denial of rehearing en banc in Range of Motion and its decision in Jacki Easlick, LLC v. CJ Emerald. The views expressed are the author’s own and do not necessarily represent the views of any client or any other party.

Contents

Abstract

In Gorham Co. v. White, the Supreme Court established that design patent infringement occurs when two designs are “substantially the same” in the eye of an ordinary observer, such that the resemblance would deceive a purchaser into buying one supposing it to be the other. Over the past two decades, courts have increasingly substituted the phrase “substantially similar” for “substantially the same.” The Federal Circuit’s Egyptian Goddess decision further introduced “plainly dissimilar” and “sufficiently distinct” language as a threshold way to dispose of cases in which non-infringement appears clear.

These verbal substitutions are not harmless. They fundamentally alter the analytical framework, shifting the factfinder’s attention from whether designs are substantially the same in overall appearance to a feature-by-feature cataloging of differences. This Article demonstrates that “substantially similar” is not “substantially the same,” that the reframing toward dissimilarity has produced a doctrinal drift away from Gorham’s holistic, deception-based test, and that this drift has been compounded by courts’ routine use of side-by-side comparison methodologies and their increasingly aggressive treatment of functional features during claim construction.

As this Article further shows, the “substantially similar” substitution is not a recent phenomenon. It is visible in the federal reporters within thirteen years of Gorham itself. The case law divides into three general categories: decisions that remain faithful to Gorham by using “similar” only descriptively while stating the rule as “substantially the same,” decisions that treat “substantially the same” and “substantially similar” as interchangeable statements of the rule itself, and decisions that affirmatively restate the rule as “substantially similar.” The second and third categories pervade the working language of design patent infringement analysis across more than a century of decisions.

The accumulated drift has now reached the Federal Circuit, and Chief Judge Moore’s dissent in Range of Motion Prods., LLC v. Armaid Co. (Fed. Cir. 2026) has brought it to the forefront of design patent jurisprudence. In August 2026 the court denied rehearing en banc, with four of the eleven participating judges voting to rehear the case. The opinions on that denial separate the two drifts this Article describes. The concurrence states the rule as “substantially the same” but defends the “plainly dissimilar” screen. The dissent attacks the screen but states the rule as “substantially similar.” This Article answers the concurrence, including its reliance on Gorham’s statement that differences are to be considered. It also examines Jacki Easlick, LLC v. CJ Emerald, in which the Federal Circuit, three days after the denial, affirmed a preliminary injunction ruling that rested on a side-by-side study, a sorting test, and the plainly dissimilar screen. The district court opinion in that case exhibits every element of the drift. The Easlick briefing defended the current framework on a number of grounds, including that the prior art comparison is a “one-way ratchet” available only to accused infringers, and that the “plainly dissimilar” threshold is a justified screen against abusive “Schedule A” mass-defendant litigation. This Article addresses those defenses, explains how the law went astray, and proposes a return to the Gorham framework as properly understood.

I. Introduction

In 1871, the Supreme Court announced the test for design patent infringement in Gorham Co. v. White: “[I]f, in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same, if the resemblance is such as to deceive such an observer, inducing him to purchase one supposing it to be the other, the first one patented is infringed by the other.”1 For over a century, this formulation guided design patent law. The test asked a single, holistic question: whether the overall visual impression of two designs was so closely aligned that a purchaser could be deceived. And it anchored that question in the practical reality of the marketplace.

Something has gone wrong. Courts increasingly have swapped “substantially similar” for “substantially the same,” treating the two phrases as if they were interchangeable.2 And in Egyptian Goddess, Inc. v. Swisa, Inc.,3 the Federal Circuit introduced “plainly dissimilar” and “sufficiently distinct” language that later decisions turned into a threshold gatekeeping screen, shifting the inquiry away from whether designs are substantially the same toward whether they can be distinguished. In the wake of that shift, courts have substituted the language of similarity for sameness, and dissimilarity for noninfringement, blurring the governing standard. Chief Judge Moore’s dissent from the panel decision in Range of Motion Products, LLC v. Armaid Co. brought these concerns into sharp relief, acknowledging that the court’s own language in Egyptian Goddess constituted a “linguistic sleight of hand” that “meaningfully changed” the law, and citing amicus briefs that “convincingly explain[ed]” how the shift produced real-world differences.4 On August 11, 2026, the court denied rehearing en banc in Range of Motion. Four of the eleven participating judges would have granted it, and the opinions on both sides of the vote engage the questions this Article addresses.5 Part IV.E answers the concurrence. Three days later, a panel of the court affirmed the denial of a preliminary injunction in a second design patent case, Jacki Easlick, LLC v. CJ Emerald, in a nonprecedential opinion.6 Part VI examines Easlick as a case study in the drift’s operation, from the district court through the appeal, and addresses two arguments its briefing placed before the court.

The shift from “substantially the same” to “substantially similar” has become so pervasive that it now occurs naturally throughout modern design patent discourse. Indeed, as Parts III.C and III.D of this Article document, the substitution is not new. It began appearing in federal decisions within 13 years of Gorham itself, and the case law from Tomkinson onward divides into three categories that together reflect more than a century of accumulated drift.7

This Article contends that the substitution of “substantially similar” and the related “plainly dissimilar” for “substantially the same” is not a harmless variation in phrasing. It is a change in the operative legal standard that alters the mode of analysis, redirects the factfinder’s attention, and tends to produce different outcomes. The Article further contends that this verbal drift has been compounded by two related methodological errors: the routine use of side-by-side comparison as the default analytical method, and the improper treatment of design features deemed functional during claim construction. Together, these errors have moved design patent infringement analysis far from the holistic, deception-centered inquiry that Gorham established, replacing it with a feature-by-feature difference-hunting exercise that systematically favors findings of non-infringement and has effectively removed the jury from its factfinding role in design patent cases.

At its core, the argument is this: the modern test for design patent infringement is no longer truly a deception test. Gorham asked whether an observer would be deceived in the marketplace. The test that has emerged in Gorham’s name asks whether designs can be analytically differentiated under controlled comparison. The two are not the same inquiry. Restoring Gorham therefore requires more than restoring its words. It requires restoring the deception inquiry those words served.

II. The Gorham Framework: “Substantially the Same”

To understand how the law has drifted, one must first understand what Gorham actually established, and in particular, what work each word in the phrase “substantially the same” is doing.

A. The Holistic Inquiry

Gorham was deliberate in its formulation. The Supreme Court explained that design patent infringement depends on “the effect of the whole design,” not on an analysis of individual features in isolation.8 The Court recognized that designed objects are perceived holistically, as unified visual impressions, and that the legal test must reflect that perceptual reality. A purchaser encountering a product in the marketplace does not mentally inventory its individual features and check them against a list. The purchaser forms an overall impression, and this impression determines whether deception occurs.

The phrase “substantially the same” is the linguistic vehicle for this holistic inquiry. The word “same” anchors the comparison in identity, in the recognition that two objects are, for practical purposes, the same thing. “Substantially” qualifies this anchor, acknowledging that perfect identity is rare, but only to the extent necessary to accommodate trivial, immaterial variations. The phrase does not invite the observer to catalog differences. It invites the observer to assess whether the designs are so alike in their overall effect that the differences are below the notice of the ordinary observer.

B. “Substantially” Defined by Deception

Critically, Gorham does not leave the word “substantially” vague. It gives the word a precise, functional meaning: two designs are “substantially the same” when “the resemblance is such as to deceive” an ordinary observer, “inducing him to purchase one supposing it to be the other.”9 The measure of “substantially” is not quantitative. It is not a percentage of shared features or a score on a similarity scale. It is consequential. It asks whether the degree of sameness is enough to produce a specific real-world outcome: the deception of a purchaser. If the resemblance deceives, the designs are substantially the same. If it does not, they are not.

C. The Calibration Mechanisms

The Gorham formulation embeds multiple calibration mechanisms within a single inquiry, each providing both precision and flexibility without requiring a separate doctrinal step.

The ordinary observer. The hypothetical ordinary observer is not a fixed, universal figure. The observer is defined by the market for the product at issue. For example, an ordinary observer of surgical instruments brings a trained eye and a keen sensitivity to formal differences that might be invisible to a casual purchaser of kitchen utensils. The same objective degree of visual difference between two designs could easily be enough to avoid deception in one product market while being entirely insufficient in another, not because the designs differ, but because the observer’s sophistication and attentiveness differ.

The purchasing context. Gorham does not ask whether two designs look the same under controlled, artificial conditions. It asks whether an ordinary observer, “giving such attention as a purchaser usually gives,” would be deceived. This places the comparison squarely in the real-world activity of purchasing, with all of its attendant variables: the amount of time a purchaser spends examining the product, whether the purchase is impulsive or deliberated, whether the price is trivial or significant, and whether the purchaser is comparing the designs side by side or recalling one from memory when encountering the other. A consumer who spends weeks researching a major purchase is far harder to deceive than one who grabs an impulse item off a shelf.

The prior art context. Egyptian Goddess later made explicit what was implicit in the original Gorham formulation: the ordinary observer is “familiar with” the prior art.10 The idea did not originate in 2008. The Sixth Circuit described the ordinary observer in 1933 as “one who, though not an expert, has reasonable familiarity with such objects, and is capable of forming a reasonable judgment when confronted with a design therefor as to whether it presents to his eye distinctiveness from or similarity with those which have preceded it.”11 On this Article’s reading of Gorham, that familiarity is not an optional add-on. The prior art is part of the very definition of the observer through whose eyes the comparison is made.

Current Federal Circuit law reads the prior art’s role more narrowly. As Part IV explains, Egyptian Goddess permits a court to end the analysis “without more” when the designs appear sufficiently distinct. The Federal Circuit restated that reading in August 2026: comparing the claimed and accused designs with the prior art “is beneficial only when the claimed and accused designs are not plainly dissimilar.”12 The disagreement turns on a distinction the case law rarely makes explicit. The prior art does two different kinds of work in this analysis, and this Article refers to them as background prior art and comparative prior art. Background prior art is the field of earlier designs with which the ordinary observer is familiar. It defines who the observer is. Under Egyptian Goddess, it sets the observer’s perceptive ability: in a crowded field, the observer is more attuned to small differences between designs, and in a sparse field, broader similarities may be enough to deceive. Comparative prior art consists of the specific references placed beside the claimed and accused designs in the three-way comparison. The “without more” language can fairly be read to refer only to comparative prior art. It cannot refer to background prior art, because the observer who would find two designs sufficiently distinct is defined, in the same paragraph of Egyptian Goddess, as an observer who is conversant with the prior art.13 A court that skips the three-way comparison in a truly easy case may be applying Egyptian Goddess as written. A court that sets aside the background prior art is not, because without it there is no ordinary observer, and without the observer there is no test.

Taken together, these mechanisms provide Gorham’s “substantially the same” test a framework that is both precise and flexible. It is precise because it points to a definable outcome: deception. It is flexible because the threshold for deception shifts with the observer’s identity, the purchase conditions, and the state of the prior art. And it keeps the inquiry anchored in holistic visual impression rather than feature-by-feature decomposition.

III. The Linguistic Drift: From “The Same” to “Similar”

The most common reaction to the argument developed in this section is that the distinction between “substantially the same” and “substantially similar” is merely semantic, a difference in phrasing without practical consequence. That reaction is understandable but mistaken. Legal standards operate through language. The words a court uses to formulate a rule shape how that rule is applied, what evidence the factfinder treats as relevant, and how the factfinder organizes its analysis. A change in formulation predictably alters how results are obtained and which results most often emerge, even when the change is small.

This is not an abstract claim. The psychological research on framing effects, developed by Tversky, Kahneman, Mussweiler, and others and discussed in Part IV.D, demonstrates that the direction in which a comparative question is framed systematically affects the judgments people reach. Survey evidence offered to the Federal Circuit in North Star points in the same direction. The phenomenon documented in this Part is the same phenomenon on a doctrinal scale. When courts adopt a different verbal standard, they conduct the inquiry differently and reach different conclusions, even when they believe they are applying the same rule. The proper test of whether a linguistic change matters is not whether the change is large, but whether it consistently produces different outcomes. By that test, the migration from “substantially the same” to “substantially similar,” and from there to “plainly dissimilar,” is consequential.

A. The Conceptual Distinction

The words “same” and “similar” are not synonyms. They describe different relationships between the objects being compared, and substituting one for the other changes the nature of the comparison in ways that are both predictable and consequential.

To say that two things are “substantially the same” is to assert that they are, for all practical purposes, the same thing, and that whatever differences exist are trivial and immaterial. The word “same” holds the comparison together. It keeps the observer’s attention on the unified visual impression and treats differences as noise to be filtered out rather than signal to be amplified.

“Similar,” by contrast, presupposes difference. Similarity is, by definition, a relationship between things that are not identical. You never say that two identical things are “similar.” You say they are the same. The very act of framing a comparison in terms of similarity carries an embedded acknowledgment that differences exist, and that acknowledgment quietly reshapes how the analysis unfolds.

Gorham itself occasionally uses the language of resemblance and similarity, but it does so descriptively and illustratively rather than as part of its formulation of the legal rule. When the Court illustrates its reasoning, as when it observes that a second carpet may bear “similar wreaths” arranged like the first’s while expressing the “same conception” in the designer’s mind, or when it characterizes the relation between features of the compared designs, “similar” appears naturally.14 When the Court formulates the governing rule, however, it returns consistently to sameness: “sameness of appearance,” “sameness of effect upon the eye,” “substantial identity,” and ultimately whether “two designs are substantially the same.”15

The same point applies to Gorham’s occasional use of difference language. When the Court applied its rule to the facts, it sometimes phrased the finding negatively, observing that there was “no substantial difference” between the designs.16 That is descriptive of the rule’s application, not a restatement of the standard. The standard itself remains “substantially the same.” Gorham said as much about differences generally: they may be “considered,” but “the controlling consideration is the resultant effect.”17 That descriptive-versus-formulational distinction is important. It locates similarity at the feature or component level and sameness at the level of overall visual impression, which is the architecture this Article attributes to Gorham’s test. And it supplies the baseline against which the drift documented in Part III.D should be evaluated. The problem this Article identifies is not the appearance of the word “similar” in a design patent opinion, which Gorham itself permits and sometimes invites. It is the migration of that word from descriptive use into the stated legal standard. As Part III.D develops, the case law divides into three categories defined precisely by how they handle that line.

There is also a structural problem with the phrase “substantially similar” that goes beyond the substitution of one word for another. The phrase “substantially the same” contains a single flexible modifier, “substantially,” applied to a fixed reference point, “the same.” By contrast, “substantially similar” introduces two layers of indeterminacy: “similar,” which is itself a relative and undefined concept, and “substantially,” which further modifies that already flexible standard. The result is a compounded ambiguity that untethers the analysis from any fixed baseline.

The move from “the same” to “similar” might be justified, implicitly or explicitly, as a way to give the infringement standard greater flexibility. The intuition is understandable. Courts may be concerned that a requirement of “sameness” sounds too rigid, too demanding, or insufficiently adaptable to the wide range of products and markets in which design patents operate. “Similarity” appears to soften the standard and provide room for practical judgment.

But this intuition rests on a misunderstanding of the Gorham framework. Gorham already provides all the flexibility the test requires, and it does so in a principled way. The flexibility does not come from diluting the standard from sameness to similarity. It comes from the calibration mechanisms embedded in the test itself: the identity of the ordinary observer, the conditions of purchase, and the state of the prior art. These mechanisms allow the threshold for deception to adjust to context without altering the core inquiry.

Replacing “substantially the same” with “substantially similar” does not add useful flexibility. It removes the anchor that makes the flexibility meaningful. Once the reference point of overall identity is lost, the analysis no longer has a stable baseline. And in the absence of that baseline, courts tend to compensate by introducing rigid analytical tools elsewhere, such as feature-by-feature comparisons, weighing of similarities and differences, and threshold screens like “plainly dissimilar.” What begins as an effort to make the test more flexible ends up making it more mechanical and more biased toward findings of non-infringement.

This dynamic helps explain why the shift to “substantially similar” has not benefited patentees. An untethered standard does not create balanced flexibility. It creates indeterminacy. And indeterminacy invites courts to resolve cases by focusing on differences, because differences are always easier to identify and articulate than overall visual sameness. Without a fixed reference point of identity, there is no principled way to say that similarities outweigh differences. The path of least resistance is to catalog the distinctions and conclude that the designs are not sufficiently similar. The intuition that “sameness” sounds stricter than “similarity,” and that the substitution therefore tilts toward patentees, misreads Gorham. Gorham’s sameness standard is already holistic and anti-dissection, which is forgiving to patentees in the right way; an untethered similarity standard invites the feature-level comparison that, as the case law bears out, systematically benefits accused infringers.

B. Similarity Invites Subtraction

When an observer is told that two things are “similar,” the natural cognitive move is to figure out how they are similar, which means identifying what they share and, inevitably, what they do not. The mind begins sorting features into two categories: shared and unshared. And once that sorting begins, the differences do not sit passively. They become salient. They attract attention. The observer starts by asking, “How different are they?” rather than “How alike are they?” The whole direction of the inquiry shifts.

A “sameness” framework resists this sorting. Because the premise is identity rather than resemblance, differences must overcome an initial presumption of unity rather than emerging naturally from the comparison. The observer’s attention remains on the overall impression, and differences register only if they are significant enough to disrupt that impression.

This difference in framing also has methodological consequences. A similarity-based framework naturally aligns with side-by-side comparison, in which the observer can move back and forth between designs to identify points of divergence. A sameness-based framework, by contrast, aligns with the marketplace conditions reflected in the Gorham record, where the observer forms an overall impression and later encounters the accused design in a different context. In that setting, memory and gestalt perception dominate, and minor differences fade rather than accumulate.

In this way, the linguistic shift toward similarity and the methodological shift toward side-by-side comparison reinforce each other. Both direct attention to differences. Both encourage feature-by-feature analysis. And both move the inquiry away from the holistic, deception-based test that Gorham established.

C. The Pre-Egyptian Goddess Roots of the Drift

The substitution of “substantially similar” for “substantially the same” did not originate with Egyptian Goddess or its progeny. It began appearing in federal design patent decisions within a generation of Gorham itself. In Tomkinson v. Willets Manufacturing Co., decided in 1884, only thirteen years after Gorham, the Circuit Court for the Southern District of New York articulated the infringement standard as whether, “to the eye of an ordinary person[,] the two are substantially similar.”18 The court’s substantive analysis was otherwise faithful to Gorham’s holistic approach, expressly rejecting feature-by-feature dissection on the ground that “persons skilled in the art are able to detect differences” that are “wholly unimportant, and unobserved by those whose pursuits are in other directions.”19 But the linguistic substitution was already complete.

By 1928, Try-Me Beverage & Compound Co. v. Metropole reflected the same pattern in a more developed form. The court in Try-Me correctly rejected two of the methodological errors this Article criticizes: it repudiated side-by-side comparison,20 and, more remarkably, anticipated by nearly a century Chief Judge Moore’s critique of dissimilarity-hunting, holding that “[t]he test of infringement is not whether dissimilarities between two designs may or not be found.”21 Yet in the same breath, the court stated the governing standard as whether “two designs are substantially similar, and if the resemblance is such as to deceive an observer.”22

Try-Me illustrates the thesis of this Article. A court fully committed to Gorham’s holistic, deception-based inquiry, one that understood intuitively why both side-by-side juxtaposition and difference-cataloging were inconsistent with that inquiry, nonetheless absorbed the verbal substitution without remarking on it. The “substantially similar” formulation had already become so naturalized in the legal vocabulary of design patents that it coexisted undisturbed alongside a faithful application of Gorham’s methodology. That is precisely the mechanism this Article identifies: the verbal drift operates below the level of conscious doctrinal choice, reshaping analysis even in the work of courts that get the larger principles right.

The pattern extends beyond Tomkinson and Try-Me. Other federal appellate decisions in the decades after Gorham used difference language in ways that fall along the same lines this Article identifies. The Third Circuit’s Monroe v. Anderson described Gorham as asking about “the sameness of, or difference between” the designs, a rule-stating use of difference alongside sameness that places Monroe in Category 2 of the taxonomy developed in the next section. By contrast, the Ninth Circuit’s Zidel v. Dexter used “differences” in the application of the rule rather than as a restatement of it, observing that “[t]he differences in designs, which under the patent law will avoid infringement, are differences which will attract the attention of the ordinary observer.” The Second Circuit’s American Fabrics Co. v. Richmond Lace Works likewise began from the Gorham-style sameness formulation, whether the designs have “substantially the same effect upon the eye,” but affirmed non-infringement because “the general appearance of the two patterns is sufficiently different.” The same taxonomic distinction applies: descriptive applications of difference are consistent with Gorham; rule-stating substitutions are not.23

The implication is important. The reframing from “substantially similar” to “plainly dissimilar” in Egyptian Goddess did not introduce the similarity vocabulary into design patent law. It inherited a standard that had been linguistically unmoored for more than a century and accelerated the consequences of that unmooring by converting an already-drifted “substantially similar” test into an outright “plainly dissimilar” gatekeeping device. Understanding that the drift has deep historical roots strengthens rather than weakens the case for correction: courts have been operating with imprecise terminology for so long that the precision of Gorham’s actual words has been forgotten.

D. Three Categories: A Taxonomy of the Case Law

The reach of the design patent case law on this question is considerably broader than the handful of Federal Circuit decisions typically cited to illustrate it. Recognizing that breadth requires a three-category taxonomy. Only two of the three represent drift. Recognizing the first, the Gorham-faithful baseline, is essential to understanding why the other two are problematic.

Category 1: Descriptive use consistent with Gorham. As Part III.A observed, Gorham itself uses the language of resemblance and similarity, but only descriptively and illustratively, never as part of its rule formulation. Cases in this category do the same. They state the governing standard as “substantially the same,” or some close cognate such as “sameness of appearance” or “substantial identity,” and when they use “similar” or “similarity,” they do so in the ordinary descriptive sense that any discussion of two visual designs naturally produces: characterizing testimony, comparing particular features, or explaining why small differences do not matter. Gorham itself is the archetype. The Supreme Court’s later design patent decision in Smith v. Whitman Saddle Co. belongs in this category, restating the test as “sameness of appearance” and “sameness of effect upon the eye” without invoking similarity as the legal standard. So do the decisions that expressly rejected side-by-side comparison, including Ashley v. Weeks-Numan Co. and Inflexible Co. v. Megibow. Ashley frames the inquiry as whether the ordinary observer “would purchase the defendant’s inkstand, believing it to be that of complainant’s,” and Inflexible articulates the test as “sameness of appearance” producing “substantial identity of appearance.” These cases use “similar” at most in service of a sameness-based rule; they do not substitute one for the other.

The existence of this baseline matters. It confirms that this Article’s thesis is not a complaint about every appearance of the word “similar” in a design patent opinion. It is a complaint about the migration of that word from descriptive use into the stated legal standard. Categories 2 and 3 document that migration in two different forms.

Category 2: Interchangeable treatment. A large body of cases treats “substantially the same” and “substantially similar” as interchangeable statements of the legal standard itself. These courts recite Gorham’s “substantially the same” language correctly, then restate or gloss it as “substantially similar” in the very next breath, or slide between the two formulations when articulating the rule. The two phrases appear together in rule-stating contexts, not merely in analytical or descriptive ones. The substitution in this category is subtler than an outright reformulation, but it is not innocent. Once “substantially similar” is admitted as an alternative way to state the rule, the cognitive pull of similarity framing begins to operate on the analysis regardless of which formulation the court happens to use in any given sentence. This pattern is visible across the Federal Circuit’s pre-Egyptian Goddess decisions, including Shelcore, L.A. Gear, KeyStone Retaining Wall Systems, Elmer, Goodyear Tire, and Contessa Food Products, as well as in district court decisions from across the country.24

Range of Motion Products, LLC v. Armaid Co. itself illustrates how deeply Category 2 runs. The majority opinion recites Gorham’s “substantially the same” language through Egyptian Goddess but then states, in footnote 3, that “the ultimate question is whether the claimed and accused designs are substantially similar,” and proceeds to describe the test throughout its infringement analysis as whether the designs are “substantially similar.” The dissent participates in the same substitution.25 So, in different ways, do the opinions on the denial of rehearing en banc, which Part IV.E discusses.

The substitution is not confined to a single doctrinal step. The Range of Motion majority articulated two independent rationales for affirming summary judgment: that the designs were “plainly dissimilar,” and, in the alternative, that they were not “substantially similar” under a three-way comparison with the prior art. Both rationales are framed throughout in the language of substantial similarity rather than substantial sameness, and both rest on side-by-side comparison as the operative methodology. The drift does not respect the boundary between threshold tests and ultimate inquiries. Once it has entered the analysis, it shapes every step of it.

Category 3: Formal reformulation. A third group of decisions, including several by the Federal Circuit itself, has affirmatively restated Gorham’s infringement inquiry in terms of “substantially similar,” subordinating or abandoning “substantially the same.” In OddzOn Products, Inc. v. Just Toys, Inc., the Federal Circuit stated that the factfinder must determine whether the patented design as a whole is “substantially similar in appearance” to the accused design.26 In Hoop v. Hoop, the court described the test for infringement as whether the second asserted invention is “substantially similar” to the first.27 And in Arminak & Associates, Inc. v. Saint-Gobain Calmar, Inc., the court framed the ordinary observer test as one asking whether the observer would be deceived because the accused design is “substantially similar” to the patented design.28 District courts applying these decisions have followed suit, with Eastern America Trio Products, Inc. v. Tang Electronic Corp. referring directly to the “Gorham test for substantial similarity,”29 and Hsin Ten Enterprise USA, Inc. v. Clark Enterprises and Tecumseh Products Co. v. Briggs & Stratton Corp. each articulating the inquiry as whether the accused design is “substantially similar in appearance” to the patented design.30 These are not passing uses of the phrase. They are working statements of the legal rule, reproduced in subsequent decisions and treatises as authoritative.31

The Arminak example is particularly consequential for reasons developed in Part V. The same decision that formally restated the Gorham inquiry as one of substantial similarity also declared that there was “no authority” supporting the principle that a detailed side-by-side comparison is improper.32 Both moves work in the same direction, and they appear in the same opinion. A court that has already recast the operative standard from sameness to similarity is institutionally predisposed to adopt a comparison methodology that foregrounds difference, because similarity and difference are the natural categories the reformulated standard generates. The verbal drift and the methodological drift are not independent phenomena; they are mutually reinforcing expressions of a single underlying shift in how the inquiry is conceived.

The reason Categories 2 and 3 produce parallel effects despite their different mechanisms is the one developed in Parts III.A and III.B. The cognitive pull of “similar” operates whenever the word enters the rule-stating vocabulary, regardless of how it gets there. Category 2’s interchangeable treatment admits the word into the rule alongside “substantially the same”; Category 3 substitutes the word outright. Either way, once “substantially similar” is available as a rule-statement, the identity anchor that “the same” supplies is compromised, and the analysis tilts toward the feature-level comparison that similarity framing naturally invites. Only Category 1, where “similar” is confined to the descriptive role it plays in Gorham, preserves the architecture of the test. That is why the distinction between descriptive and rule-stating use of the word is not a linguistic nicety. It is the distinction that separates faithful application of Gorham from drift.

E. Tracing the Post-2010 Acceleration

Against this long-standing backdrop, the most visible acceleration of the drift occurred in the years after Egyptian Goddess, as courts began using “similar” as shorthand without recognizing that the shorthand was doing substantive work. Tracing the progression reveals how the decades-old verbal convenience documented in the preceding subsections was converted into an overt doctrinal shift.

It is important to note at the outset that even though Egyptian Goddess used the phrase “plainly dissimilar,” it did not use the phrase “substantially similar.” The en banc opinion consistently uses “substantially the same” when describing the Gorham standard.33 The accelerated drift began elsewhere.

The first meaningful post-Egyptian Goddess inflection point came in Crocs, Inc. v. International Trade Commission in 2010.34 Crocs’s contribution to the drift was methodological rather than verbal. The opinion states Gorham’s test in sameness and deception, and its ultimate findings are framed in that language. But it also announced that the “proper comparison requires a side-by-side view” of the claimed design and the accused products, and it conducted its own analysis in that format. As Part V develops, that prescription converted a courtroom display into the prescribed method of analysis, and the method carries its own frame: juxtaposition directs the eye to differences just as surely as similarity language directs the mind to them. From Crocs forward, the verbal drift and the methodological drift traveled together.

The procedural drift deepened in later post-Egyptian Goddess cases that treated substantial similarity as a summary-judgment threshold. Courts framed the question as whether any reasonable factfinder could find the accused design’s overall visual effect “substantially similar” to the claimed design, reinforcing the movement away from Gorham’s deception-based formulation and toward a judicial gatekeeping inquiry focused on degree of similarity.35

The Columbia Sportswear litigation pushed the drift further. In the 2019 appeal, the court framed the comparison in terms of whether design elements affect “the overall visual impression that the designs are similar.”36 In the 2023 appeal, the court stated the operative question as whether an ordinary observer would deem “the claimed and accused designs similar enough to constitute design-patent infringement.”37 This language is even further removed from Gorham’s deception framework, restating the ultimate infringement as a graded assessment of similarity.

What is important to understand about this progression is that the shift is not just verbal. It is structural. The original Gorham test asks a binary, perception-based question: would the observer be deceived? The drifted version asks a scalar, analytical question: how similar are these, and do the similarities outweigh the differences? These are different cognitive tasks, and they produce different results. The binary deception question keeps the observer’s attention on the overall impression. The scalar similarity question invites the observer to parse individual features and weigh them against each other, which is precisely the element-by-element decomposition that Gorham sought to avoid.

IV. The Egyptian Goddess Reframing: From Similarity to Dissimilarity

The substitution of “substantially similar” for “substantially the same” was already problematic. But Egyptian Goddess compounded the problem by introducing a second, even more consequential reframing: the “plainly dissimilar” or “sufficiently distinct” threshold test.

A. What Egyptian Goddess Intended

Egyptian Goddess was rightly celebrated for overruling the point of novelty test, which had proved difficult to administer and which conflicted with Gorham’s ordinary observer framework.38 The en banc court restored the ordinary observer test as “the sole test” for design patent infringement and refined it by incorporating the context of the prior art into the observer’s identity.39 These were sound developments.

B. What Egyptian Goddess Inadvertently Did

In the course of explaining how the refined test should work, however, the court introduced language that has proved deeply consequential:40

“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, as required by Gorham. In other instances, when the claimed and accused designs are not plainly dissimilar, 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.”

What was intended as a commonsense observation, that some designs are so obviously different that a comparison with the prior art adds nothing to the analysis, became in practice a threshold gatekeeping test. Courts began applying the “plainly dissimilar/sufficiently distinct” standard as the first step of the infringement analysis, using it to dispose of cases at summary judgment without ever conducting a prior art comparison.41

Chief Judge Moore’s dissent in Range of Motion identified both the problem and much of the solution. She recognized that the reframing constituted a “meaningful[ ] change[ ]” in the substantive test, one that occurred “[w]ithout realizing it.”42 The frame of reference was inverted. Instead of asking whether two designs are substantially the same in overall appearance, which directs the factfinder’s attention to similarities, courts began asking whether the designs are “sufficiently distinct” or “plainly dissimilar,” which directs the factfinder’s attention to differences.43

Chief Judge Moore also addressed the prior art problem head-on. Responding to the majority’s defense that Egyptian Goddess “merely recognize[s] that in some cases, the claimed and accused designs may be so clearly not similar (or ‘dissimilar’) as to remove the need to consider the prior art,” she argued that courts should “always” compare the claimed and accused designs in light of the prior art, “with no special exception for plainly dissimilar designs, which has proved unworkable.”44

This is right, and for two independent reasons, one for each kind of prior art. The practical reason concerns the comparative prior art, and it is the one Chief Judge Moore emphasized: the plainly dissimilar exception has become a gatekeeping device that enables courts to dispose of infringement claims at summary judgment without ever conducting the three-way comparison that would inform the analysis. The conceptual reason concerns the background prior art, and it runs even deeper: as discussed in Part II.C, the background prior art is not an optional input that can be consulted or set aside depending on how different the designs look at first glance. It is part of the very definition that shapes who the ordinary observer is: how sophisticated, how discriminating, how likely to be deceived. Even a court that reads “without more” to excuse the three-way comparison in a truly easy case must still judge the designs through the eyes of an observer who knows the field. A court that applies the “plainly dissimilar” test without considering the background prior art has not simplified the Gorham test. It has applied a different test, one in which the observer has been stripped of the very knowledge that defines the observer’s identity.

C. Early Judicial Recognition of the Dissimilarity Problem

The recognition that an inquiry organized around identifying differences is incompatible with Gorham’s holistic test is not new. Nearly a century before the framing-effects literature gave empirical shape to the insight, the district court in Try-Me Beverage & Compound Co. v. Metropole stated the principle directly: “The test of infringement is not whether dissimilarities between two designs may or not be found.”45 Courts applying Gorham understood that an inquiry organized around the identification of differences was incompatible with a test organized around overall visual impression, even if they could not articulate the cognitive mechanism that would later be demonstrated in the psychological research. Egyptian Goddess’s “plainly dissimilar” reframing did not merely drift from Gorham in the abstract; it drifted from a line of authority that had, for good reasons of its own, expressly rejected the methodology the reframing now invites.

D. The Psychological Evidence

The claim that reframing the inquiry from similarity to dissimilarity changes outcomes is not merely a doctrinal observation. A substantial body of psychological research confirms that the direction in which a comparative question is framed systematically affects the judgments people reach.

Tversky and Kahneman demonstrated in their foundational work on framing effects that “[t]he psychological principles that govern the perception of decision problems and the evaluation of probabilities and outcomes produce predictable shifts of preference when the same problem is framed in different ways.”46 Mussweiler’s research on comparison processes in social judgment is even more directly on point: “judges who engage in similarity testing selectively generate knowledge indicating that the target is similar to the standard, whereas judges who engage in dissimilarity testing selectively generate knowledge indicating dissimilarity.”47 The framing does not merely nudge the observer’s attention. It directs the observer’s active generation of supporting evidence.

Chief Judge Moore’s dissent in Range of Motion recognized the force of this research, citing Tversky and Kahneman and other authorities.48 She also invoked the vivid analogy of the childhood puzzle that “asks you to circle the differences between two otherwise seemingly identical pictures”: anyone who has played such a game knows that hunting for differences makes them feel more significant and causes the observer to lose sight of the overall resemblance.49

Survey data has also been offered to the Federal Circuit. In an amicus brief supporting rehearing in North Star, the Institute for Design Science and Public Policy reported that respondents shown designs from Gorham and from the Apple and Samsung litigation, all held to infringe, called them “plainly dissimilar” at rates of 62 to 74 percent.50 Chief Judge Moore cited the study, and this author’s amicus brief in Range of Motion cited it as well. The results are what the framing research would predict. Because the study asked only the “plainly dissimilar” question, a follow-up that put the “substantially the same” question to a matched group of respondents under the same conditions would measure the difference between the two framings directly.

E. The Denial of Rehearing in Range of Motion

On August 11, 2026, the Federal Circuit denied rehearing en banc in Range of Motion. Four of the eleven participating judges would have granted it.51 The two opinions filed on the denial divide in an instructive way. Each is faithful to Gorham in one respect and drifts from it in the other.

Judge Cunningham’s concurrence, joined by Judge Hughes, states the rule in Gorham’s words. It asks whether designs are “substantially the same,” and it describes the easy case as one in which no ordinary observer would be “deceive[d]” or “induc[ed]” to purchase one design supposing it to be the other.52 Its one reference to “overall similarity” describes the district court’s assessment rather than restating the rule.53 In the taxonomy of Part III.D, the concurrence is Category 1. But it defends the “plainly dissimilar” screen. Chief Judge Moore’s dissent, joined by Judge Reyna, attacks the screen and would give the questions of functionality and infringement to the jury. Yet it states the Gorham test as whether “an ordinary observer would find the two designs are substantially similar in overall appearance,” and it proposes to “return us to Gorham” under that formulation.54 The dissent also quotes Gorham’s “substantially the same” language, so it treats the two phrases as interchangeable.55 That is Category 2.

The two opinions thus work as a natural experiment. The concurrence shows that restoring Gorham’s words does not by itself restore Gorham’s inquiry. A court can state the rule faithfully and still defend a screen that turns the inquiry toward differences. The dissent shows the converse. A judge can see clearly what the screen has done and still use the vocabulary with which the drift began. Both halves of the correction are needed.

The concurrence makes three arguments that this Article must answer. Its defense of the court’s role in deciding functionality at claim construction is addressed in Part VIII.

The first argument rests on Gorham itself. The concurrence quotes this passage: “We do not say that in determining whether two designs are substantially the same, differences in the lines, the configuration, or the modes by which the aspects they exhibit are not to be considered; but we think the controlling consideration is the resultant effect.”56 From it the concurrence concludes that assessing whether two designs are substantially the same “necessarily involves accounting for the ways in which they are similar and different.”57

This Article agrees with that conclusion, and so does Gorham. But the passage has two clauses, and the second one controls. Differences may be considered. What decides the case is “the resultant effect.” That is the distinction this Article has drawn throughout, between the method of analysis and the legal standard. A court may observe differences, discuss them, and take them into account. The question it must answer with them is whether the resultant effect would deceive an ordinary purchaser. The passage does not support a threshold framed in terms of dissimilarity. It points the other way. A screen that asks first whether the designs are “plainly dissimilar” makes differences the controlling consideration at the threshold, which is the one role the passage denies them. One page later, Gorham rejected a test that required “a comparison of the features which make up the two designs,” because “[s]uch a test would destroy all the protection which the act of Congress intended to give.”58

The concurrence also cites Smith v. Whitman Saddle Co., in which the Court observed that “the design of the patent had two features of difference as compared with the Granger saddle.”59 The citation helps the argument less than it appears to. The Granger saddle was prior art. The differences Whitman Saddle counted were differences between the patented design and the prior art, and the Court used them to identify what was new in the patented design, the sharp drop at the rear of the pommel. It then asked whether the accused saddles had that drop. They did not, and the Court concluded that the defendants’ saddle “could not be mistaken for the saddle of the complainant.”60 That is a comparison of the claimed and accused designs in light of the prior art, ending in a finding stated in terms of deception. It is the analysis the plainly dissimilar screen allows a court to skip.

The second argument concerns the remedy. The concurrence describes the dissent’s disagreement as “one of line-drawing: How easy is ‘truly easy-to-decide’?” If district courts have been finding designs plainly dissimilar as a matter of law when a genuine dispute of material fact remains, “the remedy is not to throw out years of well-settled design patent law but to reverse the decisions of those courts when they are appealed.”61

That remedy depends on two things the current framework does not supply. The first is a line. Reversal case by case requires a standard against which to measure the error, and the screen has none. The Federal Circuit has not said how different two designs must be before they are “plainly” dissimilar, and the dissent observes that district courts invoke the shortcut “[i]n the absence of clear standards.”62 Its answer to the concurrence is that “we are not setting the courts straight, because we are applying bad law.”63 The second is appellate review that reaches the question. Many cases in which the screen is applied never reach the Federal Circuit, and Schedule A cases, as Part VI.D explains, are rarely appealed at all. When a case does reach the court, the posture may keep the question from being decided. Three days after the denial of rehearing, the court decided Easlick. It reviewed the denial of a preliminary injunction for abuse of discretion. It affirmed a ruling in which the district court had declined to compare the prior art because there was “a substantial question” whether the designs were plainly dissimilar. It did not address that reason, and it wrote a nonprecedential opinion. Part VI.B discusses the decision. Whatever its merits, it did not draw a line.

The third argument is that Range of Motion was a poor vehicle, because the district court also conducted a three-way comparison with the prior art and the panel affirmed on both grounds.64 That is true, and it shows how far the allocation question reaches. The alternative ground was a judge’s comparison of the claimed design, the accused design, and the prior art, made on summary judgment. The dissent answered that this analysis is for the jury, because it “is deeply factual and determinative of the infringement issue in nearly all cases.”65 Part IX takes up that point.

V. The Side-by-Side Comparison Problem

The linguistic drift from “substantially the same” to “substantially similar” and “plainly dissimilar” has been compounded by a methodological shift in how courts conduct visual comparisons. In a substantial line of decisions, courts have used side-by-side comparison as the default method. This methodology produces the same difference-foregrounding effect that the reframed verbal standard produces linguistically.

A. What Gorham Actually Contemplated

The expert testimony on which the Supreme Court relied in Gorham consistently described comparisons made under realistic marketplace conditions, not side-by-side juxtaposition. One expert testified that purchasers, “seeing them apart, would mistake one for the other.”66 Another testified that if the accused design “were placed in a store different from that in which they had before seen the [patented] pattern, seven out of ten customers who buy silverware, would consider it the same pattern.”67 The scenario the Court described was one in which the purchaser carries a visual impression from one encounter and encounters the accused design at a later time, in a different place. The Court announced the deception test on that record; it did not announce a procedural rule governing how designs may be displayed in litigation. What the record supplies is the kind of encounter the test was built to police, and this Article argues that it is that encounter, not courtroom juxtaposition, that should fix the observer’s vantage.

This reflects how purchasing actually works in most product markets. Consumers rarely hold two competing products side by side and scrutinize them for differences. They form an impression of a product, carry that impression forward in memory, and encounter the competing product. Memory retains overall impressions, the gestalt, rather than granular detail. A comparison regime that mirrors this reality naturally emphasizes the holistic visual impression, which is precisely what Gorham calls for.

B. Post-Gorham Authority Rejecting Side-by-Side Comparison

Courts applying Gorham in the decades following the decision consistently refused to treat side-by-side inspection as the measure of infringement. In Ashley v. Weeks-Numan Co., the Second Circuit stated that “[i]t is not a proper test to place the two inkstands side by side, to determine whether or not there are certain differences.”68 In Inflexible Co. v. Megibow, the court found infringement despite “marked differences . . . readily perceived when placed side by side,” because those differences were “unnoticed when the laces are viewed separate and apart from each other.”69

The same principle was stated with particular force in Try-Me Beverage & Compound Co. v. Metropole, where the court adopted by quotation Judge Morris’s formulation in Coca-Cola v. Whistle that “side by side comparison in court is not a proper test.”70 What makes Try-Me especially instructive is that the court reached this conclusion as part of a broader commitment to Gorham’s holistic inquiry. The same opinion that rejected side-by-side comparison also rejected feature-by-feature dissimilarity-hunting.71 The two methodological commitments traveled together, as they should, because both follow from the same underlying understanding of what the ordinary observer test is asking.

Attorney William D. Shoemaker, in his 1929 treatise Patents for Designs, synthesized this line of authority: “[T]he test for substantial identity is to be made by ‘giving such attention as a purchaser usually gives,’ and not by making a minute side by side comparison.”72

District courts continued to recognize this principle. In Amerock Corp. v. Unican Security Systems Corp., where the designs involved kitchen cabinet hardware, the court took testimony establishing that hardware consumers generally did not view different designs simultaneously and therefore concluded that “differences between the patented and the accused design that are apparent in a side-by-side comparison will not necessarily dictate a finding of non-infringement.”73 The Hutzler court recognized the split in authority. It analyzed infringement by asking whether “a consumer who has seen [the patented] design in a store, and later sees [the accused] products, is likely to purchase the latter believing it to be the former.”74

C. The Federal Circuit’s Treatment of Side-by-Side Comparison

Despite this substantial contrary authority, the Federal Circuit declared in Arminak & Associates v. Saint-Gobain Calmar, Inc. that there was “no authority” supporting the argument that a detailed side-by-side comparison was improper.75 This assertion is hard to reconcile with the authorities cataloged above. The case cited by the Federal Circuit for this proposition, Elmer v. ICC Fabricating, Inc., states only that the court is required to “compare the patented design with the accused design.” It says nothing about whether that comparison should be conducted side by side.

Modern Federal Circuit authority has gone further than Arminak. In Crocs, the court stated that “[t]he proper comparison requires a side-by-side view of the drawings of the ’789 patent design and the accused products,” and it conducted its own analysis in that format.76 District courts have treated that statement as prescribing the method.77 Not all modern authority runs the same way; juries have been instructed that the test “does not require a side-by-side comparison” and that minor differences that do not change the overall visual impression do not preclude infringement.78 But juxtaposition is the prevailing practice in the decisions collected in this Article, and a candid account of current law must begin there. The argument of this Part is therefore not that present doctrine forbids side-by-side comparison. It plainly does not. The argument is that a side-by-side display is legitimate only as a display. When the ability to tell the designs apart under juxtaposition becomes the measure of infringement, the method has displaced the standard, and it is that displacement, not the visual aid, that the older authority correctly condemned.

The side-by-side approach does to the visual analysis precisely what the word “similar” does to the conceptual analysis: it foregrounds differences. When two designs are placed next to each other, the human eye naturally darts between them, seeking points of divergence. The perceptual mode shifts from holistic impression to analytical comparison. Differences that would be invisible to a purchaser working from memory become immediately apparent when the designs are juxtaposed. The method creates exactly the conditions for the feature-by-feature difference-hunting that Gorham’s holistic test was designed to prevent.

D. The Connection to Utility Patent Methodology

The preference for side-by-side comparison likely reflects, at least in part, an importation of analytical habits from utility patent practice. In utility patent infringement, the standard methodology involves a claim chart that maps each element of the asserted claim against corresponding elements of the accused product, analyzing them one by one. This element-by-element approach makes sense for utility patents because the claim defines a set of discrete functional elements, and infringement turns on whether each element is present.

Design patents protect something fundamentally different: the ornamental impression created by the design as a whole. The “claim” of a design patent is a visual whole, and the infringement question is whether that whole is replicated closely enough to deceive a purchaser. Importing the element-by-element methodology of utility patent practice into design patent analysis converts a gestalt comparison into a feature inventory, destroying the very thing the test is supposed to evaluate.

This critique is not a categorical rule against placing two designs in view at once. A tribunal adjudicating infringement will often see the patented and accused designs together, whether through demonstrative exhibits, expert testimony, or appellate review. The point is narrower. The conditions under which a court views the designs do not, and should not, define the operative legal standard. The ordinary observer the law cares about is the one who encounters products in the marketplace, with the attention and memory conditions that purchasing actually involves. When courts allow litigation presentation to set the test, they substitute the convenience of how designs reach a courtroom for the inquiry Gorham requires.

VI. The Drift in Application: Jacki Easlick, LLC v. CJ Emerald

The preceding Parts traced the drift doctrinally. This Part traces its application through a single case, from the district court to the Federal Circuit. In Jacki Easlick, LLC v. CJ Emerald, the district court denied a preliminary injunction in a design patent case, and on August 14, 2026, the Federal Circuit affirmed in a nonprecedential opinion.79 The district court opinion contains, in a single decision, nearly every element of the drift this Article has described: the interchangeable use of “substantially the same” and “substantially similar,” the “plainly dissimilar” threshold, side-by-side comparison as the default method, and a ruling made without any prior art comparison. And because the district court handled functionality correctly, unlike the court in Range of Motion, the case presents the reframing and methodology problems separately from the functionality problem. The Federal Circuit’s affirmance then shows how those problems fare on appeal when the rule is stated correctly.

A. The District Court’s Opinion

Jacki Easlick invented the TOTE HANGER®, a hook for hanging handbags on closet rods, and protected its design with U.S. Design Patent No. D695,526. Her companies sued sixty-seven online sellers in the Western District of Pennsylvania and obtained an ex parte temporary restraining order against all of them. At the preliminary injunction stage, only one defendant appeared and opposed the motion: AccEncyc US, an online marketplace seller. The district court granted the preliminary injunction as to ten defendants who did not oppose it and denied the injunction as to AccEncyc, holding that the plaintiffs had shown neither a likelihood of success on infringement nor irreparable harm.80 Four aspects of the court’s infringement analysis deserve attention, and each illustrates a problem discussed earlier in this Article.

First, the language. The court stated the rule correctly, quoting Gorham’s “substantially the same” formulation in full.81 One page later, it concluded that the plaintiffs had not “shown a likelihood of substantial similarity in the claimed design and the accused design,”82 and it ended the analysis by concluding that they had not shown “a likelihood that the accused design is substantially the same as their claimed design.”83 Both phrases appear as statements of the governing standard in the same short analysis and are treated as equivalents. This is Category 2 interchangeable treatment, and it appears in the opinion the Federal Circuit reviewed, just as it appears in the Range of Motion panel opinions. The reconsideration opinion repeated it, describing what the plaintiffs had failed to show as “a likelihood of substantial similarly [sic] between the two designs.”84

Second, the “plainly dissimilar” screen. The court wrote that “[a] patentee cannot meet its burden if the claimed and accused designs are ‘plainly dissimilar.’”85 Its operative finding was that “there is a substantial question of whether the claimed and accused designs are plainly dissimilar.”86 Read carefully, that sentence defeats the patentee with uncertainty about dissimilarity. Under Egyptian Goddess, the exception applies only when dissimilarity is plain. A “substantial question” whether the designs are plainly dissimilar means dissimilarity is not plain, and that is exactly the situation in which Egyptian Goddess says the analysis “will benefit from a comparison of the claimed and accused designs with the prior art.”87 The court instead treated doubt about the screen as grounds for denying relief. On reconsideration, the court confirmed that this was why it had set the prior art aside: it “did not analyze the prior art because it determined that a substantial question exists as to whether the Tote Hanger and the Accused Product are ‘plainly dissimilar.’”88

Third, the comparison method. The court stated “[t]he Court begins its infringement analysis by conducting a side-by-side study of the Design Patent (on the left) and the Accused Product (on the right),” citing Crocs.89 The citation was accurate. Crocs itself states that the proper comparison “requires a side-by-side view” of the claimed design and the accused product, so the method on display here is not a departure from Federal Circuit guidance but an application of it.90 What follows is a list of differences: the “corkscrew-like center” of the patented design against the “laterally bent center” of the accused product; a U-shaped bottom hook joined at about forty-five degrees against a crescent-shaped hook joined at twenty to thirty degrees; “sphere-shaped caps” against “tube-like caps”; a flared tip against a plain one.91 The court then offered its own test: “Even if the pictures of the Tote Hanger and the Accused Product were mixed randomly, an ordinary observer could quickly restore them to their original order upon viewing the center of the parties’ handbag hooks.”92 The image is borrowed from Crocs, where the Federal Circuit observed that if the claimed and accused designs “were arrayed in matching colors and mixed up randomly, this court is not confident that an ordinary observer could properly restore them to their original order,” a point offered to show that deception was likely.93

The district court kept the image and inverted its office: in Crocs, difficulty sorting evidenced infringement; in Easlick, ease of sorting became a test for non-infringement. Either way, it is a sorting test performed with both designs in view. It is the opposite of the evidence Gorham relied on: the testimony that seven of ten silverware customers, seeing the accused pattern in a different store, would take it for the patented one.94 Any two designs that are not exact copies usually can be sorted when both are in view. That is simply what it means for two designs not to be identical. Gorham uses the word “substantially” precisely because the ability to tell two designs apart under close inspection is not the test. A court that measures infringement by whether the designs can be sorted when placed together is not applying a stricter version of the ordinary observer test. It is applying a different test, under conditions purchasers rarely experience.

Fourth, the prior art. The court recited that the ordinary observer “is familiar with prior art” and quoted Crocs on how the prior art directs the observer’s attention.95 The preliminary injunction opinion then considered neither kind of prior art. It did not identify the background prior art that would define the observer, and it placed no comparative prior art beside the claimed and accused designs. The omission mattered on this record. The plaintiffs’ central contention was that the closest prior art hooks lacked the 90-degree offset between the rod hook and the bag hook, the signature feature of the patented design, and that the accused product copied exactly that feature.96 Whether that contention would ultimately succeed or not, it is the kind of contention the three-way comparison exists to test. And as Part II.C explained, the background prior art defines the observer. A court that sets aside the background prior art has not simplified the analysis. It has removed the observer before the comparison begins.

The court returned to the prior art on reconsideration, in the alternative. It found that the plaintiffs’ prior art “differs from the claimed design in many respects,” including “the shapes of the hooks, the finished ends, flared tips, and the downward bent center of the hook,” and that these were “some of the same ornamental dissimilarities” that separated the claimed and accused designs.97 With “the attention of the hypothetical ordinary observer still drawn to these significant differences,” it concluded, applying the prior art would not have changed the result.98 That is a three-way comparison of a kind, but it runs in one direction. It identifies departures from the prior art that the accused design does not share. Its prior art discussion does not mention the departure the plaintiffs said the accused design did share, the 90-degree offset. Under Egyptian Goddess, copying a feature that “departs conspicuously from the prior art” makes the accused design “more likely to be regarded as deceptively similar.”99 An analysis that counts only the departures the accused design lacks has used the prior art for one purpose only. Part VI.C returns to that question.

B. The Federal Circuit’s Decision

The Federal Circuit affirmed.100 Its opinion states the rule correctly. It quotes Gorham’s “substantially the same” formulation in full, and it approved the district court’s focus on whether an ordinary observer “would be deceived into thinking” that the accused design was the same as the patented design.101 It does not state the rule as “substantially similar.” Like the concurrence in the denial of rehearing in Range of Motion, it belongs in Category 1. And it affirmed the ruling that Part VI.A describes.

On the method, the panel reproduced the district court’s side-by-side comparison and agreed that the district court had properly “recited the significant differences between the ornamental features of the two designs.”102 It quoted the district court’s sorting test in support.103 It held that “[c]onsidering distinguishing features” in this way “is not legal error,” and it added that “it is hard to imagine how a court could reasonably conduct this analysis without pointing such features out.”104 Its authority was International Seaway: “The mandated overall comparison is a comparison taking into account significant differences between the two designs . . . .”105

This Article agrees that considering distinguishing features is not legal error. Gorham says that differences are to be considered, and International Seaway says they are to be taken into account. The question is what the court asks of them. Here the district court asked whether an ordinary observer, with both designs in view, could “quickly restore them to their original order.”106 The ability to sort two designs placed side by side is a finding about whether they can be told apart. It is not a finding about deception under the conditions in which purchasers encounter products. The panel treated the first as support for the second, and the opinion does not explain why the one answers the other. That is the substitution Part V describes, now affirmed on appeal. Taking differences into account is what Gorham requires. Making the ability to find them the measure of infringement is not.

On the prior art, the panel quoted Ethicon: comparing the designs with the prior art “is beneficial only when the claimed and accused designs are not plainly dissimilar.”107 It did not address the reason the district court gave for skipping that comparison, which was a substantial question whether the designs were plainly dissimilar. As Part VI.A explained, a substantial question whether dissimilarity is plain means that it is not plain. The panel turned instead to the alternative ruling on reconsideration. It agreed “that the prior art submitted by Jacki Easlick differs from the claimed design in several respects” and held that the district court had not abused its discretion.108

On functionality, the panel affirmed the construction that Part VII discusses. The district court had excluded the two attached hooks and their vertical configuration as functional while keeping the shapes of the hooks, the flared tip, the 90-degree offset, and the spheres on the ends within the claim. The panel declined to reach the appellee’s argument that the offset was itself functional.109 On Schedule A practice, the panel noted the law professors’ amicus brief and stated that “[t]he propriety of the ‘Schedule A’ practice is not before us.”110

Two cautions apply. The decision is nonprecedential, and it reviewed a preliminary injunction ruling for abuse of discretion. It does not hold that a sorting test is the measure of infringement, or that a court may skip the prior art whenever there is a substantial question whether designs are plainly dissimilar. But the decision shows what the remedy proposed by the concurrence in Range of Motion looks like in practice. The concurrence said that errors in applying the screen should be corrected by reversal on appeal. Easlick was an appeal. The rule was stated correctly, the district court’s reliance on the screen went unaddressed, and the ruling was affirmed.

C. The One-Way Ratchet Question

The Easlick briefing also raised an issue that has run beneath this Article’s prior art discussion: whether the prior art comparison applies only to patentees. The amici stated the position plainly: the use of the prior art in the design patent infringement analysis “is a one-way ratchet” that “can be used to narrow the presumptive scope of a claim but cannot be used to broaden it.”111 On this view, the first-step comparison of the designs “in a vacuum” fixes a design patent’s “presumptive scope,” and the prior art enters, if at all, only at the second step and only to the accused infringer’s benefit.112 The appellee argued the same, contending that the plaintiffs’ reliance on the prior art was an improper attempt to expand their patent.113 The appellants responded that the three-way comparison can demonstrate infringement. They cited L.A. Gear, where the affirmance rested in part on the finding that nothing in the prior art suggested “the appearance of the claimed design as a visual whole”; Avia Group, which directed comparison of the patented design’s overall appearance against the prior art; and Crocs, which explained that the prior art draws the observer’s attention “to those aspects of the claimed design that differ from the prior art.”114

The response could have rested on more direct authority. Egyptian Goddess itself states the two-way principle: “If the accused design has copied a particular feature of the claimed design that departs conspicuously from the prior art, the accused design is naturally more likely to be regarded as deceptively similar to the claimed design, and thus infringing.”115 And the Federal Circuit reaffirmed the point in 2023, explaining that comparison prior art “can also cut the other way” by highlighting similarities between the claimed and accused designs.116

The ratchet metaphor assumes the very structure this Article has criticized: a two-step framework in which an abstract comparison fixes a baseline and the prior art then adjusts it. On Gorham’s framework, as Part II.C explained, the background prior art does not adjust anything. It defines the observer. An observer familiar with a field in which no design offsets the carrying hook ninety degrees from the rod hook will see that offset as the visual signature of the claimed design. When an accused product copies exactly the feature that separates the claimed design from everything else in the field, the prior art makes deception more likely, not less. Calling that a “broadening” of the claim confuses calibration with expansion. The claim’s scope has not changed; the patentee must still prove that the overall appearances would deceive. What the background prior art supplies is the knowledge of the eye through which the appearances are judged, and that knowledge can point in either direction. The case law already recognizes half of the principle: when the claimed design is close to the prior art, small differences between the accused and claimed designs “assume more importance.”117 The other half follows from the same logic, and Egyptian Goddess and Columbia Sportswear state it expressly. When the claimed design departs sharply from the prior art, and the accused design shares the departure, the shared departure assumes more importance. Nothing in the case law limits the observer’s knowledge to the defendant’s use, and L.A. Gear and Crocs confirm that the three-way comparison has supported findings of infringement as well as findings of non-infringement.

The ratchet does address a real concern. A patentee should not be able to turn distance from the prior art into ownership of a concept (here, the general idea of offset hooks) rather than of the particular claimed shape. The amici pressed exactly this point, observing that design patents “protect shapes, not concepts.”118 But that concern is handled directly by the shape-versus-concept line that Ethicon draws and that Part VII of this Article defends. The claim covers the particular ornamental expression shown in the drawings, and no amount of distance from the prior art turns it into a claim on a design concept.

The Federal Circuit did not decide the question in terms. It agreed that the prior art “differs from the claimed design in several respects,” and it did not discuss whether the accused design shared any of the claimed design’s departures from the prior art.119 The district court’s alternative analysis, as Part VI.A explained, counted only the departures the accused design lacked. The case was thus decided with the prior art working in one direction only, which is the direction the ratchet permits. Whether the three-way comparison may also support a finding of infringement is a question Egyptian Goddess and Columbia Sportswear answered, and Easlick did not revisit it.

D. The Schedule A Objection

The law professors’ amicus brief makes a broader argument that this Article should address, because it may be the strongest institutional defense of the current framework. Easlick, the amici showed, was a “Schedule A” case: a mass-defendant infringement action in which sixty-seven unrelated online sellers, selling by the plaintiffs’ own account at least four different kinds of products, were joined in a single sealed complaint, in tension with the Patent Act’s same-accused-product joinder requirement.120 Schedule A cases are rarely appealed (the amici counted this as only the third such appeal to reach the Federal Circuit) because the asset freeze pushes most defendants to settle or default regardless of the merits.121 Against that backdrop, the amici’s defense of the framework is simple: the particularized scrutiny that finally occurred at the preliminary injunction stage is what the system got right, and the first defendant to appear and contest the allegations won under it.122

The abuses the amici brief are real, and the scholarship describing them is persuasive.123 But consider what kind of abuses they are and what kind of remedies they require. Improper joinder is a § 299 problem. Generic, undifferentiated infringement allegations are a Rule 8 problem. Ex parte restraints granted at scale, asset freezes eighty times the accused sales, nominal bonds, and default judgments entered against products already found likely non-infringing are problems of ex parte practice, proportionality, and case management. These problems are procedural, and every remedy for them is procedural: joinder enforcement, pleading scrutiny, discipline in ex parte practice, and properly calibrated restraints and bonds. The amici’s own brief documents these failures.124 None of them requires a change to the substantive test for design patent infringement. And in Easlick itself, the substantive screen did not fix any of the abuses. The temporary restraining order issued and the assets froze before any dissimilarity analysis occurred, and the default judgments issued after the finding of likely non-infringement.

As a cure for Schedule A abuse, the plainly dissimilar screen comes too late to prevent the harm of ex parte process and does nothing to stop the settlements and defaults that drive the model. What it reliably does is cut short the analysis in every design patent case, including the ordinary two-party case with a meritorious claim and a defendant fully able to defend itself. Changing the infringement standard to police litigation abuse punishes meritorious patentees for the conduct of abusive ones and leaves the abuse in place.

The clearest lesson of the Easlick record in fact points the other way. When a defendant finally appeared, and the motion was contested, the district court had all the tools it needed to deny unwarranted relief. It construed the claim, weighed the evidence, and found the showing insufficient on two independent grounds. The problem with Schedule A practice is that defendants so rarely can appear, and that is a problem of process with answers in process.

The Federal Circuit’s decision is consistent with that view. The court noted the amici’s argument and set it aside, stating that the propriety of the practice was not before it.125 It then decided the appeal under the ordinary standards for preliminary injunctions, and it held that the ex parte restraining order against sixty-seven defendants “does not restrain the district court’s separate analysis of a motion for a preliminary injunction.”126 Nothing in the decision relied on the Schedule A posture to justify the infringement analysis.

E. Wang v. Schedule A Defendants

Events since the argument underscore that lesson, and they show the drift still spreading. In April 2026, the same district judge decided another Schedule A design patent case, Wang v. Schedule A Defendants, in which the plaintiff had obtained an ex parte temporary restraining order against a schedule of online sellers of an ergonomic string-trimmer handle. As in Easlick, exactly one defendant appeared and contested the motion, and the contested motion failed: the court denied the preliminary injunction and dissolved the restraining order, this time finding the asserted design patents vulnerable to anticipation.127 The adversarial process, once engaged, worked. But the opinion shows something more. Applying the ordinary observer test to the validity question, the court held that the only difference between the claimed designs and the prior art, an arm attached at a “slightly angled” rather than “a perfect 90 degree angle,” was “not a meaningful difference,” and that the designs were “substantially the same.”128

The governing standard was correct. International Seaway makes the ordinary observer test applicable to both anticipation and infringement, and the Wang court applied it through Gorham’s deception formulation.129 But the court also assessed the prior art in the vocabulary of the Egyptian Goddess screen, block-quoting Range of Motion’s infringement framework and finding the prior art not “sufficiently distinct” and “plainly dissimilar.”130 That language provides no independent standard for anticipation. In its infringement setting, as the decisions traced in Part IV apply it, the screen lets courts end the infringement analysis without the prior art comparison that follows when designs are not plainly dissimilar. In an anticipation case, the prior art reference is the very object of the required comparison, so the screen has no threshold to guard. The phrase may state the converse of substantial sameness, but it cannot do the work it performs in infringement. The opinion even quotes § 103, the obviousness statute, on its way to an anticipation holding.131 Wang ultimately made the required substantial sameness determination. The drift lies in treating an infringement shortcut as part of the anticipation inquiry at all.132 That is the mechanism this Article has traced since Tomkinson: a phrase detaches from the rule it was written to serve and begins conducting analyses it was never designed for, without the court remarking on the substitution.

Nothing in this Article prejudges the correct outcome of Easlick under a restored Gorham framework. The accused hook may or may not deceive an ordinary observer familiar with the handbag-hanger field, given the attention a purchaser of a low-cost closet accessory usually gives. The point is narrower. That question, the deception question as purchasers would face it, is the one the district court never reached. The court used the word “deceived,” but it answered the question by asking whether the designs could be told apart when studied side by side, described its answer in the language of “substantial similarity” and substantial sameness interchangeably, and ended the case on a “substantial question” about a dissimilarity screen whose own conditions were not met. One preliminary injunction opinion, and its affirmance, measure how completely the drift has become the doctrine.

VII. The Functionality Problem: What “Factoring Out” Should and Should Not Mean

The third major error, compounding the linguistic and methodological drift, concerns the treatment of functional features in design patent claim construction. Courts have increasingly used claim construction to strip design patents of features deemed “functional,” dramatically narrowing claim scope before the infringement comparison ever begins.

A. The Correct Principle

The principle requires a distinction that courts may miss. When a design contains elements that serve functions, what must be factored out of the infringement analysis is the functional necessity of the elements’ existence and arrangement, not the ornamental shapes of the elements themselves. The fact that a tool needs a handle, or that a hand tool’s components must be arranged in a particular configuration to work, cannot support a finding of infringement. But the particular ornamental shape the designer gave the handle remains fully within the claim’s protection. The existence goes out. The ornamental shapes stay in.

Richardson v. Stanley Works illustrates the distinction.133 The case involved a multi-function carpentry tool that combined a hammer, a stud-climbing tool, and crowbar. The district court construed the claim to distinguish the functional aspects, the fact that the tool has a hammer head, a crowbar, and jaws arranged in a particular working configuration, from the ornamental aspects, which were the particular shapes and appearances given to those elements. The Federal Circuit approved, stating “[t]he district court here properly factored out the functional aspects of Richardson’s design as part of its claim construction.”134 The infringement comparison then proceeded on the ornamental aspects.

The correct approach, then, is not to excise functional elements from the claim. It is to recognize that the claim protects the ornamental appearance of the design as a whole, including the ornamental aspects of functional elements, while the mere presence and arrangement of functionally necessary elements cannot alone establish infringement.

The terminology itself may be part of the problem. The phrase “factoring out” suggests removal, subtraction, elimination. It invites courts to treat functional elements as if they were stricken from the claim entirely, leaving only a residue of purely decorative flourishes.

A better term might be “account for.” Courts should account for the functional aspects of a design by recognizing that functional necessity cannot support a finding of infringement, while keeping the ornamental appearance of every element, functional or not, within the scope of the protected design.

If Richardson’s “factor out” language is kept, then another approach is to refer to the factoring out of “concepts” rather than design “elements.” That usage has support in the case law, at the construction step where the filtering belongs. OddzOn approved a claim construction limiting the patent “to its overall ornamental visual impression, rather than to the broader general design concept of a rocket-like tossing ball.”135 The Easlick district court applied the same principle, stating “[b]ecause there is a functional aspect, the concept of two attached hooks must be excluded from the scope of the claimed design.”136 The Federal Circuit affirmed that construction.137 In each instance, the thing identified as outside the claim was the concept. The ornamental appearance of the elements embodying the concept remained within it. A court that factors out the concept has nothing left to subtract from the design, because the concept was never part of the claimed appearance to begin with.

This is not a trivial point about word choice. As Parts III and IV of this Article demonstrate, the vocabulary courts use shapes the analysis they conduct. A court that believes its task is to “factor out” functional elements will look for things to remove. A court that accounts for functionality, or that keeps the traditional phrase but factors out only the concept, will preserve the design as a whole while assigning appropriate weight to the sources of its visual character.

B. What Ethicon and Sport Dimension Got Right

In Ethicon Endo-Surgery, Inc. v. Covidien, Inc., the district court had construed the design patent claims at such a high level of abstraction that the ornamental shapes of functional elements disappeared from the analysis. The Federal Circuit reversed, explaining that the district court had performed its functionality analysis “at too high a level of abstraction, focusing on the general concepts of an open trigger, torque knob, and activation button rather than the ornamental designs adorning those elements.”138 The court returned to the point at the infringement stage, emphasizing that “the trigger has a particular curved design, the torque knob has a particular flat-front shape, and the activation button has a particular rounded appearance.”139 Those particular appearances were the protected design.

Sport Dimension, Inc. v. Coleman Co. reinforced the point, though less cleanly. The Federal Circuit held that the district court erred by eliminating the armbands and side torso tapering from the claim entirely; the elements remained in the claim “to the extent that they contribute to the overall ornamentation of the design.” The court simultaneously instructed that the factfinder “should not focus on the particular designs of these elements when determining infringement, but rather focus on what these elements contribute to the design’s overall ornamentation.”140 The instruction’s first half preserves the elements; its second half points away from their particular designs. This Article reads Ethicon as supplying the governing principle, the preservation of particular ornamental shapes within the claimed whole, and reads Sport Dimension’s instruction as a caution against treating any single element’s design as dispositive, not as license to disregard it.

C. How Range of Motion Went Wrong

The district court in Range of Motion applied the functionality framework in a way that, on this Article’s reading, cannot be squared with Ethicon and Sport Dimension. At claim construction, the court identified a series of elements in the patented arm-therapy device as functional, including the arched shape, the leg lengths, and the roller carriage, and effectively removed their ornamental aspects from the claim. What remained for comparison were the aspects of the design the court deemed “largely ornamental.”141 The Federal Circuit majority saw no conflict: it affirmed the construction and held that Sport Dimension permits the functionality factors to guide claim construction. Section D explains why this Article reads Ethicon and Sport Dimension differently. Dissenting from the denial of rehearing, Chief Judge Moore identified the uncertainty that follows: once a judge determines that an aspect of a design is functional, “it is entirely unclear what happens next.”142 The author’s companion materials on functionality propose an answer.143

The error is structural. Once the ornamental shapes of functional elements are excluded at claim construction, the infringement comparison operates on a skeleton of the actual design. The overall visual impression, the very thing Gorham protects, has been dismantled before the comparison begins. The accused product is then compared not against the design as the ordinary observer would perceive it, but against a judicially edited abstraction.

This sequencing compounds the other errors this Article describes. A claim, already narrowed by aggressive functionality filtering, enters an infringement analysis already tilted by similarity framing and side-by-side methodology. Each step subtracts from the patentee’s position, and the subtractions multiply.

D. Why the Court Fell Into the Error: The Overuse of the Berry Sterling/PHG Functionality Factors

The Range of Motion district court did not invent its approach. It applied a set of factors articulated by the Federal Circuit in Berry Sterling Corp. v. Pescor Plastics, Inc. and elaborated in PHG Technologies, LLC v. St. John Cos. for a different purpose: determining whether a design patent is invalid because its design is dictated by function.144 The factors include whether the protected design represents the best design, whether alternative designs would adversely affect the article’s utility, whether there are concomitant utility patents, whether advertising touts particular features as having specific utility, and whether the design has elements clearly not dictated by function. Sport Dimension then stated that these validity-oriented factors “may serve as a useful guide for claim construction functionality as well.”145

Start with the question each inquiry asks. The validity inquiry asks whether the design as a whole is dictated by function, such that the patent should not exist at all. That is a demanding standard, and the Berry Sterling/PHG factors were developed to answer it.146 The claim construction inquiry, by contrast, asks how to describe the protected design so that the factfinder compares the right things. Importing a whole-design invalidity framework into an element-by-element claim construction exercise converts a shield against wholly functional patents into a scalpel for trimming ornamental content out of valid ones.

The concurrence in the denial of rehearing in Range of Motion draws the same distinction. It explains that “functionality plays different roles in invalidity and claim construction.” For validity, the question is whether the overall appearance is dictated by function. At claim construction, the task is “identifying the design’s functional elements in an effort to guide the jury as to the scope of the claim.”147 This Article agrees. But if the two inquiries are distinct, a tool built to answer one is not for that reason suited to answer the other. The distinction the concurrence draws is a reason to return the Berry Sterling/PHG factors to the validity inquiry, not a reason to keep them at claim construction.

When a validity-oriented framework is applied element by element at claim construction, the practical consequence is exactly the one Ethicon and Sport Dimension warned against: elements that serve any identifiable function are treated as presumptively excludable, their ornamental shapes disappear from the comparison, and the claim shrinks to whatever the court regards as purely decorative. Design patents protect the ornamental appearance of useful articles; nearly every element of a useful article serves some function. A framework that filters elements by asking whether they are functional will, in most cases, filter out most of the design, making infringement difficult to prove.148

That is the analytical path the Range of Motion district court followed. The court organized its claim construction around the PHG factors, weighed the presence of a concomitant utility patent and advertising that touted utilitarian advantages, and concluded that the arched shape and other elements were functional, with the consequences described above. The court treated the factor test as the governing framework for construction rather than as a validity screen, and the Federal Circuit affirmed the resulting construction. Under Ethicon and Sport Dimension, the proper course was to acknowledge the functions those elements serve while preserving their particular ornamental shapes in the claim. Under the PHG factors as the district court applied them, the shapes themselves were treated as functional and removed. A utility patent covering a device’s mechanism says nothing about whether the particular visual form given to that mechanism is ornamental. Nor does advertising touting leverage and whole-body usability show that the visual form is dictated by function.149

The lineage of the factor test confirms the mismatch. Berry Sterling announced the factors in the context of assessing whether a design was invalid as dictated by function. PHG applied them to the same validity question.150 Ethicon discussed them in determining whether a design was primarily functional and therefore invalid.151 Not until Sport Dimension did the Federal Circuit suggest the factors could guide claim construction, and it did so in a single sentence without analysis.152 The suggestion has since hardened into practice, with district courts treating the factor test as the standard framework for functionality at claim construction. The result is a doctrinal short circuit: a test designed to answer whether a patent should exist is being used to decide how much of the patent’s design survives into the infringement comparison.

None of this is to say that functionality has no role at claim construction. Sport Dimension itself shows the appropriate, modest role: a court may note that certain elements serve functions and instruct the factfinder that the functional utility of those elements cannot support an infringement claim, while preserving the elements’ ornamental appearance within the claim. What the court may not do is use a validity framework to excise the ornamental shapes of functional elements from the claim. The distinction tracks the line drawn in Part VII.A: the concept, meaning the elements’ existence and arrangement, goes out; the ornamental shapes stay in. The Berry Sterling/PHG factors were developed for validity analysis. Sport Dimension stated that they “may serve as a useful guide” at claim construction, and the Range of Motion majority relied on that statement; this Article contends that the guidance has been converted from a modest aid into an element-filtering framework that eliminates precisely what Ethicon and Sport Dimension preserve. This Article proposes to return the factor test to the validity context for which it was developed; on the reading advanced here, that is the most direct way to prevent the error’s repetition. Courts should account for, rather than eliminate, the functional aspects of the design.153

VIII. The Seventh Amendment Problem: Functionality Is Not Always Claim Construction

The functionality problem described in Part VII has a constitutional dimension that has received insufficient attention. The practice of resolving functionality questions at claim construction rests on the premise that functionality is a question of law for the court. That premise derives from Markman v. Westview Instruments, Inc., which held that claim construction in utility patent cases is a matter of law.154 But the extension of Markman to design patent functionality determinations is far from obvious. No controlling authority holds that these determinations belong to a jury; Range of Motion reiterates that claim construction, including its functionality component, is for the court. The court divided on the point when it denied rehearing en banc. The concurrence would keep the determination with the court at claim construction, and the dissent would give it to the jury.155 This Part therefore advances a constitutional argument rather than a description of settled law: that, in its current form, the practice works a significant transfer of fact-finding authority from juries to judges, and that the transfer raises Seventh Amendment concerns Markman, a utility patent claim construction decision, does not resolve.

A. What Markman Authorized

Markman held that “the construction of a patent, including terms of art within its claim, is exclusively within the province of the court.”156 The Court reasoned that judges are better equipped than juries to construe written instruments, analogizing patent claims to statutes and contracts. The decision rested heavily on the documentary character of the task: claim construction involves reading a written document and determining the meaning of its terms, an exercise in textual interpretation for which “judges, not juries, are the better suited.”157 The Court emphasized the “importance of uniformity in the treatment of a given patent” and the value of appellate review of legal determinations in achieving it.158 Markman also left two questions open. Because precedent supported classifying claim construction as a question for the court, the Court did not decide “the extent to which the Seventh Amendment can be said to have crystallized a law/fact distinction,” or whether post-1791 precedent classifying an issue as one of fact would trigger the Seventh Amendment’s protections.159 Neither question has been answered for design patents.

B. What Courts Are Actually Doing

Design patents are different. A design patent’s claim is a drawing. Egyptian Goddess itself recognized the mismatch, cautioning that “the preferable course ordinarily will be for a district court not to attempt to ‘construe’ a design patent claim by providing a detailed verbal description of the claimed design.”160 The en banc court warned that verbal descriptions risk placing “undue emphasis on particular features of the design” rather than the “design as a whole.”161 The Supreme Court made a similar point in 1886, observing that a claimed design “is better represented by the photographic illustration than it could be by any description.”162 The Supreme Court has likewise recognized, in the trademark context, that questions turning on the perceptions of ordinary consumers are quintessentially factual: “[w]hen the relevant question is how an ordinary person or community would make an assessment, the jury is generally the decisionmaker that ought to provide the fact-intensive answer.”163

Consider what a court actually does when it determines at claim construction that the arched shape of a therapy device is “functional.” The court examines the product, considers how it works, evaluates whether the shape contributes to its operation, weighs evidence such as utility patents and advertising, and reaches a conclusion about the relationship between form and function in a physical artifact. Nothing about that inquiry involves interpreting the meaning of words in a written instrument. It is an evaluative judgment about a tangible object, informed by evidence, of exactly the kind juries have historically made.

That has the character of factfinding, not of textual interpretation, and labeling it “claim construction” does not change its nature. And when the finding strips ornamental shapes from the claim before the jury ever sees the case, or supports summary judgment on the ground that what remains is “plainly dissimilar,” the label operates to withdraw from the jury a determination that, on the argument advanced here, the Seventh Amendment commits to it.

The concurrence in the denial of rehearing gives two reasons for keeping the determination with the court. The first is Markman’s concern for “uniformity in the treatment of a given patent,” which the concurrence says “applies equally to both design patents and utility patents.”164 The second is that defining a design patent’s scope “involves the interpretation of a legal document and determining the bounds of a property right,” like the description in a deed.165 Both reasons fit the parts of design patent claim construction that do interpret a document, such as the effect of broken lines, of claim language naming the article, and of statements made during prosecution. They fit less well the question whether a particular shape could have been different and still worked. That question is answered from evidence outside the patent, including utility patents, advertising, and testimony about alternative designs. As the dissent observed, none of the answers to the Sport Dimension/PHG questions “can be found in the asserted design patent itself.”166 Uniformity is a reason to give judges the interpretation of documents. It is not a reason to give them the weighing of evidence about a physical object.

The concurrence also relies on Teva, under which the construction of a claim remains a question of law even though “subsidiary factfinding is sometimes necessary.”167 Teva addressed a task that consists mainly of reading a document, with factfinding as the exception. In design patent claim construction the proportions are reversed. The dissent put it directly: “Design patent claim construction is nearly all fact.”168

C. The “Purely Functional” Line

Egyptian Goddess said that a trial court can usefully guide the factfinder by “distinguishing between those features of the claimed design that are ornamental and those that are purely functional.”169 The authorization is best read in light of the principles developed in Part VII: a court may identify aspects of a design whose presence and arrangement are dictated by function. It may instruct the jury that functional necessity cannot support infringement. What the authorization does not support is judicial removal of ornamental shapes from the claim on the ground that the elements bearing those shapes serve functions. When a court goes further and decides that particular shapes are themselves functional, contested evaluative questions about real-world objects, it crosses from construction into factfinding. The Range of Motion district court crossed that line when it determined that the arched shape was functional notwithstanding record evidence of alternative designs. That determination then controlled the summary judgment analysis, with the court concluding that the remaining protected elements “appear to be largely ornamental.”170

The point is not that a judge can never resolve functionality. Where the facts are undisputed and no reasonable jury could find otherwise, summary adjudication is available for functionality as for any other factual question. The point is that functionality determinations that turn on disputed evidence, alternative designs, the significance of utility patents, the import of advertising claims, and the visual contribution of particular shapes are, on the argument advanced here, jury questions, and routing them through “claim construction” effectively withdraws a dispositive factual issue from the jury.

D. The Inconsistency with Validity and Trade Dress

The current allocation is also internally inconsistent. When functionality is raised as an invalidity defense, whether the design as a whole is dictated by function, the question is treated as factual, and the Berry Sterling/PHG factors guide the factfinder.171 In trade dress law, functionality is likewise a question of fact.172 Yet when the same kind of question arises at claim construction in a design patent case, whether particular elements or shapes are functional, it is suddenly a question of law for the court. The same kinds of evidence and the same evaluative judgment go to a different decisionmaker depending on the procedural label attached. Nothing in Markman compels that inconsistency, and the Seventh Amendment counsels strongly against it.

The concurrence in the denial of rehearing answers that the two inquiries differ. Validity asks whether the overall appearance is dictated by function, while claim construction identifies functional elements to guide the jury.173 That is correct as to their subjects, and Part VII.D relies on the same distinction. But a difference in subject does not change the kind of question. Whether the form of a whole design, or of one element, is dictated by function is answered in both settings by weighing evidence about a physical object, and since Sport Dimension, by consulting the same factors. As the dissent put it, the question answered by “the identical factors” becomes “a question of law for the judge” only “because we place it under the claim construction umbrella.”174

E. The Connection to the Berry Sterling/PHG Problem

The constitutional and doctrinal problems identified in Part VII are connected. The migration of the Berry Sterling/PHG validity factors into claim construction is what converted functionality from a jury question into a judge question. When the factors lived exclusively in validity analysis, the factfinder applied them. When Sport Dimension suggested they could guide claim construction, the same factual inquiry moved to the bench. The doctrinal correction proposed in Part VII, returning the factor test to validity analysis where it belongs, would simultaneously cure most of the constitutional problem, because functionality determinations would again be made by the factfinder in the context of an invalidity defense, subject to the ordinary standards for summary adjudication. This constitutional analysis reinforces the concern raised in Part VII about importing validity frameworks into claim construction. The two errors are one error, viewed from different angles.

IX. The Summary Judgment Problem

The doctrinal errors documented in the preceding Parts converge at summary judgment, where their combined effect is to remove design patent infringement from the jury with a frequency that would be startling in any other area of law involving questions of fact.

Design patent infringement is a question of fact.175 The ultimate inquiry, whether an ordinary observer would be deceived into purchasing one design supposing it to be the other, turns on human perception, marketplace conditions, and the visual impact of designs on a hypothetical purchaser. These are exactly the kinds of assessments that juries exist to make. Whether two designs would deceive a purchaser is often a question on which reasonable minds can differ, which is the classic definition of a jury question under familiar summary judgment criteria.

Under ordinary summary judgment standards, a court must view the evidence in the light most favorable to the non-moving party and may grant judgment only if no reasonable jury could find for that party.176 A substantial line of decisions, however, shows courts granting summary judgment of non-infringement, using the “plainly dissimilar” screen to resolve as a matter of law a question that is supposed to be factual.177 In Ethicon, summary judgment of non-infringement was affirmed following a claim construction that abstracted away the ornamental shapes of functional elements.178 In North Star, summary judgment was affirmed under the “plainly dissimilar” framework.179 In Range of Motion, the district court granted summary judgment even while acknowledging that the designs “share a ‘broad design concept’” and that “at a conceptual level they look quite similar.”180 And as Part VI shows, the same gatekeeping dynamic now operates at the preliminary injunction stage as well, and the Federal Circuit has affirmed it there. The dynamic has reached the pleadings too: courts now apply the screen on motions to dismiss, ending cases before any record exists.181 As Chief Judge Moore observed in dissent, the court’s errant language “has and will result in the near-complete removal of the jury from its fact-finding role in design patent infringement.”182 Dissenting from the denial of rehearing, she was blunter: “We have messed up design patent infringement and essentially eliminated any role for the jury over what are quintessential jury-type fact questions.”183

The mechanism is now familiar. The “plainly dissimilar” screen invites the court to catalog differences. Side-by-side comparison makes differences salient. Functionality filtering removes ornamental content from the claim, making the remaining differences loom larger still. Each doctrine independently tilts toward non-infringement; in combination, they have made summary judgment a common disposition rather than an exceptional one across the decisions collected in this Article. A legal framework under which judges repeatedly resolve a question of fact against the party entitled to a jury determination is a framework in tension with the Seventh Amendment, whatever the verbal formulation of the governing test.

The concurrence in the denial of rehearing defended summary judgment on grounds of judicial economy, in cases where “no reasonable jury could find the patent infringed.”184 This Article agrees that summary judgment is proper in those cases. The objection is to a screen that decides which cases those are without a standard, and without the prior art that defines the observer through whose eyes a jury would be asked to look.

X. The Path Forward

The errors documented in this Article share a common origin: the gradual replacement of Gorham’s holistic, deception-based inquiry with a decomposed, difference-focused methodology. The corrections follow from the diagnosis. The Federal Circuit declined to take up these questions en banc in Range of Motion, with four of the eleven participating judges voting to rehear the case. It should take them up in an appropriate future case and reaffirm that the test for design patent infringement, announced by the Supreme Court in Gorham and confirmed in Egyptian Goddess as “the sole test,” is whether the designs are substantially the same in the eye of the ordinary observer.185

First, courts should restore Gorham’s verbal formulation. The Supreme Court’s standard is “substantially the same,” not “substantially similar,” and certainly not “plainly dissimilar.” Words matter. The formulation of the test directs the factfinder’s attention, and only the sameness formulation directs attention to the holistic visual impression that Gorham protects.

Second, courts should eliminate the “plainly dissimilar” threshold as a freestanding gatekeeping device. As Chief Judge Moore recommended, courts performing the infringement analysis should “always” compare the claimed and accused designs in light of the prior art, with no special exception for designs a judge regards as plainly dissimilar.186 At a minimum, every infringement analysis must be conducted through the eyes of an observer familiar with the background prior art. The background prior art defines the ordinary observer, and an analysis that proceeds without it is an analysis conducted through the eyes of no one.

Third, courts should conduct the comparison under marketplace conditions rather than laboratory conditions. A side-by-side display may assist the factfinder, and current precedent endorses such a display. The needed correction is narrower and more fundamental: the ability to distinguish two designs when juxtaposed should not be treated as an answer to the marketplace deception question. The older authority understood the difference between a display and a test, and it preserved the perspective of a purchaser who encounters the designs separately, with the attention and memory that real purchasing involves.

Fourth, courts should confine functionality filtering to its proper role. The functional necessity of an element’s existence and arrangement is accounted for; the ornamental shape of every element remains in the claim. The Berry Sterling/PHG factors, as Part VII.D explains, are a poor tool for that inquiry and should return to the validity context for which they were developed. Where functionality is genuinely disputed, it is a question for the factfinder.

Fifth, courts should restore the jury to its constitutional role. Design patent infringement is a question of fact. Summary judgment remains available where no reasonable jury could find deception, but the current practice, in which the combination of dissimilarity screening, side-by-side comparison, and functionality filtering makes judgment as a matter of law the norm, inverts the constitutional allocation of decision-making authority.

XI. Conclusion

Gorham’s “substantially the same” formulation, measured by marketplace deception and applied through the eyes of an ordinary observer familiar with the prior art, is a precise instrument. Each element does work. The substitution of “substantially similar,” the inversion to “plainly dissimilar,” the juxtaposition of designs side by side, and the judicial excision of ornamental content under the banner of functionality have together replaced that instrument with a different one, a difference-detection protocol that no longer measures what the law is supposed to measure.

The consequences are not abstract. Cases that belong with juries now end at summary judgment. As Chief Judge Moore wrote in dissenting from the denial of rehearing, under the correct test, cases like Range of Motion, North Star, and Ethicon “would all go to juries.”187

The Appendix to this Article catalogs representative decisions spanning the years 1871–2026. It organizes them into three categories: decisions faithful to Gorham, decisions that treat sameness and similarity as interchangeable, and decisions that have reformulated the test outright. The taxonomy demonstrates that the drift is not the product of any single decision or any single court. It accumulated, one borrowed phrase at a time, across more than a century.

The difference between “substantially the same” and “substantially similar” or “plainly dissimilar” is the difference between asking whether a purchaser would be deceived and asking whether a judge can tell two pictures apart. The first is the question Gorham announced. The second is the question that, as this Article has shown, has come to govern in practice. This Article has traced how that happened and how it can be undone. No statute and no Supreme Court decision compels the current framework; the correction is the Federal Circuit’s to make. What remains is for the courts to undo it.

Appendix

Representative Cases Using “Substantially Similar” in Design Patent Infringement Analysis

This appendix organizes representative decisions into the three categories developed in Part III.D. Category 1 comprises decisions faithful to Gorham, stating the rule as “substantially the same” (or a close cognate) and using “similar” only descriptively. Category 2 comprises decisions treating “substantially the same” and “substantially similar” as interchangeable statements of the rule. Category 3 comprises decisions that affirmatively restate the rule as “substantially similar.”

Case Court & Year Operative Language / Notes
Category 1: Descriptive use consistent with Gorham (rule stated as “substantially the same”; “similar” used only descriptively)
Gorham Co. v. White, 81 U.S. (14 Wall.) 511 U.S. 1871 Archetype. States the rule as whether “two designs are substantially the same”; uses “similar” only descriptively (e.g., “similar wreaths” may convey the “same conception”).
Smith v. Whitman Saddle Co., 148 U.S. 674 U.S. 1893 States the test as “sameness of appearance, — in other words, sameness of effect upon the eye.” Does not use “substantially similar” as a rule statement.
Ashley v. Weeks-Numan Co., 220 F. 899 2d Cir. 1915 Frames the inquiry as whether the ordinary observer “would purchase the defendant’s inkstand, believing it to be that of complainant’s.” Expressly rejects side-by-side comparison.
Inflexible Co. v. Megibow, 251 F. 924 D.N.J. 1918 Articulates the test as “sameness of appearance” producing “substantial identity of appearance.” Finds infringement despite differences “readily perceived when placed side by side.”
Range of Motion Prods., LLC v. Armaid Co., No. 2023-2427, ECF No. 85 (Cunningham, J., concurring in denial of reh’g en banc) Fed. Cir. 2026 States the rule as “substantially the same” and describes the easy case in Gorham’s deception terms; uses “similarities and differences” descriptively. Defends the “plainly dissimilar” screen, showing that faithful wording alone does not restore the inquiry. See Part IV.E.
Jacki Easlick, LLC v. CJ Emerald, Nos. 2024-1538, -1826 (nonprecedential) Fed. Cir. 2026 Quotes Gorham’s “substantially the same” in full and does not state the rule as “substantially similar.” Affirms a ruling built on a side-by-side study and a sorting test, holding that considering distinguishing features “is not legal error.” See Part VI.B.
Category 2: Interchangeable treatment (“substantially the same” and “substantially similar” both used as rule statements)
Shelcore, Inc. v. Durham Indus., Inc., 745 F.2d 621 Fed. Cir. 1984 Quotes Gorham’s “substantially the same” formulation but affirms a finding that the designs were not “substantially similar in overall appearance.”
L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117 Fed. Cir. 1993 States that infringement requires the accused design to be “substantially the same” while describing the inquiry in terms of “substantial similarity.”
KeyStone Retaining Wall Sys., Inc. v. Westrock, Inc., 997 F.2d 1444 Fed. Cir. 1993 Quotes Gorham; uses “similarity” and “substantially dissimilar” language in applying the test.
Elmer v. ICC Fabricating, Inc., 67 F.3d 1571 Fed. Cir. 1995 Quotes Gorham; concludes no reasonable jury could find the designs “substantially similar.”
Intex Recreation Corp. v. Hasbro, Inc., 3 F. Supp. 2d 1102 C.D. Cal. 1998 Quotes Gorham’s “substantially the same” standard while framing the operative inquiry as whether the designs are “substantially similar.”
Goodyear Tire & Rubber Co. v. Hercules Tire & Rubber Co., 162 F.3d 1113 Fed. Cir. 1998 Quotes Gorham; discusses whether the designs as a whole were “substantially similar.”
Moen Inc. v. Foremost Int’l Trading, Inc., 38 F. Supp. 2d 680 N.D. Ill. 1999 Recites “substantially the same” through Gorham while analyzing whether the accused faucet design is “substantially similar.”
Rockport Co. v. Deer Stags, Inc., 65 F. Supp. 2d 189 S.D.N.Y. 1999 Quotes Gorham and proceeds to evaluate “substantial similarity” of shoe designs as the operative question.
Metrokane, Inc. v. Wine Enthusiast, 185 F. Supp. 2d 321 S.D.N.Y. 2002 States the accused design is compared to determine whether it is “substantially similar in appearance” to the patented design, alongside Gorham’s formulation.
Contessa Food Prods., Inc. v. Conagra, Inc., 282 F.3d 1370 Fed. Cir. 2002 Quotes Gorham while describing the ordinary-observer analysis as finding the designs “substantially similar.”
Brown Jordan Int’l, Inc. v. Mind’s Eye Interiors, Inc., 236 F. Supp. 2d 1152 D. Haw. 2002 Recites both “substantially the same” and “substantially similar” as statements of the test.
ADC Telecomms., Inc. v. Panduit Corp., 200 F. Supp. 2d 1022 D. Minn. 2002 Interchangeable rule statements in the ordinary-observer analysis.
Bernhardt L.L.C. v. Collezione Europa USA, Inc., 280 F. Supp. 2d 485 M.D.N.C. 2003 Uses “substantially similar” language while quoting Gorham.
Durdin v. Kuryakyn Holdings, Inc., 440 F. Supp. 2d 921 W.D. Wis. 2006 Quotes Gorham; applies the ordinary-observer test in terms of whether the designs are “substantially similar.”
PHG Techs., LLC v. St. John Cos., 529 F. Supp. 2d 852 M.D. Tenn. 2007 States the inquiry as whether the patented design is “substantially similar in appearance” to the accused design, while also quoting Gorham. (The Federal Circuit’s earlier decision in the same litigation is the source of the functionality factors discussed in Parts VII–VIII.)
Richardson v. Stanley Works, Inc., 597 F.3d 1288 Fed. Cir. 2010 Quotes Gorham-faithful formulations, including deception language, but frames its operative conclusion in similarity terms (“the overall visual effect of the Fubar was not substantially similar to that of the Stepclaw”). Correct on functionality; participates in the verbal drift.
Jacki Easlick, LLC v. CJ Emerald, No. 2:23-cv-2000, ECF No. 75 (aff’d, Nos. 2024-1538, -1826 (Fed. Cir. 2026)) W.D. Pa. 2024 Quotes Gorham’s “substantially the same” in full, then concludes plaintiffs did not show “a likelihood of substantial similarity” and, in the same analysis, that the designs were not shown to be “substantially the same.” Conducts a self-described “side-by-side study” (citing Crocs); attributes “plainly dissimilar” to Amini Innovation, a pre-Egyptian Goddess decision; finds a “substantial question of whether the claimed and accused designs are plainly dissimilar” without conducting any prior art comparison. On reconsideration, describes what plaintiffs failed to show as “a likelihood of substantial similarly [sic].” Affirmed, Fed. Cir. Aug. 14, 2026 (nonprecedential). See Part VI.
Range of Motion Prods., LLC v. Armaid Co. (majority), 166 F.4th 981 Fed. Cir. 2026 Recites Gorham’s “substantially the same” through Egyptian Goddess but states that “the ultimate question is whether the claimed and accused designs are substantially similar,” and describes the test throughout as “substantially similar.”
Range of Motion Prods., LLC v. Armaid Co. (Moore, C.J., dissenting), 166 F.4th 981 Fed. Cir. 2026 Criticizes the “plainly dissimilar” reframing while itself describing the required analysis as “the substantial similarity analysis required by Gorham.” Illustrates how deeply Category 2 usage runs.
Range of Motion Prods., LLC v. Armaid Co., No. 2023-2427, ECF No. 85 (Moore, C.J., dissenting from denial of reh’g en banc) Fed. Cir. 2026 Quotes Gorham’s “substantially the same” but would “return us to Gorham, where the test for design patent infringement is whether an ordinary observer would find the two designs are substantially similar in overall appearance.” See Part IV.E.
Category 3: Formal reformulation (rule affirmatively restated as “substantially similar”)
Tomkinson v. Willets Mfg. Co., 23 F. 895 C.C.S.D.N.Y. 1884 Thirteen years after Gorham: “if to the eye of an ordinary person the two are substantially similar, it is enough.” Substantively holistic; linguistically drifted.
Try-Me Beverage & Compound Co. v. Metropole, 25 F.2d 138 D.S.C. 1928 States the standard as whether “two designs are substantially similar, and if the resemblance is such as to deceive an observer.” Same opinion rejects side-by-side comparison and difference-hunting.
Blumcraft of Pittsburgh v. Citizens & S. Nat’l Bank of S.C., 286 F. Supp. 448 D.S.C. 1968 Uses the “two designs are substantially similar” formulation as the rule.
OddzOn Prods., Inc. v. Just Toys, Inc., 122 F.3d 1396 Fed. Cir. 1997 Factfinder must determine whether the patented design as a whole is “substantially similar in appearance” to the accused design.
E. Am. Trio Prods., Inc. v. Tang Elec. Corp., 97 F. Supp. 2d 395 S.D.N.Y. 2000 Refers directly to “the Gorham test for substantial similarity.”
Hsin Ten Enter. USA, Inc. v. Clark Enters., 149 F. Supp. 2d 60 S.D.N.Y. 2001 Articulates the inquiry as whether the accused design is “substantially similar in appearance” to the patented design.
Hoop v. Hoop, 279 F.3d 1004 Fed. Cir. 2002 Describes the test as whether the second asserted invention is “substantially similar” to the first.
Tecumseh Prods. Co. v. Briggs & Stratton Corp., 295 F. Supp. 2d 902 E.D. Wis. 2003 Articulates the inquiry as whether the accused design is “substantially similar in appearance” to the patented design.
Arminak & Assocs., Inc. v. Saint-Gobain Calmar, Inc., 501 F.3d 1314 Fed. Cir. 2007 Frames the ordinary observer test as whether the accused design is “substantially similar” to the patented design. Same opinion declares “no authority” bars side-by-side comparison.

Note: This appendix is illustrative, not exhaustive. Additional decisions could be added to any category. The Category 2 and Category 3 decisions are discussed in Part III.D; Category 1 cases are discussed in Parts II, III.A, IV.E, V, and VI.B.

This presentation is published by Design Patent Pro LLC as general educational and informational material about United States design patent practice. It is not legal advice, and it is not a substitute for the advice of an attorney. Viewing this presentation, downloading any accompanying materials, or contacting Design Patent Pro does not create an attorney-client relationship. Reasonable efforts were made to ensure accuracy, but the law changes and errors can occur, so nothing here should be relied upon as a substitute for independent research. ©2026 Robert G. Oake, Jr. All rights reserved.