Matthew Bender & Company, Inc., Hyperlaw, Inc., Intervenor-Plaintiff-Appellee v. West Publishing Co. And West Publishing Corporation

158 F.3d 674, 48 U.S.P.Q. 2d (BNA) 1560, 1998 U.S. App. LEXIS 30790, 1998 WL 764837

Opinion

Judge SWEET dissents in a separate opinion.

JACOBS, Circuit Judge:

West Publishing Co. and West Publishing Corp. (collectively “West”) publish compilations of reports of judicial opinions (“case reports”). Each case report consists of the text of the judicial opinion with enhancements that for the purposes of this case can be put in two categories: (i) independently composed features, such as a syllabus (which digests and heralds the opinion’s general holdings), headnotes (which summarize the specific points of law recited in each opinion), and key numbers (which categorize points of law into different legal topics and subtopics), and (ii) additions of certain factual informa tion to the text of the opinions, including parallel or alternative citations to cases, attorney information, and data on subsequent procedural history. HyperLaw, Inc. publishes compact disc-read only memory (“CD-ROM”) compilations of Supreme Court and United States Court of Appeals decisions, and intervened as a plaintiff to seek a judgment declaring that the individual West case reports that are left after redaction of the first category of alterations the independently composed features), do not contain copyrightable material. West now appeals from a judgment of the United States District Court for the Southern District of New York (Martin, /.), following a bench trial, granting declaratory judgment in favor of HyperLaw. Matthew Bender & Co. v. West Publishing Co., No. 94 Civ. 0589, 1997 WL 266972 (S.D.N.Y. May 19, 1997).

It is true that neither novelty nor invention is a requisite for copyright protection, but minimal creativity is required. Aside from its syllabi, headnotes and key numbers— none of which HyperLaw proposes to copy— West makes four different types of changes to judicial opinions that it claimed at trial are copyrightable: (i) rearrangement of information specifying the parties, court, and date of decision; (ii) addition of certain information concerning counsel; (iii) annotation to reflect subsequent procedural developments such as amendments and denials of rehearing; and (iv) editing of parallel and alternate citations to cases cited in the opinions in order to redact ephemeral and obscure citations and to add standard permanent citations (including West reporters). All of West’s alterations to judicial opinions involve the addition and arrangement of facts, or the rearrangement of data already included in the opinions, and therefore any creativity in these elements of West’s ease reports lies in West’s selection and arrangement of this information. In light of accepted legal conventions and other external constraining factors, West’s choices on selection and arrangement can reasonably be viewed as obvious, typical, and lacking even minimal creativity. Therefore, we cannot conclude that the district court clearly erred in finding that those elements that HyperLaw seeks to copy from West’s case reports are not copyrightable, and affirm.

BACKGROUND

West obtains the text of judicial opinions directly from courts. It alters these texts as described above to create a case report, and then publishes these case reports (first in advance sheets, then in bound volumes) in different series of “ease reporters.” 1 These case reporter series cover all state and federal courts and are collectively known as West’s “National Reporter System.” Two series of case reporters are at issue in this case: the Supreme Court Reporter, which contains, inter alia, all Supreme Court opinions and memorandum decisions; and the Federal Reporter, which contains, inter alia, all United States Court of Appeals opinions designated for publication, as well as tables showing the disposition of unpublished cases.

HyperLaw markets two compilations that cover approximately the same ground: Supreme Court on Disc, an annual CD-ROM disc containing opinions of the United States Supreme Court starting from 1990; and Federal Appeals on Disc, a quarterly CD-ROM disc containing nearly all opinions (published and unpublished) of the United States Courts of Appeals from January 1993 on. 2 Currently, HyperLaw obtains the text of the opinions directly from the courts. However, Hyper-Law intends to expand its CD-ROM product to include any recent cases it could not obtain directly from the courts (and attorney information that is omitted from slip opinions by certain circuits), as well as pre-1990 Supreme Court cases and pre-1993 court of appeals cases that are cited in recent Supreme Court and court of appeals eases (so that users can jump to those cases). Hyper- Law intends to achieve this expansion by copying West’s case reports (after redacting the syllabi, headnotes and key numbers) from the Supreme Court Reporter and the Federal Reporter. The total number of opinions HyperLaw intends to copy is unclear; but HyperLaw’s President Alan Sugarman testified that in time it could reach 50 percent of Supreme Court and court of appeals decisions published by West.

Following the commencement of suit by Matthew Bender & Co. in the United States District Court for the Southern District of New York seeking a judgment declaring that Bender’s insertion of star pagination to West’s case reporters in its CD-ROM version of judicial opinions did not infringe West’s copyright, HyperLaw intervened and requested the same relief. In addition, Hy-perLaw sought a declaration that Hyper-Law’s redacted versions of West’s case reports contain no copyrightable material and thus may be copied without infringement. On the star pagination issue, the district court granted summary judgment to Bender and HyperLaw, and final judgment was entered pursuant to Fed.R.Civ.P. 54(b). (We affirm that ruling in a separate opinion issued today). See Matthew Bender & Co. v. West Publ’g Co., 158 F.3d 693 (2d Cir.1998). But the district court denied summary judgment allowing HyperLaw to copy redacted versions of West’s case reports, and conducted a bench trial on this issue.

The principal trial witness was Donna Bergsgaard, the manager of West’s manuscript department. She specified four kinds of alterations made by West to the opinions that it publishes in the Supreme Court Reporter and Federal Reporter and that Hyper-Law intends to copy: (i) the arrangement of prefatory information, such as parties, court, and date of decision; (ii) the selection and arrangement of the attorney information; (iii) the arrangement of information relating to subsequent procedural developments; and (iv) the selection of parallel and alternative citations.

Following the bench trial, the district court ruled that West’s revisions to judicial opinions were merely trivial variations from the public domain works, and that West’s case reports were therefore not copyrightable as derivative works. Matthew Bender & Co., 1997 WL 266972 at *4. In reaching this conclusion, the district court reviewed each type of alteration and found that “West does not have a protectible interest in any of the portions of the opinions that HyperLaw copies or intends to copy” because West’s altera- ' tions lack even minimal creativity. Id.

DISCUSSION

I

As a threshold matter, West argues that the judgment should be vacated and the action dismissed as moot because HyperLaw failed to demonstrate through admissible evidence a sufficiently real intent to copy West’s case reports in quantities beyond what West conceded (immediately prior to trial) would be permissible under the fair use doctrine, i.e., one to two percent of its published reports of Supreme Court and court of appeals cases.

Early in the- litigation, the district court held a justiciability hearing and concluded that HyperLaw had the ability and intent to copy certain elements of West’s ease reports and that if HyperLaw did so, it faced a reasonable apprehension of suit. But a dispute remained as to the scope of HyperLaw’s intended copying. Prior to trial, West argued that HyperLaw intended no copying of West’s case reports other than the few recent court decisions that HyperLaw had not obtained from the courts, and West preemptively conceded that such copying was permitted under the fair use doctrine. Hyper-Law contested West’s characterization of the scope of its intended copying, and pointed to an affidavit (submitted earlier in the litigation) in which Alan Sugarman, HyperLaw’s president, expressed HyperLaw’s intent to copy opinions from earlier years cited in opinions included on their CD-ROM discs.

Faced with West’s last-minute concession, the district court reserved decision on West’s motion to dismiss for mootness and proceeded with the bench trial, at which Hypei'Law would have the opportunity to offer testimony on the percentage of West case reports it intended to copy. After the close of evidence, West moved to dismiss for mootness on the ground that HyperLaw neglected to adduce such evidence. The district court, however, permitted HyperLaw to remedy its omission by offering Sugarman’s affidavit on that subject for the truth of the matters stated therein, subject to West’s opportunity to cross-examine Sugarman. During this cross-examination, Sugarman testified that HyperLaw intended to copy most older eases cited in recent Supreme Court and court of appeals decisions included on the CD-ROM discs; he estimated that this copying might approximate half of West’s case reports. On the basis of Sugarman’s testimony, the district court concluded that a live controversy existed. We cannot find that this determination was in error.

West’s mootness argument amounts to a claim that the district court erred in admitting Sugarman’s testimony following the close of evidence. A district court’s decision to reopen the proof to allow a party to submit additional evidence is subject to its sound discretion. See Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 331, 91 S.Ct. 795, 802-03, 28 L.Ed.2d 77 (1971); see also Fed.R.Civ.P. 59(a) (“On a motion for a new trial in an action tried without a jury, the court may open the judgment if one has been entered, take additional testimony, amend findings of fact and conclusions of law or make new findings and conclusions, and direct the entry of a new judgment.”). The district court’s decision to receive Sugarman’s testimony in evidence is therefore reviewed for abuse of discretion. See Air et Chaleur, S.A. v. Janeway, 757 F.2d 489, 495 (2d Cir.1985); see also Garcia v. Woman’s Hosp., 97 F.3d 810, 814 (5th Cir.1996) (“Among the factors the trial court should examine in deciding whether to allow a reopening are the importance and probative value of the evidence, the reason for the moving party’s failure to introduce the evidence earlier, and the possibility of prejudice to the non-moving party.”).

Sugarman’s testimony should have been offered earlier; but (i) the testimony was important for determining the existence of a live controversy, (ii) West was afforded a full opportunity for cross-examination, and (iii) West points to no specific prejudice from the delay. Furthermore, the district court’s decision served judicial efficiency; otherwise, the case would have been dismissed as moot without preclusive effect, HyperLaw would have simply amended its complaint to allege its intent to engage in more extensive copying, and the case would have been re-tried. The district court’s admission of Sugarman’s testimony did not amount to an abuse of discretion.

II

Works of the federal government are not subject to copyright protection; the text of judicial decisions may therefore be copied at will. 17 U.S.C. § 105. Federal judicial opinions may, however, form part of a compilation. The Copyright Act defines “compilation” as “a work formed by the collection and assembling of preexisting materials or of data that are selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship.” 17 U.S.C. § 101. West has filed a certificate of copyright registration for every paperbacked advance sheet and bound permanent volume of the Supreme Court Reporter and Federal Reporter, and each certificate characterizes the copyrighted work as a “compilation.” Under Feist Publications, Inc. v. Rural Telephone Serv. Co., 499 U.S. 340, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991), an infringement claim for a compilation has two elements: “(1) ownership of a valid copyright, and (2) copying of constituent elements of the work that are original.” Id. at 361, 111 S.Ct. at 1296.

But HyperLaw has not signaled its intent to copy the text of every case included in particular volumes of West ease reporters or the case reporters’ selection and arrangement of cases; 3 HyperLaw’s intent is to copy particular, though numerous, individual case reports. HyperLaw seeks a declaratory judgment that these case reports — after removal of the syllabus, headnotes, and key numbers — contain no copyrightable material.

' A HyperLaw contends that each case report should be analyzed as a derivative work, which is defined under the Copyright Act as, inter alia, “[a] work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship.” 17 U.S.C. § 101. The district court adopted this view and analyzed the individual case report as a derivative work, but found it wanting in the requisite originality. West contends that each case report is a compilation, ie., a collection of facts that have been distinctively selected and arranged. No one claims that a case report is anything other than a derivative work or a compilation.

The House Report on the 1976 Copyright Act distinguishes between a derivative work and a compilation:

Between them the terms ... comprehend every copyrightable work that employs preexisting material or data of any kind. There is necessarily some overlapping between the two, but they basically represent different concepts. A “compilation” results from a process of selecting, bringing together, organizing, and arranging previously existing material of all kinds, regardless of whether the individual items in the material have been or ever could have been subject to copyright. A “derivative work,” on the other hand, requires a process of recasting, transforming, or adapting “one or more preexisting works”; the “preexisting work” must come within the general subject matter of copyright set forth in section 102, regardless of whether it is or was ever copyrighted.

H.R.Rep. No. 94-1476, at 57 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5670; see also 1 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 3.02, at 3-5 (1998) (“[Wjhile a compilation consists merely of the selection and arrangement of pre-existing material without any internal changes in such material, a derivative work involves recasting or transformation, ie., changes in the preexisting material, whether or not it is juxtaposed in an arrangement with other preexisting materials.”).

We think that West’s case reports have elements of both types of works. West compiles (and selects) the factual information it includes in each case report — the type of task usually involved in creating a compilation. On the other hand, West rearranges prefatory and citation information included in judicial opinions, steps that tend toward the making of a derivative work rather than a compilation. In addition, all of West’s decisions are constrained by West’s main project, which is to enhance the judicial opinions without altering their texts in any substantive or appreciable way.

We need not categorize West’s ease reports as either derivative works or compilations in order to decide this case. Copyright protection is unavailable for both derivative works and compilations alike unless, when analyzed as a whole, they display sufficient originality so as to amount to an “original work of authorship.” See 17 U.S.C. § 101 (defining a “derivative work”, inter alia, as a work containing alterations “which, as a whole, represent an original work of authorship”); id. (defining a compilation as requiring that “the resulting work as a whole constitutes an original work of authorship”). The originality required for copyright protection is essentially the same. See Feist, 499 U.S. at 358, 111 S.Ct. at 1294 (“Originality requires only that the author make the selection or arrangement independently ... and that it display some minimal level of creativity.”); L. Batlin & Son, Inc. v. Snyder, 536 F.2d 486, 490-91 (2d Cir.1976) (in banc) (“[Wjhile a copy of something in the public domain will not, if it be merely a copy, support a copyright, a distinguishable variation will. * * * [Tjo support a copyright there must be at least some substantial variation, not merely a trivial variation such as might occur in the translation to a different medium.”). As West and HyperLaw seemingly agree, the question presented is whether West’s alterations to the case reports, when considered collectively, demonstrate sufficient originality and creativity to be copyrightable.

B

The district court found that the elements of the West case reports for which West seeks copyright protection lack sufficient originality or creativity to be protecta-ble&emdash;whether considered separately or together. Because we treat the question of whether particular elements of a work demonstrate sufficient originality and creativity to warrant copyright protection as a question for the factfinder-here the judge&emdash;we will not reverse the district court’s findings unless clearly erroneous. See, e.g., Victor Lalli Enters., Inc. v. Big Red Apple, Inc., 936 F.2d 671, 673 (2d Cir.1991) (per curiam) (“Generally, we review a district court’s determination of whether a work is sufficiently original to merit copyright protection under the clearly erroneous standard.”); Financial Information, Inc. v. Moody’s Investors Serv., Inc., 808 F.2d 204, 207-08 (2d Cir.1986) (same); see also Woods v. Bourne Co., 60 F.3d 978, 991 (2d Cir.1995) (noting that our review of originality determination is for clear error); Weissmann v. Freeman, 868 F.2d 1313, 1322 (2d Cir.1989) (holding that the district court’s finding of lack of originality or variation in derivative work was reviewable for clear error). The dissent (at [page 1], footnote 1) argues that de novo review would be more appropriate, and that our precedents to the contrary are doubtful in light of Feist. Feist, however, did not address standard of review, and the clear error standard retains vitality in our precedents, both before and after Feist. In 1995, we acknowledged that the question of copyright-ability entails the kind of conclusion that would often justify de novo review, but that most courts, including this Court, review for clear error. See Woods, 60 F.3d at 991 (Feinberg, J.) (citing 1 William F. Patry, Copyright Law and Practice 145 n. 106 (1994) (cataloguing cases)).

The only elements of a work that are entitled to copyright protection are those that are original. See Feist, 499 U.S. at 361, 111 S.Ct. at 1296; Mid America Title Co. v. Kirk, 59 F.3d 719, 721 (7th Cir.1995). The “originality” standard requires that the work result from “independent creation” and that the author demonstrate that such creation entails a “modicum of creativity.” See Feist, 499 U.S. at 346, 111 S.Ct. at 1288; see also Key Publications, Inc. v. Chinatown Today Publ’g Enters., Inc., 945 F.2d 509, 512-13 (2d Cir.1991) (“Simply stated, original means not copied, and exhibiting a minimal amount of creativity.”).

According to West, the required originality and creativity inhere in four elements of the case reports that HyperLaw intends to copy:

(i) the arrangement of information specifying the parties, court, and date of decision;

(ii) the selection and arrangement of the attorney information;

(iii) the arrangement of information relating to subsequent procedural developments such as amendments and denials of rehearing; and

(iv) the selection of parallel and alternative citations. 4

Each element either adds or rearranges preexisting facts, in themselves unprotecta-ble, and so West is not entitled to protection for these elements of its case reports unless it demonstrates creativity in the selection or arrangement of those facts. See Feist, 499 U.S. at 348, 111 S.Ct. at 1289. Because many of the cases relating to compilations assess the creativity involved in selecting and arranging information, we look to those- cases for guidance.

The Copyright Act protects original and minimally creative selection of preexisting, unprotected materials (such as facts) for inclusion in a work, as well as original and creative arrangement of those materials. See Eckes v. Card Prices Update, 736 F.2d 859, 863 (2d Cir.1984) (“[Selectivity in including otherwise non-protected information can be protected expression.”); 1 Nimmer, supra, § 3.04[B][2], at 3-31 (“If originally combined, a selection or arrangement of underlying materials that are themselves unori ginal may support copyright protection.”). Feist tells us:

The compilation author typically chooses which facts to include, in what order to place them, and how to arrange the collected data so that they may be used effectively by readers. These choices as to selection and arrangement, so long as they are made independently by the compiler and entail a minimal degree of creativity, are sufficiently original that Congress may protect such compilations through the copyright laws.

Feist, 499 U.S. at 348, 111 S.Ct. at 1289 (citations omitted). 5 However, not every such compilation or decision on selection or arrangement is sufficiently creative to be protected.

The creative spark is missing where: (i) industry conventions or other external factors so dictate selection that any person composing a compilation of the type at issue would necessarily select the same categories of information, see, e.g., Victor Lalli Enters., 936 F.2d at 672 (charts of winning numbers in illegal gambling operations); see also Mid America Title Co., 59 F.3d at 722 (title examiner’s report; “[selecting which facts to include in this compilation of data was not a matter of discretion based on Mid America’s personal judgment or taste, but instead it was a matter of convention and strict industry standards”), or (ii) the author made obvious, garden-variety, or routine selections, see Feist, 499 U.S. at 362, 111 S.Ct. at 1296 (concluding that the selection and arrangement of a white pages in which the publisher had chosen to include name, town and telephone number, and to arrange these listings in alphabetical order were entirely “typical” and “garden-variety”); BellSouth Adver. & Publ’g Corp. v. Donnelley Info. Publ’g, Inc., 999 F.2d 1436, 1444 (11th Cir.1993) (m banc) (holding that the categories for the organization of material in a yellow pages directory lacked creativity where many of the selected headings, such as “Attorneys” or “Banks” are so obvious and many others “result from certain standard industry practices”).

Thus, when it comes to the selection or arrangement of information, creativity inheres in making non-obvious choices from among more than a few options. See, e.g., Hearn v. Meyer, 664 F.Supp. 832, 847 (S.D.N.Y.1987) (“Copyright protection is afforded rarely where a fact permits only a narrow continuum or spectrum of expression.”). For example, in Kregos v. Associated Press, 937 F.2d 700, 704 (2d Cir.1991), the decision to express a pitcher’s performance in terms of nine statistics from “at least scores of available statistics about pitching performance available to be calculated from the underlying data and therefore thousands of combinations of data that a selector can choose to include in a pitching form” was not necessarily obvious or self-evident. We therefore concluded that the district court erred in granting summary judgment on the uneopyrightability of the pitching forms. Id. at 704-05, 711; see also American Dental Ass’n v. Delta Dental Plans Ass’n, 126 F.3d 977, 979 (7th Cir.1997) (holding taxonomy of dental procedures creative after noting that they “could be classified ... in any of a dozen different ways”). However, selection from among two or three options, or of options that have been selected countless times before and have become typical, is insufficient. Protection of such choices would enable a copyright holder to monopolize widely-used expression and upset the balance of copyright law.

In sum, creativity in selection and arrangement therefore is a function of (i) the total number of options available, (ii) external factors that limit the viability of certain options and render others non-creative, and (iii) prior uses that render certain selections “garden variety.” See, e.g., 1 Patry, supra, at 196 (“As a general principle, the greater the amount of material from which to select, coordinate, or arrange, the more likely it is that a compilation will be protectible. On the other hand, where less material is available, it is less likely that a compilation, even if original, will be protectible, since de minimis efforts, including selections, are not subject to copyright.”).

C

We proceed to assess the originality and creativity underlying the elements of West’s case reports that HyperLaw seeks to copy.

1. Captions, Courts, and Date Information

West claims that originality inheres in the following enhancements:

• The format of the party names — the “caption” — is standardized by capitalizing the first named plaintiff and defendant to derive a “West digest title,” and sometimes the party names are shortened {e.g., when one of the parties is a union, with its local and national affiliations, West might give only the local chapter number, and then insert etc.). 6

• The name of the deciding court is restyled. E.g., West changes the slip opinion title of “United States Court of Appeals for the Second Circuit” to “United States Court of Appeals, Second Circuit.”

• The dates the case was argued and decided are restyled. E.g., when the slip opinion gives the date on which the opinion was “filed,” West changes the word “filed” to “decided.”

• The caption, court, docket number, and date are presented in a particular order, and other information provided at the beginning of some slip opinions is deleted (such as the lower court information, which appears in the West case syllabus).

We do not think that the district court committed clear error in finding that these changes are insubstantial, unoriginal, and uncreative. Reference to a case by the names of the first plaintiff and first defendant is a garden variety decision. See, e.g., The Bluebook: A Uniform System of Citation rule 10.2.1(a) (16th ed.1996) (hereinafter “The Bluebook”). The same is true of West’s manner of shortening long case names. See, e.g., id. rule 10.2.1(i) (“Cite a union name exactly as given in the official reporter, except that: (i) only the smallest unit should be cited ... (ii) all craft or industry designations ... should be omitted_”). Even if these choices regarding which words to capitalize and shorten to form the West digest title were an original inspiration, we doubt the decisions to shorten the titles or capitalize certain letters would be copyrightable. See Secure Servs. Tech., Inc. v. Time & Space Processing, Inc., 722 F.Supp. 1354, 1363 n. 25 (E.D.Va.1989) (“Size of print ... is not copyrightable.”); 37 C.F.R. § 202.1(a) (1998) (“Words and short phrases such as names [and] titles ... mere variations of typographic ornamentation [or] lettering” are not copyrightable.). Nor does West’s overall choice concerning which procedural facts to include at the start of the case report demonstrate the requisite originality or creativity: The names of the parties, the deciding court, and the dates of argument and decision are elementary items, and their inclusion is a function of their importance, not West’s judgment. Cf. Kregos, 937 F.2d at 702 (noting that “there can be no claim of a protectable interest in the categories of information concerning each day’s [baseball] game,” including the teams, starting pitchers, the game time, and the betting odds).

2. Attorney Information

The second claimed creative element is West’s selection and arrangement of attor ney information. The Supreme Court slip opinions and some slip opinions of the courts of appeals omit some or all of the information about counsel that West compiles from the docket sheets and from other sources. For Supreme Court opinions, West lists the arguing counsel and the lawyer’s city and state of practice; for court of appeals decisions, West lists the names of briefing attorneys as well as arguing attorneys, and specifies each lawyer’s city and state of practice and law-firm or agency affiliation.

Here again we agree with the district court that West’s decisions lack a modicum of creativity. Like the name, town and telephone number included in Feist’s telephone directory, the information West includes — attorney names, firms and cities of practice — is entirely “typical” and “garden-variety.” See, e.g., Skinder-Strauss Assocs. v. Massachusetts Continuing Legal Educ., Inc., 914 F.Supp. 665, 676 (D.Mass.1995) (noting that “[i]n compiling a Massachusetts directory of lawyers and judges, ... [t]he ‘selection’ of other directory data, including the attorney name, address, telephone and fax numbers, year of bar admission, and so forth are ... unoriginal and determined by forces external to the compiler”); cf. Key Publications, Inc. v. Chinatown Today Publ’g Enters., Inc., 945 F.2d 509, 514 (2d Cir.1991) (emphasizing the creativity inherent in designating phone-book classifications of particular interest to Chinese-Americans). In fact, most courts (this one included) provide the very same information in their slip opinions.

West’s decision to provide more information about counsel in the court of appeals case reports, and less in the Supreme Court case reports does not strike a creative spark because the options available to a publisher are simply too limited. West’s claim illustrates the danger of setting too low a threshold for creativity or protecting selection when there are two or three realistic options: West lists only the arguing attorneys and city of practice, while United States Law Week lists the arguing and briefing attorneys, their firm affiliations and city and state of practice. If both of these arrangements were protected, publishers of judicial opinions would effectively be prevented from providing any useful arrangement of attorney information for Supreme Court decisions that is not substantially similar to a copyrighted arrangement.

3. Subsequent History

West’s case reports reflect certain subsequent procedural developments, such as orders amending an opinion or denying rehearing. The district court found that West’s alteration of opinions to reflect these subsequent case developments does not reflect an exercise of originality or creativity, in part because West’s realistic options are limited. We cannot say that this was clear error.

In most appeals, the only subsequent development is a denial of rehearing. West’s manuscript manager Bergsgaard described two possible ways to reflect this action: (i) a file line, inserted at the beginning of the case just after the date of the original decision, which simply states “rehearing denied” and the date of the denial, or (ii) a table containing the same information. West has chosen to reflect denial of rehearing via a file line. Neither this choice, nor the actual language used to reflect the denials (“rehearing denied” followed by the date), is creative or requires judgment.

As the trial testimony reflects, subsequent orders sometimes reflect more extensive changes or additions to opinions, such as amendments to the original opinion (ranging from minimal to extensive), or subsequent opinions upon denial of rehearing or in dissent from denial of rehearing or rehearing in banc. The actual text of any amendments to opinions or new opinions issued with the denial of rehearing will always be the court’s. But West points to the following available options for reflecting these changes, and claims that its choice from among these options is creative: (i) printing the order in full at the end of the opinion; (ii) altering the text of the opinion to reflect the amendments; (iii) publishing the order separately from the original opinion in a different volume and cross-referencing to the original opinion; or (iv) reprinting the original opinion in full with the changes reflected in the text.

Almost never will these decisions present more than one or two realistic or useful options, and almost always the choice among them will be dictated by the timing of the court action. For example, an order amending an opinion will either (i) reflect the exact locations in the opinion where changes are to be made, in which case the obvious preference is to alter the text of the opinion, or (ii) include some general language modifying the opinion, in which case the obvious preference is to print the order at the end of the text of the opinion. If the opinion has not yet gone to print in an advance sheet or a bound volume, West will actually make these changes in the original opinion. If the opinion has already been printed in the bound volume, then West (once again) has only binary options: (i) print only the order and cross-reference to the original opinion, or (ii) reprint the opinion in full, incorporating the changes and/or publishing the order at the end of the opinion. As set forth in the margin, West’s exemplar confirms how little judgment is exercised even in a case with a complicated subsequent procedural history. 7 West also adds a file line explaining the action, but the line merely sets forth the court’s action. Given the few practical options available to West, and the fact that the choice among these narrow options will be dictated by the timing and nature of the court’s action, we do not think the district court clearly erred in determining that this element of West’s case reports does not demonstrate sufficient creativity to be protecta-ble.

A Parallel or Alternate Citations

As the district court recognized, the element of West’s ease reports that raises the closest question as to creativity is West’s emending of the citations, as follows:

(a) West inserts parallel citations when the judicial opinion does not, e.g., (i) for citations to Supreme Court opinions, it inserts parallel citations to United States Reports, Supreme Court Reporter, and Lawyer’s Edition, (ii) for citations to a state court, West inserts parallel citations to the official reporter and West’s regional reporter, and (iii) for citations to looseleaf, specialized, or electronic reporters, West inserts parallel citations to a West National Reporter System (“NRS”) reporter or Westlaw (West’s on-line database);

(b) West substitutes some court citations, e.g., if the court cites to a slip opinion that has been published in an NRS reporter, West will substitute a citation to the NRS reporter; and

(c) West adds citations when the opinion refers to a case by name but does not insert a citation.

The district court concluded that “[i]n most instances the determination of which parallel citations to include ... reflect[s] no level of originality,” and that the “selections made tend to conform to the standard of the legal profession and appear consistent with those recommended in A Uniform System of Citation.” Matthew Bender & Co., 1997 WL 266972 at *4.

We cannot find that the district court’s conclusion was unreasonable. West claims that it exercises careful judgment as to which sources are most useful to legal practitioners. However, almost every one of West’s decisions relating to citation alterations is inevitable, typical, dictated by legal convention, or at best binary. See 1 Patry, supra, at 196-97 (“Even where theoretically there is a large number of items to choose from, functional, commercial, or legal constraints may limit, or even bar, protectibility.”)- And each case report exhibits only one or two decisions on how to alter citations.

West has issued a series of memoranda to its editors that contain guidelines for citation alterations (a complete list of the eitational instructions is set out in the margin 8 ). Most of West’s citation guidelines need no discussion because they involve obvious, garden-variety decisions to cite to West NRS case reporters and to Westlaw whenever the court has cited to sources that are not easily accessible, such as looseleaf or daily or weekly reporter services or slip opinions; there are few options to begin with, and West’s case reporters and Westlaw have the widest availability and have essentially become the standard citation to case law. See, e.g., The Bluebook, supra, at 165-225 (recommending citation to West federal and regional reporters for all federal and state courts, with the exception of the Supreme Court, for which Supreme Court Reporter is listed as a secondary cite); id. rule 10.3.1 (recommending citation to West regional reporters, and if not published therein, to a “widely used computer database” or to a “service,” in that order of preference). No evaluative judgment is involved in a decision to cite to those two sources (especially given West’s self-interest in including citations to its own products).

As for the other guidelines, the following represent the citation decisions that receive the most emphasis by West, and that represent most, if not all, of the examples of citation alterations included in cases that West offered into evidence at trial:

• West’s decision to insert a citation to the denial of certiorari only when the denial pre-dated the opinion is necessary to avoid anachronism, and is in any event a choice among two or three options at most. 9 See, e.g., BellSouth Adver. & Publ’g Corp. v. Donnelley Info. Publ’g, Inc., 999 F.2d 1436, 1441 (11th Cir.1993) (in banc) (holding that determination of closing date after which no changes were included in the yellow pages at issue did not demonstrate creativity in selection because “any collection of facts ‘fixed in any tangible medium of expression’ will by necessity have a closing date” (footnote omitted)).

• As proof of its creativity in electing to parallel-cite to United States Reports, Supreme Court Reporter, and Lawyer’s Edition, West lists eight other reporters of Supreme Court opinions that do not parallel-cite to these sources, and to which West does not parallel-cite. But all of these reporters are daily or weekly updates intended to provide quick copies of the slip opinions, not to serve as research tools or permanent records, and they are not recognized as standard sources for citation in the legal profession. West’s decision to omit parallel citation to those reporters requires no evaluative judgment, and the decision by the publishers of those reporters and services to omit parallel citation to the same reporters cited by West reflects the more limited purpose of those reporters rather than the creativity of West’s own choice. 10 In the end, West decided to include every permanent record of Supreme Court opinions — i.e., it makes no “selection” at all.

• Nor do we see any creativity in West’s decision to cite to official state reporters as well as regional NRS reporters for state court decisions. These are almost always the only two realistic choices (again, this hardly amounts to “selection”), and a decision to cite to an official reporter can hardly be said to be anything other than typical. 11

One useful way to appreciate how little creativity inheres in West’s citation decisions is to consider what West’s competitors would have to do to avoid an infringement claim were we to find West’s citation decisions copyrightable. Competitors such as Hyper-Law seeking to create a useful case report would need to engage in their own original selection of parallel and alternate citations. But while some generally useful information which does not appear in West’s case reports could be included (such as citations to LEXIS instead of Westlaw), most of the information a researcher would find useful (such as citations to West NRS reporters) already have been added to West’s case reports; West gives few examples of other useful supplementary information. A competitor that included these alterations, notwithstanding their inclusion in West’s case reports, could have no confidence that an infringement claim could be avoided, especially given our warning in Key Publications that infringement cannot be avoided by pointing to isolated differences from a copyrighted work. See Key Publications, Inc. v. Chinatown Today Publ’g Enters., Inc., 945 F.2d 509, 514 (2d Cir.1991) (holding that a compilation copyright is not so “thin” as to be “anorexic,” and does not allow “subsequent compilers to avoid infringement suits simply by adding a single fact to a verbatim copy of the copyrighted compilation, or omitting in the copy a single fact contained in the copyrighted compilation”). One way of saying that West’s “choices” are obvious and typical is that a competitor would have difficulty creating a useful case report without using many of the same citations. Affording these decisions copyright protection could give West an effective monopoly over the commercial publication of case reports (at least those containing supplemental citations).

Nor do we think the district court erred in concluding that the combination of these citation decisions is unprotectable. West’s particular decisions about which parallel citations to insert are driven in each instance by the court’s decision to cite to a certain case, and thus each editorial choice is independent of the others. The cumulative effect of these citation decisions is a piling up of things that are essentially obvious or trivial (albeit helpful), each in its discrete way in its discrete spot. The whole does not disclose or express an overall creative insight or purpose, such as a set of statistics that together allow the ranking of a group of ball players, or a designation or highlighting of phone numbers that together allow the user of a phone book to enjoy an unusual or particular convenience. The combined effect of West’s non-creative citation decisions cannot be said to be creative, on such a theory or any other theory that West advances. Ve conclude that the district court did not clearly err in deciding that West’s citation alterations display insufficient creativity to be protectable.

Finally, West’s overall decision to add attorney information, subsequent history, and additional citation information exhibits little, if any, creative insight; most courts already provide attorney information, and opinion accuracy mandates inclusion of subsequent history.

West’s editorial work entails considerable scholarly labor and care, and is of distinct usefulness to legal practitioners. Unfortunately for West, however, creativity in the task of creating a useful case report can only proceed in a narrow groove. Doubtless, that is because for West or any other editor of judicial opinions for legal research, faithfulness to the public-domain original is the dominant editorial value, so that the creative is the enemy of the true. 12

Our decision in this case does not mean that an editor seeking to create the most accurate edition of another work never exercises creativity. 13 As West argues, our deci sions establish a low threshold of creativity, even in works involving selection from among facts. But those cases involved the exercise of judgments more evaluative and creative than West exercises in the four elements of the ease reports that HyperLaw intends to copy. For instance, in Kregos thousands of different permutations of pitching statistics were available for inclusion in the publisher’s pitching chart. See Kregos, 937 F.2d at 704; see also Eckes v. Card Prices Update, 736 F.2d 859, 863 (2d Cir.1984) (baseball card guide which selected 5,000 “premium” baseball cards from among 18,000 eligible baseball cards was copyrightable). In Key Publications, we found sufficient creativity because the author of the yellow pages “excluded from the directory those businesses she did not think would remain open for very long.” 945 F.2d at 513. In CCC Information Services, we found sufficient creativity in the selection of optional car features and number of years’ models to be included in a used-car price compilation. 44 F.3d at 67. And in Lipton v. Nature Co., 71 F.3d 464 (2d Cir.1995), the author “selected [the terms included in the work] from numerous variations of hundreds of available terms.” Id. at 470. In each of these cases, the compiler selected from among numerous choices, exercising subjective judgments relating to taste and value that were not obvious and that were not dictated by industry convention. See Mid America Title Co. v. Kirk, 59 F.3d 719, 723 (7th Cir.1995) (distinguishing cases involving exercise of subjective judgment). 14

Similarly, in Weissmann v. Freeman, 868 F.2d 1313 (2d Cir.1989), a textual derivative-work case, we found sufficient creativity where the author of the derivative work had drawn on earlier joint works with another professor to create a document that contained the following new elements from the previous version of the document:

(1) a selection and arrangement of photo illustrations and associated captions; (2) references to recent reports in the pertinent literature; (3) selection, condensation, and description of additional source material; (4) several new textual additions; (5) substantial rearrangement of the manner and order of presentation of material contained in the parties’ prior joint works; and (6) the addition of a section on “congenital disorders,” a revised treatment of “chronic cholecystitis,” and the incorporation of Dr. Freeman’s “false positive” studies.

Id. at 1322. As this passage demonstrates, the alterations inserted in the derivative work were by no means obvious or driven by professional convention, and resulted in substantial changes to the substance and flow of the piece. No such substantial variations characterize West’s case reports.

CONCLUSION

The district court did not clearly err in concluding that the elements of West’s case reports that HyperLaw seeks to copy are not copyrightable. The judgment of the district court is affirmed.

SWEET, District Judge:

The key issue in this appeal is whether West’s Supreme Court Reporter and Federal Reporter case reports in the context of its overall reporter citation system meet the constitutional and statutory requirement of creative originality. Because the majority imposes a standard that demands significantly more than the “modicum” of originality required by Feist Publications, Inc., v. Rural Telephone Service Company, Inc., 499 U.S. 340, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991), and far more than the “non-trivial” variation required by this Court for derivative-work and compilation copyright protection, I respectfully dissent. 1

The Copyright Act protects both derivative works and compilations, and I agree with the majority that West’s case reports have elements of both. The standard for copyrightability set forth in Feist, as the majority notes, is applicable whether West’s editorial work is analyzed in terms of derivative work or compilation. See e.g., Atari Games Corp. v. Oman, 979 F.2d 242, 244-45 (D.C.Cir.1992) (Feist applicable to [audio-visual] compilation); 2 W. Patry, Copyright Law and Practice 1225 (1994) {Feist applicable to derivative works). Contrary to the majority’s holding, however, I find that West’s selection and arrangement of factual annotations to public domain judicial opinions, considered as a whole, is copyrightable.

Originality alone — whether the “author make[s] the selection or arrangement independently (i.e. without copying that selection or arrangement from another work)” — is not sufficient. Feist, 499 U.S. at 358, 111 S.Ct. 1282. The work must also “display some minimal level of creativity.” Id. Creativity for copyright purposes is not a philosophical question: the “creative spark” need only pass “the narrowest and most obvious limits.” See Bleistein v. Donaldson Lithographing Company, 188 U.S. 239, 251, 23 S.Ct. 298, 47 L.Ed. 460. The “modicum of creativity” requires simply that the author prove “the existence of ... intellectual production, of thought, and conception.” Feist, 499 U.S. at 362, 111 S.Ct. 1282 (quoting Burrow-Giles Lithographic Co. v. Sarony, 111 U.S. 53, 59-60, 4 S.Ct. 279, 28 L.Ed. 349 (1884)); see also Key Publications Inc. v. Chinatown Today Publishing Enterprises, Inc., 945 F.2d 509, 514 (2d Cir.1991). {“[D]e minimis thought withstands originality requirement”).

Thus, while the majority is correct that it is “not a goal of copyright law” to encourage the creation of compilations which lack “sufficient creativity,” it is well-established that the required level of creativity is “extremely low.” Feist, 499 U.S. at 345, 111 S.Ct. 1282. 2

In Feist, Rural’s alphabetical arrangement by surname, and selection of name, town and telephone number to include in a telephone book, was “practically inevitable.” Id. at 363, 111 S.Ct. 1282. Although the facts need not be presented in an “innovative or surprising way,” Rural’s choice was “so mechanical or routine as to require no creativity whatsoever.” Feist, 499 U.S. at 362, 111 S.Ct. 1282. Indeed, the copyright claimant in Feist had no real choice how to arrange a white pages directory, particularly given that state law prescribed the selection of data, and that only a few basic decisions were involved. Id. Also significant was that only Rural possessed the underlying data. Id. at 363, 111 S.Ct. 1282.

Here, West has made choices to make its reporters and its citation system valuable. West makes dozens of multi-part, variable judgments, and there is no evidence that any of West’s choices are commonplace, “practically inevitable,” dictated by law, or that they follow any external guidelines. On the contrary, the record demonstrates that West makes a number of substantive, editorial choices — without court direction or approval — in determining the content and expression of its ease reports. West’s judgments involve assessments of “readability,” clarity, completeness, availability (present and future) of sources, and other subjective considerations related to making the reports more useful.

Specifically, West asserts that originality inheres in the following aspects of its editorial process: (1) its decisions about when to add parallel citations and which parallel citations to add; (2) its substitution of “alternative citations” when its editors deem that the original citation should be improved upon in terms of usefulness, currency, or accuracy; (3) its addition of its own citations when none are provided; (4) its internal revision and correction of citations; (5) its expansion and completion of citation page references; (6) its creation, selection, and arrangement of additional text to reflect subsequent case developments; (7) its selection and arrangement of data for attorney summaries, and (8) its revision and reorganization of captions, court lines, date lines, and other prefatory material. Because these choices express thought and are not inevitable, West’s annotations do not fall in the “narrow category” of works which are not copyrightable.

The fact that federal judges publish written opinions differently than West is sufficient reason to conclude that West’s version requires some “thought” and is sufficiently “creative” to satisfy the modicum necessary for copyrightability. If a federal judge chooses to cite only to the United States Reporter, include minimal attorney information in his or her written opinion, or not provide a cite for a referenced case, then an alternative choice to provide parallel citations, expand attorney information, and cite the case cannot be deemed so “typical,” “garden-variety,” “obvious” or “inevitable” to prohibit copyrightability. Cf. Feist, 499 U.S. at 362-63, 111 S.Ct. 1282.

The majority dissects each element of West’s editorial process and then extrapolates that “the cumulative effect of these citation decisions is a piling up of things that are essentially obvious or trivial (albeit helpful), each in its discrete way in its discrete spot.”

West’s originality, however, cannot be determined by the sum total of whatever (creativity) remains after each individual component is atomized. Indeed, this Court recently warned against the dangers of basing copyrightability analysis on an approach which isolates each element or ignores the “protectible expression within an unprotecti-ble element.” Softel, Inc. v. Dragon Medical and Scientific Communications, Inc., 118 F.3d 955, 964 (2d Cir.1997). See also 3 M. Nimmer, Copyright § 13.03[F][5], at 13-145 n.345.1 (explaining that the fact that Hamlet’s soliloquy can be reduced into un-proteetible words does not mean that the soliloquy as a whole lacks originality for copyright purposes).

West’s selection of particular annotations for each case must be considered a whole, not individually. See Key Publications, 945 F.2d at 514 (issue is “whether the arrangement ... viewed in the aggregate, is original”). In defining a derivative work, the issue is whether “modifications” “represent an original work of authorship” must be considered “as a whole.” 17 U.S.C. § 101 (emphasis added) (definition of derivative work). The same legislative command is repeated in the definition of compilations, which provides that a compilation is “a work formed by the collection and assembling of preexisting materials ... that are selected, coordinated, or arranged in such a way that the resulting work as a lohole constitutes an oiiginal work of authorship.” 17 U.S.C. § 101 (emphasis added). The cumulative and collective originality manifest in West’s case reports satisfies the “de minimis ” level needed for the work as a whole to be copyrightable. In deed, it must be assumed that the originality and the consequent utility of the West citation system is precisely the reason that Hy-perlaw seeks to use West as its verbatim source.

In my view the decision of the majority is not consistent with the post -Feist case Key Publications, 945 F.2d at 509. In Key Publications, this Court held that classified directories of Chinese-American businesses was copyrightable because, among other things, plaintiff excluded enterprises she believed would not remain in business. This selection “indicates thought and creativity in the selection of businesses included.” Id. at 513. If there is a modicum of originality in deciding which businesses are likely to stay open for awhile, the test is surely also met by deciding in a system designed to assist legal research for example, which sources are sufficiently useful, available, or permanent to stand alone, and which require the addition of an electronic parallel citation.

Contrary to the majority’s view, the alternative factual annotations selected by West are not comparable to the cases where courts have denied copyright protection based on lack of originality. In Victor Lalli Enterprises, Inc. v. Big Red Apple, Inc., 936 F.2d 671 (2d Cir.1991), it was undisputed that the publisher’s selection and arrangement of fact categories was exactly the same as that of all racing-chart publishers. Id. at 672. In Financial Information, Inc. v. Moody’s Investors Service, Inc., 808 F.2d 204 (2d Cir.1986), cert. denied, 484 U.S. 820, 108 S.Ct. 79, 98 L.Ed.2d 42, it was beyond dispute that the five basic facts listed on the Daily Bond Cards were an “inevitable” choice devoid of originality. In this case, by contrast, all of West’s basic choices involve subjective judgment.

In Skinder-Strauss Associates v. Massachusetts Continuing Legal Education, Inc., 914 F.Supp. 665 (D.Mass.1995), the court held that the legal directory is copyrightable “as a whole,” and therefore the question was substantial similarity, not copyrightability. Id. at 677. Although the Skinder-Strauss court held that individual elements, such as a calendar of Christian and Jewish holidays, were not copyrightable, here West does not seek copyrightability of individual facts, but rather seeks to prevent verbatim copying of the case report as a whole.

The copyright granted West is thin, but it is sufficient to protect against the verbatim digital copying proposed by Hyperlaw. This result protects the advancement of science and the arts, while not permitting Hyperlaw to undermine any incentive for West to annotate judicial opinions selectively. If West’s competitors were authorized to scan West’s editorial enhancements systematically and, in effect, to copy its citation system, the economic incentive to engage in this kind of original and productive enterprise would largely evaporate.

There is no danger here that granting West’s copyright protection to its annotations provides them a monopoly over the “idea” of publishing judicial opinions. When the number of ways data can be organized is so limited that its expression merges with the idea, copyright may be denied. However, here “there are a sufficient number of ways of expressing the idea ... to preclude a ruling that the idea has merged into its expression.” Kregos v. Associated Press, 937 F.2d 700 (2d Cir.1991). In Kregos, this Court held that “the past performances of baseball pitchers can be measured by a variety of statistics,” and is copyrightable. Here, too, opinions can be, and are, written with a variety of citation combinations and other facts either included or not. This case is not like Matthew Bender & Co. v. Kluwer Law Book Publishers, Inc., 672 F.Supp. 107 (S.D.N.Y.1987), where the court concluded that the categories in the plaintiffs chart (amount, case, plaintiff event, injury, and relevant data) are “the only sensible ones which could have been used to compile the data.” Id. at 112.

To the extent that the West selection of factual annotation may seem obvious to anyone familiar with legal sources, it may be because of West’s success in the market. 3 There is no support for the proposition that West’s success in achieving an “industry standard” citation arrangement obligates them to donate the material to the public domain. Cf. BellSouth, 999 F.2d at 1444 (industry standard copied from industry association).

For the reasons stated, I conclude the summary judgment granted in favor of Hy-perlaw should be reversed.

1 . West also functions as the publisher of slip opinions — i.e., the official versions of court opinions — for the Fifth and Eleventh Circuits, and West does not claim a copyright for any alterations it makes to those slip opinions other than its syllabi, headnotes, and key numbers.

2 . The record seems to indicate that since the commencement of this suit, HyperLaw has combined these two products into one.

3 . Whether insertion of star pagination to West’s case reporters amounts to copying of West’s arrangement of cases is a separate question we address in the other opinion issued today. See Matthew Bender & Co. v. West Publ’g Co., 158 F.3d at 693.

4 . West initially claimed some creativity in its corrections to the text of opinions, but it has abandoned this claim, presumably because these corrections either are trivial (i.e., punctuation or spelling), or else (nearly always) approved by the courts by order or informal means.

5 . We have previously explained that the protection of compilations “is consistent with the objectives of the copyright law ... to promote the advancement of knowledge and learning by giving authors economic incentives ... to labor on creative, knowledge-enriching works” because compilations “that devise new and useful selections and arrangements of information ... contribute to public knowledge by providing cheaper, easier, and better organized access to information.” CCC Info. Servs., Inc. v. Maclean Hunter Mkt. Reports, Inc., 44 F.3d 61, 65-66 (2d Cir.1994). Without that financial incentive, such compilations would not be created. Id. at 66. Nevertheless, it is not a goal of copyright law to promote the production of compilations which lack sufficient creativity.

6 . West offered as evidence at trial memoranda to its editors concerning caption alterations. Review of these memoranda demonstrates that they are simply concerned with accurate presentation of caption information. For example, they clarify the accurate family name for parties of Chinese or Spanish origin so that the editor capitalizes the proper name in the caption. They also set forth a uniform system for abbreviating certain words, which is not original or copyrightable. See, e.g., Brief English Systems, Inc. v. Owen, 48 F.2d 555, 556 (2d Cir.1931) (“There is no literary merit in a mere system of condensing written words into less than the number of letters usually used to spell them out.”).

7 . West focuses on its case report for Roulette v. City of Seattle, decided in March 1996 and published at 78 F.3d 1425 (9th Cir.1996). In September 1996, there issued an order amending the opinion upon denial of rehearing and suggestion for rehearing in banc, and two opinions dissenting from the denial of rehearing in banc. The order amending the opinion made substantial substantive changes to the opinion, and so West decided to republish the full opinion with the changes reflected in the text at 97 F.3d 300 (9th Cir.1996). Moreover, it printed the full order (with a note indicating that the amendments to the opinion were included in the text) with the dissenting opinions at the end of the opinion. Finally, it added a file line reflecting those developments. None of these decisions demonstrates the exercise of any creativity; in fact, with the exception of the file line, all were dictated by the changes made to the opinion by the court. The Ninth Circuit did the same thing, i.e., republished the original opinion incorporating the amendments and printed the order and dissents in full following the text.

8 . West has issued the following guidelines for citation alterations:

(1) follow certain prescribed formats for abbreviation of Rules of Civil Procedure, state rules, and statutes.

(2) add a citation to Westlaw or to an NRS case reporter for citations to certain looseleaf services when the case cannot be directly retrieved through Westlaw using the looseleaf service citation.

(3) substitute an NRS case reporter cite for 15 publications, almost all of which are daily or weekly reports or journals, including: Arizona Advance. Reports, California Daily Appellate Report, 9th Circuit Daily Journal, Florida Law Weekly, Oklahoma Bar Journal, Texas Supreme Court Journal, and United States Law Week.

(4) add an NRS case reporter or Westlaw cite for 15 publications, some of which are topical publications and others of which are competitor publications, including: American Law Reports (A.L.R.), Board of Tax Appeals, Federal Sentencing Reporter, Idaho Bankruptcy Court Reports, Pennsylvania Fiduciary, and Virgin Islands Reports.

(5) for citations to cases appearing in tables in West case reporters, include a Westlaw cite if Westlaw contains the full-text of the summary order.

(6) if the court's citation indicates that a petition for certiorari has been filed and the Supreme Court has acted on the petition:

(a) if the petition has been denied and the date of the denial was prior to the date the opinion was decided, add the citation for the denial of certiorari.

(b) if the petition was denied after the opinion was decided, do not add a citation for the denial of certiorari.

(c)if the petition was granted, check with the court to see if the case citation should be retained.

(7) if a court includes a short form citation for a case which has not yet been cited in full in the opinion, insert the full citation.

(8) if a court cites to a slip opinion, add the NRS case reporter citation. If this citation is unavailable prior to publication, add a West-law citation.

(9) add extension pages for citations if the court’s NRS citation includes an extension page.

(10) change internal page references in the slip opinion to dashes so that a reporter page number can he inserted by West editors.

(11) add a Westlaw or NRS case reporter citation for public utility citations.

(12) do not remove citations in opinions to public domain citation systems in those jurisdictions that have adopted such citation systems, including the Sixth Circuit and South Dakota.

(13) add a Westlaw citation when the court includes a LtíXIS citation, and in certain situations, remove the LEXIS citation.

(14) add a citation for a case referenced in the text of the opinion without citation unless that case appears on a list of 300 popular case names.

(15) add parallel citations for Supreme Court decisions to United States Reports, Supreme Court Reporter, and Lawyer's Edition.

(16) add parallel citations for state court decisions to an NRS regional case reporter and the state’s official reporter.

9 . Even West notes in its brief, "Every publisher has at least three options [on how to update citations in slip opinions] — to leave citations as they appear in the opinion, to update them as of the date of case-report publication, or to update them only as of the date of the original order."

10 . West maintains a list of 300 cases deemed by West to be so well known that West will leave intact one-parly references to these cases without expanding the case citation. This practice involves an undeniably creative, but minor, insight into which cases are within an ordinary lawyer's frame of reference, and the expression of that selection would be protectable. Nevertheless, West offered no evidence as to how often these cases are cited without a full citation in the judicial opinions that HyperLaw seeks to copy and we will not find infringement for what seems to be de minimis copying of protected material. Cf. Warner Bros. Inc. v. American Broad. Cos., 720 F.2d 231, 242 (2d Cir.1983) (noting that de minimis rule permits "the literal copying of a small and usually insignificant portion of the plaintiff's work”).

11 . West emphasizes that The Bluebook's rule dispenses with citation to official reporters, but it ignores another Bluebook rule that practitioners cite to official state reporters for cases decided by the state in which the practitioner is filing papers. See The Bluebook, supra, rule P.3. Furthermore, another citation guide recommends citing to both the official reporter and the NRS reporter on all occasions. See The University of Chicago Manual of Legal Citation 15 (1989) ("When citing to a state case, indicate the volume and first page of the case for both the official and commercial reporters.”).

12 . HyperLaw argues that we should apply the merger doctrine to bar the copyrightability of West's alterations to judicial opinions.

The fundamental copyright principle that only the expression of an idea and not the idea itself is protectable has produced a corollary maxim that even expression is not protected in those instances where there is only one or so few ways of expressing an idea that protection of the expression would effectively accord protection to the idea itself.

Kregos v. Associated Press, 937 F.2d 700, 705 (2d Cir.1991) (citation omitted). HyperLaw claims that the idea of a case report has effectively merged with West’s expression of that concept.

We decline to invoke the merger doctrine in this case. First, "[o]ur Circuit has considered this so-called 'merger' doctrine in determining whether actionable infringement has occurred, rather than whether a copyright is valid." Id.; see also CCC Info. Servs., Inc. v. Maclean Hunter Market Reports, Inc., 44 F.3d 61, 72 n. 26 (2d Cir.1994) ("In this circuit, consideration of the merger doctrine takes place in light of the alleged copying to determine if infringement has occurred, rather than in analyzing the copyright-ability of the original work.”). In addition, under this approach, "if a defendant has actually copied the plaintiff's work, it is unlikely to be allowed to rely on merger to avoid liability,” Kregos, 937 F.2d at 716 (Sweet, J., concurring in part and dissenting in part). HyperLaw seeks a declaratory judgment holding West’s case reports non-copyrightable and allowing it actually to copy West's case reports. It therefore cannot avail itself of the merger doctrine.

Second, West's work does not constitute a "building block[] of understanding,” which we have indicated is the type of expression to which we will consider applying the merger doctrine. See CCC Info. Servs., 44 F.3d at 71.

13 . It is true that some types of editing require little creativity. See, e.g., Grove Press, Inc. v. Collectors Publication, Inc., 264 F.Supp. 603, 605 (C.D.Cal.1967) (“Plaintiff made approximately forty thousand changes from the Verlag copy in producing its edition. These changes consisted almost entirely of elimination and addition of punctuation, changes of spelling of certain words, elimination and addition of quotation marks, and correction of typographical errors. These changes required no skill beyond that of a [1967] high school English student and displayed no originality. These changes are found to be trivial.”). In addition, convention and external forces may, as here, limit the practical choices available so as to eliminate any creativity.

On the other hand, preparing an edition from multiple prior editions, or creating an accurate version of the missing parts of an ancient document by using conjecture to determine the probable content of the document may take a high amount of creativity. See, e.g., Abraham Rabino-vich, Scholar: Reconstruction of Dead Sea Scroll Pirated, Wash. Times: Nat'l Wkly. Edition, Apr. 12, 1998, at 26 (discussing scholar’s copyright infringement claim in Israeli Supreme Court relating to his reconstruction of the missing parts of a "Dead Sea Scroll” through the use of "educated guesswork” based on knowledge of the sect that authored work).

14 . HyperLaw argues also that West has failed to comply with the requirement of 17 U.S.C. § 403 that it identify those parts of the government work in which it does not claim a copyright. Because we find that the elements of West’s case reports that HyperLaw seeks to copy are not protectable, we need not reach this issue.

1 . Although not dispositive since I believe that the district court’s determination is clearly erroneous, review should be de novo. Whether West’s annotations are copyrightable requires application of the legal standard imposed by the constitution and copyright statute to the undisputed facts. To the extent prior Second Circuit opinions have reviewed issues of copyrightability for clear error, they appear inconsistent with the Supreme Court's de novo review of a similar issue in Feist (impliedly, although not expressly, reversing district court after de novo review), inconsistent with the de novo standard applied by this Court in other mixed questions in copyright law, see e.g. American Geophysical Union v. Texaco Inc., 60 F.3d 913, 918 (2d Cir.1994) (fair use is mixed question subject to de novo review); Carter v. Helmsley-Spear, Inc., 71 F.3d 77, 85 (2d Cir.1995) (work for hire status legal conclusion reviewed de novo, although factual finding of each relevant factor reviewed for clear error); Folio Impressions, Inc. v. Byer California, 937 F.2d 759, 766 (2d Cir.1991) (substantial similarity reviewed de novo since credibility not at issue), or distinguishable, see Financial Information, Inc. v. Moody’s Investors Service, Inc., 808 F.2d 204, (2d Cir.1986) (noting that district court relied “on its evaluation of the credibility of witnesses — which we are ill-disposed to disturb on appeal” whereas here facts are not in dispute).

2 . Indeed, ”[m]ost applications of Feist have recognized the circumscribed sphere to which its holding applies, ruling that it invalidates the copyright only in the most banal of works, such as the white pages of a phone book." 1. M. Nimmer, Copyright § 3.04[B][2], p.3-33 (footnotes omitted).

3 . The contention that all of West’s enhancements are trivial is somewhat ironic given that what motivates this litigation, it is assumed, is the desire to make money by copying West's valuable editorial work.