concurring specially:
It seems to me that the District Court’s error was broader and more serious than the majority’s analysis concludes, and I write separately to highlight some of those differences.1
This case reveals the critical need to see the “big picture” when attempting to determine what constitutes fair use of copyrighted work. It also highlights how the temptation to apply traditional statutory interpretation principles to a common law concept can lead to serious error.
While copyright is a creature of statute, the doctrine of fair use has always been governed by judicially-created common law principles. See 4 Nimmer on Copyright § 13.05 (2014) (“Nimmer”) (stating that fair use was, and remains, a “judge-made rule of reason”). The Copyright Act of 1909 (and earlier versions) did not even mention fair use, notwithstanding that it had been around since “the infancy of copyright protection[.]” Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 575-76, 114 S.Ct. 1164, 1169-70, 127 L.Ed.2d 500 (1994) (discussing early cases). When Congress enacted a new copyright statute in 1976, it recognized the existence of fair use (for the first time) in Section 107 of the Act, which provided that “fair use ... is not an infringement of copyright” and summarized the four non-exhaustive factors that courts should consider in analyzing claims of fair use: (1) the purpose of the use, (2) the nature of the work, (3) the size and significance of the copied portion, and (4) the effect of the use on the value or potential market for that work. 17 U.S.C. § 107(1)-(4). As the Supreme Court of the United States has observed, however:
Congress meant § 107 “to restate the present judicial doctrine of fair use, not to change, narrow, or enlarge it in any way” and intended that courts continue the common-law tradition of fair use adjudication.
Campbell, supra, 510 U.S. at 577, 114 S.Ct. 1164 (citations omitted). Thus, it is important to keep in mind that fair use analysis does not require conventional statutory interpretation or the mechanical application of a checklist; it is, instead, rooted in the “270-year-old tradition of judge-made law and in judicial common sense — the mortar of the common law.” Pierre N. Leval, Nimmer Lecture: Fair Use Rescued, 44 UCLA L.Rev. 1449, 1454 (1997) (“Nimmer Lecture ”).
While Section 107 was well intended, judges have had difficulty applying statutory recognition to common-law adjudication, and this case presents a good example of that situation. Courts frequently focus on the word-bites of the statute instead of the fair use doctrine itself. As Judge Leval has bluntly described, there was “chaotic confusion that resulted from the statutory recognition” in 1976, and, as courts “looked for answers in the wrong *1285place, fair use jurisprudence became a comedy of miscommunication.... The doctrine was thus broken into a haphazard assortment of nonfunctional fragments, its core elements forgotten.” See Nimmer Lecture, supra, 44 UCLA L.Rev. at 1450; Nimmer, supra, at § 13.05 (noting fair use has been called “‘the most troublesome [area] in the whole law of copyright’ ” and is said to be “ ‘so flexible as virtually to defy definition’ ”) (citations omitted). Nonetheless, despite the confusion that statutory incorporation of the doctrine has brought about, one thing should be reasonably clear from the name of the doctrine: a proper fair use analysis should focus primarily on the use of the work, not on the user. So, in analyzing fair use in a given case, the court should step back a little, just as you would at an art museum, and view the work and its use in its entirety. Viewed in this way, and after applying traditional common law principles to the use at issue here, this is a rather simple case. Checking the four statutory factors to ensure that they have been considered merely affirms the conclusion that what GSU is doing is not fair use.2
This case does not involve an individual using a single copyrighted work, nor does it involve a single course, a single professor, or even a one-time use of “multiple copies for classroom distribution.” See Campbell, supra, 510 U.S. at 579 n. 11, 114 S.Ct. 1164. Nor, in my opinion, should it be confined to the seventy-four specific instances of infringement that were the focus during trial. Rather, this case arises out of a university-wide practice to substitute “paper coursepacks” (the functional equivalent of textbooks) that contained licensed copyrighted works with “digital coursepacks” that contained unlicensed, copyrighted works. This was done for the vast majority of courses offered at GSU and, as will be seen, it was done primarily to save money.3 As Plaintiffs have pointed out:
*1286What makes GSU’s disregard for copyright law in relation to the creation and dissemination of digital coursepacks [most] remarkable is its stark contrast to GSU’s legally compliant conduct when it supplies the same readings in physical form. It is stipulated that when GSU faculty make precisely the same uses of excerpted copyrighted books to create paper coursepacks for students, GSU has paid permissions fees to [Plaintiffs] and the other publishers of those works.
(emphasis added). Thus, Defendants have the burden to show by a preponderance of the evidence why this aggregated use in electronic form is fair use — when the exact same use in paper form is not. In my view, they have not even come close to doing so. Plaintiffs’ brief then goes on to cite two important cases in this area:
Two key decisions from the 1990’s, Princeton Univ. Press v. Mich. Document Servs., Inc., 99 F.3d 1381 (6th Cir.1996) (en banc), and Basic Books, Inc. v. Kinko’s Graphics Corp., 758 F.Supp. 1522 (S.D.N.Y.1991) (the “Cour-sepack Cases”), established that the nontransformative, verbatim copying inherent in the preparation of course-packs, with the resulting prospect of significant market harm to the authors and publishers of the materials, did not constitute fair use. Significantly, GSU stipulated [in the District Court] that it has for years abided by these requirements for paper coursepacks, routinely seeking permission from copyright holders and paying the requisite fees when printing and selling paper coursepacks.
(emphasis added). The twice-emphasized stipulation referenced above — which is arguably the most meaningful fact in this case — specifically provided that: “GSU pays permissions fees when copyrighted content is used in hardcopy coursepacks.” Thus, James Palmour (whose job at GSU was to arrange coursepacks and evaluate the materials included therein “to determine if copyright laws are violated or not”) testified that, when he prepared paper coursepacks, “if anything was copyrighted I obtained the copyright permission and paid the royalties to the publishers.... ” Palmour Dep. at 16, 24-25, 30. He did this whenever the work had already been published in a book or journal because, in that situation, “I assume that it needs copyright permission” before it could be put in a paper coursepack. Id. at 30-31, 34. Notably, however, that same assumption was not made (and the permissions fees were not paid) when GSU moved to digital coursepacks, which (as Palmour explained) was done because they were “easier”, “cheaper” and “convenient.” Id. at 129,134-35.4
In short, GSU went from paper course-packs (for which they had “obtained the copyright permission and paid the royalties to the publishers”) to digital course-packs (for which they did not), and they did this not because there was any real, difference in the actual use but, rather, in large part to save money.5
*1287In light of the undisputed fact that GSU paid the permissions fees whenever copyrighted works were included in paper coursepacks, but did not pay those fees when “they put the same material on E Reserves”, resolution of this case (which involves admitted extensive verbatim copying in undisputed non-transformative format, resulting in complete market substitution) seems clear. See Greenberg v. National Geographic Society, 533 F.3d 1244, 1257 (11th Cir.2008) (en banc) (stating that the transfer of a work “from print to digital form ... does not create a different balance of copyright protection ... because copyright protection is media neutral”) (citing New York Times Co. v. Tasini 533 U.S. 483, 121 S.Ct. 2381, 150 L.Ed.2d 500 (2001)). Inasmuch as both sides in this case stipulated at trial (and Palmour’s testimony further made clear) that royalty payments were made when copyrighted works were provided to students in paper form, providing the same or similar materials in an electronic form should not change the calculus in any meaningful or significant way. The “use” of a copyright protected work that had previously required the payment of a permissions fee does not all of a sudden become “fair use” just because the work is distributed via a hyperlink instead of a printing press.6
To the extent the majority concludes that reliance on the media neutrality doctrine is “misplaced”, and that the Course-pack Cases are not controlling since they involved commercial copyshops, see Maj. Op. at 1260-61, 1263-64, I respectfully disagree. The media neutrality doctrine should not be limited to copyrightability, but, rather, should apply equally in a fan-use analysis. There is no transformative technology transition here. The digital format is merely another way of displaying the same paginated materials as in a paper format and for the same underlying use. Electronic reproduction is faster, cheaper, and almost unlimited in its scope and duration, but there is no discernable difference in its use, purpose, and effect. And while the Coursepack Cases are, of course, not binding on us, I find their legal and factual rationale persuasive and would reverse for largely the same reasons set out in those decisions.7
*1288The majority allows that the-Coursepack Cases may “provide guidance” in conducting the fair use analysis, but it concludes that the cases do not dictate the outcome here since every case must be decided on its own facts and by individual application of the fair use factors. I do not disagree that each individual case must be decided on its own facts, as I have already set out. However, in undertaking its analysis in this case, I believe the majority has misapplied several of those factors — or at least certain parts thereof. I will now proceed to highlight where I part company with the majority’s analysis of the fair use factors (and to briefly add to some portions of the analysis with which I agree).
It appears that the majority finds that the first factor (i.e., the purpose of the use) favors a finding of fair use based primarily on the fact that GSU is a not-for-profit university using the copyrighted material for educational purposes. Neither churches, charities, nor colleges get a free ride in copyright, however. The test is ultimately the same for them as it is for everyone else: is the use “fair” under the specific circumstances? While I agree that educational use is an important factor to consider, and there is much to recommend in the majority’s thoughtful analysis and detailed consideration of this issue, see Maj. Op. at 1261-67 — which stands in stark contrast to the District Court’s perfunctory (two page) analysis8 — I simply cannot agree that the first factor weighs in favor of fair use just because the works are being used for educational purposes at a non-profit university.9
In Campbell, the Supreme Court expressly stated that the “central purpose” of the investigation under the first factor is to determine “whether the new work merely supersedes the objects of the original creation” or, instead, whether it adds “something new” with an additional purpose or different character. See 510 U.S. at 578-79, 114 S.Ct. 1164 (citations, quotation marks, and alterations omitted); see also Nimmer, supra, § 13.05[A][1][b]. The Supreme Court went on to state in this context that “the mere fact that a use is *1289educational and not for profit does not insulate it from a finding of infringement.” Campbell, supra, 510 U.S. at 584, 114 S.Ct. 1164. Thus, rather than the focus being on the distinction between commercial and non-profit — or educational and non-educational — the real “backbone” of the first factor is the “crucial” distinction between transformative use (which tends to support fair use) and non-transformative use that supersedes the original work (which cuts against it). See Nimmer Lecture, supra, 44 UCLA L.Rev. at 1460-61; Nimmer, supra, § 13.05[A][1][b]. The use of the works in this case, as the majority opinion notes and the Defendants have not really contested, was obviously non-transforma-tive. In switching from paper to digital coursepacks, the professors at GSU superseded the original works, and displaced and substituted the “textbooks” which contained the copyrighted works for which Plaintiffs had previously been paid. In my view, this use compels the conclusion that the first factor weighs heavily against fair use (despite that the user is a non-profit university).
As for the second factor (i.e., the nature of the work), the District Court held that it weighs in favor of Defendants “because all of the excerpts are informational and educational in nature and none are fictional[.]” See Cambridge Univ. Press v. Becker, 863 F.Supp.2d 1190, 1242 (N.D.Ga.2012). This conclusion was error, as the majority opinion holds.- Admittedly, the “nature of the work” encompasses a huge spectrum of considerations, often bridging the field of copyrighted works. Characterizing the nature of poetic works, visual art, or musical compositions and performances is quite different from that of written textual materials. For some works, the contested use also may affect the “nature” of the work. A proper analysis of the second factor should not focus on whether a work is fiction or nonfiction; rather, in my opinion, the focus should be on the originality and creativity of the work and its value to the public (or, as here, to the academic community).10 Here, the works were all original, and it cannot be denied that they had value. To be sure, their value is demonstrated by the fact that they were each selected and utilized for classroom purposes by professors to illustrate the pedagogical point for which they were chosen. I thus agree with the majority that the District Court erroneously held that factor two weighs in favor of fair use for informational or educational works of non-fiction “in every instance.” Such works can be, and frequently are, original and valuable works that contain evaluative, analytical, or subjectively descriptive material. I also agree with the majority that factor two is either neutral here or weighs against fair use.11
*1290In considering the third factor (i.e., the size and significance of the portion that was copied), the majority concludes that “[t]he District Court’s blanket 10 percent- or-one-chapter benchmark was improper.” Maj. Op. at 1271. I agree for all the reasons that the majority has identified.12 To the extent the majority then goes on to reject Plaintiffs’ claim that the Classroom Guidelines should further inform the analysis under the third factor, I disagree. The guidelines — which expressly deal with fair use of copyrighted material in the classroom context and place limits on not-for-profit educational copying — are a compromised and negotiated arms-length agreement that Congress had asked for, and was fully aware of and took into account, at the time that Section 107 was enacted. They provide, inter alia, strict word count limits on allowable copying, such as the lesser of an excerpt from a prose work of not more than 1,000 words or 10 percent of the work. While the majority opinion is correct that the guidelines do not create a “hard evidentiary presumption” or “carry force of law”, they are still important to this analysis as the Coursepack Cases (and other cases) have recognized, and I believe they deserve more weight and consideration than the majority has allowed. See Princeton Univ. Press, supra, 99 F.3d at 1390 (noting that there are “strong reasons” to consider the Classroom Guidelines as they “evoke a general idea ... of the type of educational copying Congress had in mind”). With respect to factor three, they provide a standard of presumed reasonableness. If they are properly considered and applied to the facts of this case, I agree with the Plaintiffs that they would confirm that the Defendants “helped themselves overmuch.” See Campbell, supra, 510 U.S. at 587, 114 S.Ct. 1164.13
And lastly, with respect to the fourth factor (i.e., the effect of the use on the value or potential market for that work), there are two inquiries: (1) the extent of the market harm caused by the conduct of the alleged infringer, and (2) whether such conduct—if unrestricted and widespread— would result in a substantially adverse impact on the potential market for the original. See Campbell, supra, 510 U.S. at 590, 114 S.Ct. 1164. “In particular, the adverse effect with which fair use is primarily concerned is that of market substitution.” Peter Letterese & Assocs., Inc. v. World Institute of Scientology Enters., Int’l, 533 F.3d 1287, 1315 (11th Cir.2008). The threat of market substitution in this case, as the majority opinion rightly notes, is severe. However, after noting that digital licenses were not available for some of the copyrighted works at issue, the majority opinion concludes “the District Court held that the fourth factor weighted in favor of fair use. We find that the District Court’s analysis under the fourth factor was correct. ...” See Maj. Op. at 1278.
I disagree and am instead persuaded by the Plaintiffs’ arguments, which highlight multiple legal and factual errors in the District Court’s analysis of the fourth fac*1291tor. In short, establishing market harm does not require a showing of lost profits, nor is it dependent on the availability of a digital license.14 Rather, what counts is whether “some meaningful likelihood of future harm exists.” Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 451, 104 S.Ct. 774, 793, 78 L.Ed.2d 574 (1984). The Sixth Circuit undertook- a proper market-harm analysis in one of the Coursepack Cases when it assessed the potential impact on publishers if unauthorized reproduction of scholarly works for paper eoursepacks was replicated nationwide:
If copyshops across the nation were to start doing what the defendants have been doing here, this revenue stream would shrivel and the potential value of the copyrighted works of scholarship published by the plaintiffs would be diminished accordingly.
Princeton Univ. Press, supra, 99 F.3d at 1387. One could substitute “universities” for “copyshops” in the above quoted passage (which, as I indicated earlier, would be appropriate since the underlying use is the same in both cases) and would have to reach the same conclusion.
Therefore, I would go further than does the majority and conclude that both the District Court’s methodology and its analysis were flawed. The Defendants’ use fails under any objective common law “big picture” adjudication of fair use and also fails on a work-by-work analysis under three of the four factors (while the remaining factor is either neutral or weighs' against Defendants). It has been said that fair use is best and most precisely explained by the following paraphrase of the Golden Rule: “ ‘Take not from others to such an extent and in such a manner that you would be resentful if they so took from you.’ ” See Nimmer, supra, at § 13.05[A] (citation omitted). If GSU or the individual defendants held copyrights for which other universities had always sought — and paid for — permission to copy and provide to students on paper, I am confident they would be “resentful” if those universities (in an effort to save money) just stopped paying the permissions fees and instead provided the same or similar materials by digital means. GSU’s -use of Plaintiffs’ copyright protected works without compensation was, in a word, unfair.
I thus concur with the decision to reverse and remand this case.
. As will be seen, my disagreement with the majority opinion is limited to its fair use analysis. Since I agree with the majority with respect to vacating the injunction, declaratory relief, and attorney fee award, I will not separately address those subjects here.
. The statutory factors are not exhaustive, and there is an unlimited universe of possible things to consider, which automatically makes it improper to apply the factors—as the District Court did—in a straight arithmetic manner. See Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L.Rev. 1105, 1110 (1990) ("The factors do not represent a score card that promises victory to the winner of the majority.”).
. The majority notes that, during the Spring 2009 term, paper coursepacks were offered for about fifteen courses, compared to hundreds of courses that utilized the digital cour-sepacks. Plaintiffs contend on appeal that the District Court erred by performing a work-by-work analysis which focused on each individual work rather than analyzing fair use in the context of this broad ongoing practice at GSU. The majority opinion rejects this argument, concluding that "[cjourts must apply the fair use factors to each work at issue. Otherwise, courts would have no principled method of determining whether a nebulous cloud of alleged infringements purportedly caused by a secondary user should be excused by the defense of fair use.” See Maj. Op. at 1258-59 & n. 20. If we were writing on a blank slate in the normal case, I would agree that the court should focus on the details. But, we are not writing on a blank slate in a normal case. This is a unique (but much easier) case as it is undisputed that GSU had always paid permissions fees to use the copyrighted works in paper form, but it refused to do so when it used the same or similar copyrighted work in digital form. The fact that GSU previously paid the permissions fees is strong, if not conclusive, evidence that the underlying use at issue was not “fair use;” indeed, it is practically an admission that it was not. Thus, in my view, there is no “nebulous cloud” surrounding what happened here that would otherwise require a work-by-work analysis. Rather, it seems clear that the challenged use (replacing paper coursepacks comprised of licensed copyrighted works with digital versions comprised of unlicensed works) is an aggregated one that should be viewed in the "big picture.” I believe the District Court’s work-by-work methodology— on the facts of this specific case — was an un*1286necessary undertaking and serious error which led that court astray.
. The only cost associated with the digital coursepacks was apparently incurred when • the materials were printed out. As Palmour testified:
Q. And so if the professor opts to use a coursepack, a printed coursepack, they have to pay permission fees under your system?
A. Yes.
Q. And if they put the same material on EReserves, it’s just the printing cost, is that right?
A. Yes.
Palmour Dep. at 135 (emphasis added).
. To illustrate, Palmour testified about an instructor who had recently wanted to use four chapters from a psychology textbook for her paper coursepack. When she found out how *1287much the publisher intended to charge (25 cents per page), the instructor just reduced the amount that she wanted to copy to get it under 20% of the whole book (per the Regents' Guide as described by the majority) and then put it on EReserves, thereby avoiding paying any copyright fee.
. I recognize that digital permissions (or licenses) were not offered for all of Plaintiffs' works. Nevertheless, the mere fact that a license may have been available in paper form but not in digital form — perhaps for market reasons — should not mean that users only have to pay for the first but have carte blanche under fair use for the second. The underlying use is the same in both cases. So, if a license was only available for a work in paper form, GSU could presumably pay for it (as it always had), or participate in the Academic Repertory License Service (which it did not), and “make the desired number of photocopies or scan and place the electronic excerpt online.” Cf. Cambridge Univ. Press v. Becker, 863 F.Supp.2d 1190, 1212-15 (N.D.Ga.2012) (describing how permission services offered by the Copyright Clearance Center (CCC) function in the context of licensing works for use by universities and other educational institutions).
. Although the Coursepack Cases involved copying by commercial print shops for a nonprofit university'and not, as here, copying by a non-profit university, as I have already indicated, and as the majority also notes, see Maj. Op. at 1264, fair use analysis should focus primarily on the use, not on the user. The use at issue in this case and in the Coursepack Cases (specifically, nontransformative, extensive, and verbatim copying of copyrighted protected works for the inclusion in university “coursepacks” — a commercial substitution) *1288and the effect on the market for those protected works is exactly the same.
. As one of the amici notes: “To say that the court’s 'analysis' of the first fair use factor was cursory is an understatement. The court looked no further than the nonprofit status of GSU and the fact that teaching was involved to find that 'the first fair use factor favors Defendants.' ... A determination that copies of protected works are being made by a nonprofit entity for use in education should be the beginning, not the end, of the discussion.” Brief of Marybeth Peters, Ralph Oman and Jon Baumgarten as Amici Curiae in Support of the Appellants Recommending Reversal at 8.
. The Supreme Court said almost thirty years ago that ‘‘[t]he crux of the profit/non-profit distinction is not whether the sole motive of the use is monetary gain but whether the user stands to profit from exploitation of the copyrighted material without paying the custom-aiy price.” See Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 562, 105 S.Ct. 2218, 2231, 85 L.Ed.2d 588 (1985). Defendants obviously "stand to profit” from exploiting the copyrighted works at issue here, if nothing else by virtue of the fact that GSU is no longer paying the fees that it previously deemed appropriate. As a result, GSU students no longer have to pay for classroom materials and books in digital form that they (through tuition and/or other fees) used to pay for in paper form. See note 5 supra. Buying and selling books or other textual material is clearly a commercial endeavor (even in a university setting), and the substitution of digital texts for paper texts does not change that commercial aspect. In any event, even if GSU did not "stand to profit” from the use at issue, as will be seen, the Supreme Court has emphasized that the most pertinent inquiry under the first factor is not profit vs. nonprofit (or educational vs. noneducational), but, rather, transformative vs. non-transfor-mative.
. To say, as the District Court impliedly did, that a work of fiction is deserving of more copyright protection than a work of non-fiction, without regard for the originality and value of the works involved, would suggest (for example) that a pulp fiction novel should be entitled to more protection than Darwin's The Origin of Species, which would, of course, be absurd.
. Both the District Court and the majority have noted that the Coursepack Cases reached opposite conclusions on factor two. In Princeton University Press, the Sixth Circuit held (albeit apparently based on stipulation and with little discussion) that the second factor weighed against fair use as the factual, non-fictional, and scholarly works in that case "contained creative material or ‘expression’ ” and were "certainly not telephone book listings. ...” 99 F.3d at 1389. In Basic Books, by contrast, the District Court stated as a “general rule of thumb” that non-fictional works were deserving of less protection than fictional works for purposes of the second factor. 758 F.Supp. at 1532-33. I agree with the conclusion reached by the Sixth Circuit.
. But, I also believe, unlike the majority, that the District Court erred in its treatment of edited compilations and its calculations and methodology in measuring the amount taken by the Defendants.
. Plaintiffs argue in their brief, and I could not agree more, that:
In dismissing the Guidelines as irrelevant, the district court noted that they are "so restrictive that no book chapters in this case ... would qualify for fair use.” But unless the object of the fair-use exercise is to validate GSU's ongoing practices (which it is not), rather than to require that those practices comply with the quantitative (and other) limits Congress had in mind when it adopted section 107, the district court’s reasoning cannot stand.
. The market for certain of the works at issue may be quite small (and thus the copyright holders may have made a decision to not market the work in a digital form), but the material can be reproduced digitally, see note 6 supra, and the copyright, still must be respected and protected.