1 West is more properly referred to as Thomson West, or West, a Thomson business, after the acquisition of the West Company by Thomson in 1996 but in this article I will use the shorthand form of the company's name familiar to most lawyers.Google Scholar

2 Although pre-indexed research is often thought of as book-based research, because researchers are familiar with the process of working through volumes of regional, federal, and decennial digests, pre-indexed research is available as part of both the Westlaw and LexisNexis legal databases.Google Scholar

3 West's key numbers are the form of pre-indexing most familiar to most lawyers, but other forms of pre-indexing – most notably the annotation approach of the American Law Reports, or A.L.R. – are still available to legal researchers. A recent decision by West, however, means that A.L.R. annotations are less available than they once were. Whereas A.L.R. annotations have been available in both LexisNexis and Westlaw, West has decided to remove the annotations from LexisNexis from January, 2008, meaning that they will only be available in print form and on Westlaw. Press Release, American Law Reports and Westlaw: An Exclusive Arrangement Highlighted in New Westcast Podcast (August 10, 2007), available at http://www.thomson.com/content/pr/tlr/tlr_legal/230857 (accessed September 6, 2007).Google Scholar

4 Boolean logic is a term honoring George Boole, a British mathematician whose work into symbolic logic proved invaluable to the computer scientists who first developed the relational approach to obtaining information from a computer database. Marilyn Walter, Retaking Control of Legal Research, 43 J. Legal Educ. 569, n.l (1993).Google Scholar

5 A discussion of the benefits and disadvantages of these two research methods can be found at notes 59–108 and accompanying text.Google Scholar

6 See, e.g., Amy E. Sloan, Basic Legal Research: Tools And Strategies 340 (3d ed. 2006 (“For many research projects, a combination of [print and electronic sources] will be necessary for complete, accurate, and efficient research.”)Google Scholar

7 Legal researchers performing pre-indexed searches can research the legal issues defined by the indexers with efficiency and ease, but the approach has recognized and significant limitations. See, infra., notes 74 – 82 and accompanying text.Google Scholar

8 John West's digesting sets were first published in 1897, completing the process of systematized legal research begun by West in 1879 with the publication of his Northwestern reporter and continued by the inclusion of the national reporter system, covering all state and federal jurisdictions, which was completed over the next ten years. Lynn Foster & Bruce Kennedy, The Evolution of Research: Technological Developments in Legal Research, 2 J. App. Prac. & Proc. 275, 276–77 (2000).Google Scholar

9 The effects of this move away towards computer-assisted legal research and away from book-based legal information have been recognized for some time. See, e.g., Gary J. Bravy & K. Celeste Feather, The Impact of Electronic Access on Basic Library Services: One Academic Law Library's Experience, 93 Law Libr. J. 261 (2001)(tracking precipitous declines in both student photocopying and in book reshelving, both indicators that students are using computer-assisted legal research and book-based legal information less); Erica V. Wayne & J. Paul Lomio, Book Lovers Beware: A Survey of Online Research Habits of Stanford Law Students, 6–7 (Robert Crown Law Library Legal Research Paper Series, Research Paper No. 2) (2005)(reporting a three year increase in the number of first year students who reported doing all or most of their legal research online). See also, infra, n. 47 (description of surveys of practicing lawyers suggesting increasing reliance on computer-assisted legal research among practitioners).Google Scholar

10 The move away from print-based legal information has at least two identifiable causes. First, of course, is the move to the internet and away from books that has been identified among practicing lawyers (see, e.g., supra at n. 9). The second cause of this move is bottom-line driven. Put simply, the cost of maintaining a print library is increasing, both in terms of the books themselves and in terms of the dollar cost associated with their storage. For an analysis of these costs, see, Ian Gallacher, Forty-Two: The Hitchhiker's Guide to Teaching Legal Research to the Google Generation, 39 Akron L. Rev. 151, 193–96 (2006); Kendall Svengalis, Legal Information Buyer's Guide And Reference Manual (“Svengalis”), 25 (2005).Google Scholar

11 Having done away with a print library, it is highly unlikely that a law firm, for example, will be able to justify the expense associated with the reacquisition of legal information in print form.Google Scholar

12 LexisNexis is a division of Reed Elsevier Inc.Google Scholar

13 They offer natural language research options as well, but the natural language search engines are, in essence, devices that translate natural language search queries into Boolean searches. Walter, supra n. 4, at 572, n. 19.Google Scholar

14 Casemaker is an example of an online legal research tool that is available only to bar members in states whose bar associations have joined the Casemaker consortium. For information on the Casemaker site, see http://www.casemaker.us/page.php?page=overview (accessed August 20, 2007).Google Scholar

15 Findlaw, a free online legal information site owned by West's corporate parent, Thomson, provides access to state and federal court opinions but the historical coverage is spotty and varies from jurisdiction to jurisdiction. In order to discover what coverage the site has for a specific jurisdiction, the user must check each library within the Findlaw database. For information on Findlaw's coverage. See http://lp.findlaw.com/ (accessed August 20, 2007). A fee-based alternative, Fastcase, has more coverage, but admits that its federal appellate coverage extends back only as far as “1 F.2d.” See https://www.fastcase.com/Corporate/Home.aspx (accessed September 7, 2007).Google Scholar

16 Loislaw, an online legal information database owned by Wolters Kluwer, claims to offer “up-to-date cases, statutes, rules and regulations, and other primary law for all 50 states and federal jurisdictions,” but charges an unspecified amount to enroll in the service. For further information on Loislaw, see http://www.loislaw.com/ (accessed August 20, 2007).Google Scholar

17 The present author has proposed that law schools band together to form a consortium to publish the law and make it accessible to all. Ian Gallacher, Cite Unseen: How Neutral Citation and America's Law Schools Can Cure Our Strange Devotion to Bibliographical Orthodoxy and the Constriction of Open and Equal Access to the Law, 70 Albany L. Rev. 491, 529–34 (2007). Some tentative steps are already being taken to provide free access to the law. See, e.g., Altlaw.org, a site established in collaboration between the Columbia Law School Program on Law and Technology and the Silicon Flatirons Program at the University of Colorado law school providing free access to a full-text searchable database of Supreme Court and Federal Appellate decisions from “the last decade or so.” Information found at http://altlaw.org/about (accessed August 20, 2007). And in August, 2007, the President and CEO of Public.Resource.Org, Inc. wrote to the President and CEO of Thomson North American Legal, announcing that his organization had “begun the process of scanning the Federal Reporter, the Federal Supplement, and the Federal Appendix” and that it intended to “extract[] the public domain content and republish[] it on the [i]nternet for use by anyone. Letter from Carl Malamud, President and CEO, Public.Resource.Org, Inc., to Peter Warwick, President and CEO, Thomson North American Legal (August 14, 2007) available at http://bulk.resource.org/courts.gov/ (accessed August 20, 2007). In a New York Times story about Public.Resource.Org, Inc.'s action, John Shaughnessy, a spokesman for Thomson, is quoted as having said: “We have received the letter from Public Resource and Mr. Malamud raises a number of interesting but complex points. We are looking at them now and then will be in touch directly with Mr. Malamud.” John Markoff, A Quest to Get More Court Rulings Online, and Free N. Y. Times, August 20, 2007 (“Times Article”), available at http://www.nytimes.com/2007/08/20/technology/20westlaw.html?ex=1345262400&en=9595a33c5fec0648&ei=5090&partner=rssuserland&emc=rss (accessed August 20, 2007).Google Scholar

18 Most recently, West fought Mathew Bender and Hyperlaw, a company attempting to publish and sell CD-ROM disks of case law. Bender and Hyperlaw won in both the United States District Court for the Southern District of New York and the Second Circuit. Matthew Bender & Co. v‥ West Publ'g Co., 158 F.3d 674 (2d Cir.1998) and Matthew Bender & Co. v. West Publ'g Co., 158 F.3d 693 (2d Cir.1998), cert. denied, 526 U.S. 1154 (1999). The victory was a pyrrhic one, however: Alan Sugarman, the President and CEO of Hyperlaw, was quoted as saying that the legal battle “cost me a lot of money, and when it was all said and done I was wiped out financially, so I went back to the practice of law.” Times Article, supra, n. 17. West also sought, and obtained, an injunction preventing Lexis’ then publisher from using West pagination in Lexis-produced versions of cases. West Publ'g Co. v. Mead Data Cent., Inc., 799 F.2d 1219 (8th Cir. 1986). West later blocked another publisher from using its page numbers (Oasis Publ'g Co v. West Publ'g Co., 924 F.SuPp. 918 (D.Minn. 1996)), despite an intervening Supreme Court decision that cast doubt on the validity of the Mead Data decision. Feist Publications, Inc. v. Rural Telephone Service Co., 449 U.S. 340 (1991). Most significantly, for the future of open access to the law, West intervened in a Freedom of Information Act (“FOIA”) case seeking to make the JURIS database available to the public. JURIS (“Justice Retrieval and Inquiry System”) was created in 1971 by the Department of Justice (“DOJ”), building on a collection of Supreme Court opinions collected by the United States Air Force under the acronym FLITE (“Finding Legal Information Through Electronics”). James H. Wyman, Freeing the Law: Case Reporter Copyright and the Universal Citation System, 24 Fla. St. U. L. Rev. 217, 254 (1996). The DOJ's purpose in developing JURIS was to have a collection of federal case law its attorneys could use. Id. When the burden of data entry and management became too much for the DOJ to handle internally, it contracted with West to provide those services. Id. West's contract, however, allowed it to remove all information entered by it under the terms of the contract if it chose to terminate its relationship with DOJ. Id. West indeed did terminate its relationship with the DOJ in 1993, and it exercised its right to remove all the case law added to JURIS during the ten year term of the contract. Id. Despite this, a non-profit group brought a FOIA request to have access to at least the data remaining in JURIS after West's removal of all West-entered material. Tax Analysts v. Dep't of Justice, 913 F.Supp. 599, 600 (D.D.C. 1996). West intervened in the action, claiming a “substantial interest” in the JURIS database, and helped to obtain a ruling that JURIS was not an “agency record” within the contemplation of FOIA. Id. at 601, 607; Wyman, supra, at 254.Google Scholar

19 The Altlaw.org site is full-text searchable, albeit somewhat inelegantly, (see http://altlaw.org/search/advanced) demonstrating that Boolean searching is achievable within an open access legal information site.Google Scholar

20 West has been digesting cases for over 100 years and has had the benefit of that time and labor of a substantial number of indexers.Google Scholar

21 The magnitude of the process is one of the reasons the author proposed American law schools – institutions with a large body of intelligent, skilled, and motivated students – as the appropriate location for an open access legal information effort. See, Gallacher, supra n. 17, at 531–33.Google Scholar

22 It should not be necessary to say that no indexing effort could use, in any way, West's ubiquitous, and most certainly copyrighted, key number system.Google Scholar

23 Footnotes 26–88 and accompanying text.Google Scholar

24 Footnotes 89–108 and accompanying text.Google Scholar

25 Footnotes 109–157 and accompanying text.Google Scholar

26 Robert Berring has argued that America's legal system is the most open and easily-accessed in the world. “[I]t is possible for any literate, English-speaking person to walk into a local library, perhaps one that specializes in law, but perhaps not, and find federal and state court cases, statutes, and administrative law.” Robert Berring, Essay, On Not Throwing Out the Baby: Planning the Future of Legal Information, 83 Cal. L. Rev. 615, 618 (1995). It is doubtful whether access to the law was ever as easy as Berring suggested; the patrons of many libraries, particularly in rural communities, likely had little or no access to legal materials and, at least as important, the finding tools necessary to locate relevant law and instruction on how to use them. And the situation has doubtless deteriorated since 1995, with escalating costs for purchase and storage of legal materials making it increasingly difficult for local libraries to maintain collections of legal information.Google Scholar

27 1 The Selected Writings of Sir Edward Coke 127 (Steve Sheppard ed., 2003)(1605). This belief is not universal. Even Coke, who had earlier written that lawyers should write documents in such a way that clients could understand them (“Note reader, there is great reason, that the writing should be expounded in such language, that the party may understand it, although he could read, because, by the law, he is at his peril to deliver it presently upon request, and hath not time to consult upon it with learned counsel.” Id., at 44) also wrote that statutes written in French should not be translated into English. “It was not thought fit nor convenient to publish either of those, or any of the Statutes enacted in those dayes in the vulgar tongue, lest the unlearned by bare reading without right understanding might sucke out errors, and trusting to their owne conceit might endamage themselves, and sometimes fall into destruction.” Id., at 76. Coke's view is held by some today as well. See, e.g. David Crump, Against Plain English: The Case for a Functional Approach to Legal Document Preparation, 33 Rutgers L.J. 713, 734 (2002)(arguing in favor of abstruse as opposed to “plain” English in some documents, in part because clients are more likely to take seriously legal documents that are written in difficult to understand language).Google Scholar

28 Wheaton v. Peters, 33 U.S. (8 Pet.) 591 (1834).Google Scholar

29 Banks v. Manchester, 128 U.S. 244 (1888).Google Scholar

30 Although the law has, in theory, been available for publication, the reality is that in the print world, West has been the only significant publisher. Although the Government Printing Office publishes the United States Reports, the official reporter for United States Supreme Court decisions (Svengalis, supra n. 10, at 69) the decisions of lower federal courts and twenty-nine state jurisdictions (Alabama, Alaska, Arizona, Colorado, Delaware, the District of Columbia, Hawaii, Illinois, Indiana, Iowa, Kentucky, Louisiana, Maine, Michigan, Minnesota, Mississippi, Missouri, New Jersey, New York, North Dakota, Oklahoma, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Utah, West Virginia, Wisconsin, and Wyoming) are only available in print form from West. Id. at 69, 471–586.Google Scholar

31 A “written opinion” has been defined by the federal Judicial Conference as meaning “any document issued by a judge or judges of the court, sitting in that capacity that sets forth a reasoned explanation for a court's decision,” although “[i]n the courts of appeals, only those documents designated as opinions of the court meet the definition of ‘written opinion.'” Stephen B. Burnbank, Judicial Accountability to the Past, Present and Future: Precedent, Politics, and Power, 28 U. Ark. Little Rock L. Rev. 19, 22–23 (2005)(internal quotation marks omitted) (citing Memorandum on Compliance with Website Requirements of the E-Government Act to All Chief Judges, United States Courts, from Leonidas Ralph Mecham 2 (Nov. 10, 2004)).Google Scholar

32 E-Government Act of 2002 § 205(a), Pub. L. No. 107–347, 116 Stat. 2913 (codified at 44 U.S.C. § 3501 (Supp. III 2005)). The E-Government Act provides that:Google Scholar

[t]he Chief Justice of the United States, the chief judge of each circuit and district and of the Court of Federal Claims, and the chief bankruptcy judge of each district shall cause to be established and maintained, for the court of which the judge is chief justice or judge, a website that contains the following information or links to websites with the following information:Google Scholar

(1) Location and contact information for the courthouse, including the telephone numbers and contact names for the clerk's office and justices’ or judge's chambers.Google Scholar

(2) Local rules and standing or general orders of the court.Google Scholar

(3) Individual rules, if in existence, of each justice or judge in that court.Google Scholar

(4) Access to docket information for each case.Google Scholar

(5) Access to the substance of all written opinions issued by the court, regardless of whether such opinions are to be published in the official court reporter, in a text searchable format.Google Scholar

Id.Google Scholar

33 It is also true that by no means all the population has access to the internet, thereby making the E-Government Act's promise of access to all court opinions chimerical at best. While many who do not themselves have access to the internet at home can use internet-enabled computers in public libraries, prisoners in jails typically have, at best, limited access to the internet, and therefore limited access to the opinions placed on the internet in response to the E-Government Act's provisions. And prisoners are active litigants who likely would make substantial use of legal resources were they made available to them. LexisNexis, for one, has identified a market opportunity in the prison world and has developed a service in which a CD-ROM of case law is loaded into a “kiosk” designed specially to withstand the rigors of a prison environment. See Amy Hale-Jeneke, The “Inside “ Information on New Jail Kiosks, LISP Newsletter, June 2004, available at http://www.aallnet.org/sis/lisp/news2004_1.pdf (accessed September 27, 2007).Google Scholar

34 The E-Government Act, while clearly requiring the publication of opinions in “text searchable format,” has failed to make federal case law entirely available to anyone with an internet connection. Nineteen federal courts were reported, in 2004, to be “deferring compliance with the Act's requirements as to the accessibility of written opinions ins some respect.” Burbank, supra n. 31, at 23 (citing Memorandum on Compliance with Website Requirements of the E-Government Act to All Chief Judges, United States Courts, from Leonidas Ralph Mecham 2 (Nov. 10, 2004)). And some courts who have complied with at least some part of the E-Government Act and made their opinions available on the internet have certainly not complied with the spirit of the Act. For a description of the problems associated with courts that use the Public Access to Court Electronic Records, or PACER, service as a means of disseminating their written opinions, see Gallacher, supra n. 17, at 516–19. But even had all federal courts fully complied with both the letter and spirit of the E-Government Act, the opinions would remain virtually inaccessible to researchers, and therefore useless for purposes of researching the law. The Act only requires courts to place their own opinions on the internet, thereby creating a group of stand-alone websites with no connections to each other. A search in one site, therefore, will only return the set of cases that correspond to the search for that individual site. In order to build up a comprehensive picture of federal court jurisprudence on a particular issue, therefore, the researcher would be compelled to conduct the same search in almost two hundred websites – an impractical task that no sane researcher would undertake. The promise of universal accessibility to federal court opinions apparently offered by the E-Government Act is, unfortunately, a political, and not a practical, solution: it is a chimerical illusion with no substance.Google Scholar

35 One new law student, confronted with a picture of a lawyer working against a background of reporters, is reported as having reacted “Oh my God, I'm going to have to read all those books or that lawyer … will have me for lunch, or maybe even just an appetizer.” Maureen Straub Kordesh, Navigating the Dark Morass: A First-Year Student's Guide to the Library, 19 Campbell L. Rev. 115, 115 (1996).Google Scholar

36 Alternatives to West's key number digesting approach existed during this period but they lacked the comprehensiveness or popularity of West's solution. Ultimately, the two principal alternative research systems – the American Law Reports annotations and the Lawyer's Cooperative Publishing Company's Total Client-Service Library System-came under West's control. Svengalis, supra n. 10, at 10, 75. West later sold the Lawyer's Cooperative U.S. Supreme Court Reports, Lawyer's Edition and companion Digest to Reed Elsevier, the owner of LexisNexis, as part of a consent decree entered into between West and the Department of Justice. Id. at 596.Google Scholar

37 Id., at 137.Google Scholar

38 WoltersKluwer bought Loislaw in 2000 for a price of $95 million. Id., at 14. Loislaw, formerly LOIS, was developed in 1987 to be a low-cost alternative to Westlaw and Lexis. Id., at 149. Initially, LOIS used CD-ROM as a means of disseminating and updating its libraries of state and federal opinions (although its failure to publish federal district court opinions limited its usefulness to practitioners), but moved to an internet-based distribution method in 1996. Id. Loislaw is part of the reason WoltersKluwer is the third largest provider of American legal information in terms of market share and is credited with forcing West to pay attention to the small law firm market. Id., at 15, 149.Google Scholar

39 In 2003, the last year for which data can readily be obtained, West controlled 39 percent of the legal information market, Reed Elsevier, the current owner of LexisNexis, controlled 26 percent, and Wolters Kluwer, the owner of the Loislaw legal information service and the owner of Commerce Clearing House, the Little, Brown publishing house, and Aspen Law & Business, controlled 17 percent. Id., at 15. The remaining 18 percent was spread among all other legal information publishers. Id.Google Scholar

40 Exactly what the costs of legal information are is an almost impossible question to answer. While Loislaw offers a flat fee for its services, both Westlaw and LexisNexis offer an array of charges, tailored to meet a vast array of needs. A law firm can choose between flat rate charges, calculated by analyzing the firm's past use of either LexisNexis or Westlaw, can elect to pay for database services on a transactional or time basis, or can negotiate individualized packages tailored to meet the firm's geographical and practice needs. For more information on Westlaw and LexisNexis pricing, see Gallacher, supra n. 10, at 196–97, Svengalis supra n. 10, at 140–48.Google Scholar

41 The time difference is limited to information retrieval. It is by no means clear that a research project using digests and case reporters would take, in the aggregate, any longer than a research project conducted using electronic resources.Google Scholar

42 There are already several commercial database alternatives to Westlaw and LexisNexis. Services such as Loislaw, Fastcase, Casemaker, and others all provide legal information to their subscribers. Adequate though they might be for many purposes, however, the restrictions on access or on coverage make these sites inadequate as open access and comprehensive sources of legal information. See, notes 14–16, supra and accompanying text.Google Scholar

43 Svengalis, supra n. 10, at 17. The reporter cost $262 in 1999 and $465 in 2003, an increase of $203. Id.Google Scholar

44 The price increases for print digests are so significant because these are the crucial finding tools necessary to make sense of the vast amount of information contained in the case reporters. A library could have all the available reporters on open display but without the digests to allow researchers to locate the issues analyzed in those opinions, they would have no meaningful function.Google Scholar

45 Svengalis, supra n. 10, at 17. The changes in price are even more disturbing when the full range of the change is taken into account. The Hawaii digest was originally priced at $819 in 1999 and fell $507 in that year to reach its final selling price of $312. Id. By 2001, the price of the digest had rebounded to $512, and in another two years the price was up to $1,371.50. Id. The same trajectory can be seen in the pricing for the Rhode Island digest, dropping from $1,089 to $432 in 1999, then rising to $903 in 2001 and then to $1,272.50 in 2003. Id. Svengalis attributes these increases to a conscious decision by West to “build its subscription lists prior to engineering an extensive program of bound volume revisions” and speaks of customers “lured” into buying products that West intended to increase dramatically in price during the coming four years. Id. Whether or not the reasons for these price increases were as nefarious as Svengalis believes, the reality for subscribers was a substantial increase in supplementation costs for an essential legal research tool.Google Scholar

46 For a discussion of the cost of storing print-based legal information materials in 2006, see Gallacher, supra n. 10, at 195.Google Scholar

47 A 2007 survey of Chicago-area law librarians suggests that we will see an “[e]ver-increasing reliance on electronic over print resources” in the next five years. Tom Gaylord, Chicago-Area Librarians Survey 5 (2007) (the “2007 Librarian's Survey”) (prepared for the 2007 “Back to the Future of Legal Research” Conference hosted by Chicago-Kent College of Law. A copy of the survey results is on file with the Author). The surveyed librarians did not see this trend as a good thing, noting that in the past five years they had seen “too much reliance on electronic [databases].” Id. A companion survey suggested that most senior attorneys agreed, with one respondent noting that the past five years had seen “[m]ore and more reliance on keyboard database searching, and less resort to books [which is] not necessarily a good thing because young associates often fail to develop concepts as a result of their myopic use of keywords. Sanford N. Greenburg, Chicago-Area Attorney Survey 1 (2007) (the “2007 Attorney Survey”) (prepared for the 2007 “Back to the Future of Legal Research” Conference hosted by Chicago-Kent College of Law. A copy of the survey results is on file with the Author). Another response echoed this comment, noting that “[t]he use of print materials seems to better stress and underscore the need for analysis. In contrast, online research is many time more mechanical… and many topics are missed.” Id. at 8. Asked for their suggestions for legal research training, most respondents (20 percent) answered that law schools should continue teaching print, as opposed to the 18 percent who suggested that the focus should be on “constructing better, more targeted searches.” Id. at 3.Google Scholar

48 The 2007 Librarian's survey indicated that 79.2 percent of those librarians responding to the survey had cancelled or planned to cancel subscriptions to at least some print materials. 2007 Librarian's Survey, supra n. 46, at 3. This move to computer-assisted legal research, and away from print, will likely have no effect on the price of the commercial legal information databases. Both LexisNexis and West likely will contend that they have to make up for the loss of revenue from their print products by maintaining the price of their electronic services, and both will have to carry the cost of continuing to retain a substantial number of employees in order to provide the added editorial services that make their versions of the law so valuable to practicing lawyers.Google Scholar

49 Svengalis has noted that “[s]ignificant opportunities exist for online providers who can effectively deliver primary law while undercutting the prices charged by the two major online legal services. As more courts, legislatures, and administrative agencies offer information on the World Wide Web, the role of the traditional primary law publishers will diminish in those instances where value-added content is not important.” Svengalis, supra n. 10, at 17.Google Scholar

50 West in particular is not shy about acquiring legal information providers when it sees an opportunity for enhancing its own market position and, presumably, eliminating a potential threat. The protection of its market position certainly seems to have been behind the purchase of Findlaw, a free legal information site, in 2001 for a rumored $37 million. News This Week Ticker, Nat'l L. J., Jan. 15, 2001, at A4. Even though Findlaw was “one of the most popular portals for accessing free legal information on the Internet,” (Svengalis, supra n. 10, at 14) such a purchase price seems to make little sense unless West perceived a potential threat behind Findlaw's marketing strategy. In addition to Findlaw, the Thomson Company has made significant purchases in the legal information field since its purchase of West in 1996, including Foundation Press, Federal Publications, the Harrison Company, and Andrews Publications. Id., at 15. In the same period, Reed Elsevier, the owner of the LexisNexis database, has purchased, among others, Matthew Bender & Company, Mealey's Publications, the Book Publishing Company, Courtlink, Anderson Publishing, and Gould Publishing. Id.Google Scholar

51 In addition to its purchasing activities, West has shown that it is not at all shy in pursuing litigation when it feels that is the appropriate step to take in order to protect its market position. See, supra, n. 18.Google Scholar

52 West's sponsorship of an award for federal judges has come under particular scrutiny. Sharon Schmickle & Tom Hamburger, Devitt Award is Prestigious – and Unusual; Close Involvement of Corporate Sponsor Sets it Apart, Star Tribune (Minneapolis), March 5, 1995, at 18A (describing West's involvement in Devitt Award for federal judges. Recipients of award receive $15,000 and crystal obelisk); Sharon Schmickle & Tom Hamburger, Members Accepted Gifts and Perks While Acting on Appeals Worth Millions to Minnesota Firm, Star Tribune (Minneapolis), March 5, 1995, at 16A (describing trips paid for by West and taken by award nominating committee, including several Supreme Court justices, while Court was considering cases in which West was party).Google Scholar

53 Thomson reported revenues of $3.647 billion in its legal and regulatory group in 2006, an 8 percent increase over the previous year. The Thomson Corporation 2006 Annual Report, available at http://ar.thomson.com/main.htm (accessed September 7, 2007).Google Scholar

54 This nightmarish possibility has been named the “Rupert Murdoch scenario” by Robert Berring. Robert C. Berring, The Evolution of Research: Legal Research and the World of Thinkable Thoughts, 2 J. App. Prac. & Proc. 305, 316 (2000).Google Scholar

55 Located at http://law.cornell.edu. The Legal Information Institute (“LII”) includes a series of recent and landmark Supreme Court opinions and acts as a portal to federal circuit court opinions, but its principal emphasis is on statutory and regulatory law. In this emphasis on federal statutes and regulations, the LII is reflecting the belief of one of its developers, Thomas Bruce, that “[t]he majority of people wanting legal information aren't doing formal legal research, but rather undertaking a kind of risk-management activity similar to what they might do with a site like WebMD. (And for the lawyers out there tempted to start griping about unauthorized-practice issues, the similarities to WebMD run deep: these folks almost never self-prescribe, and if they do, the effects are probably both drastic and Darwinian, so there's really nothing to fear). For [an] audience like that, judicial opinions are (most of the time) a secondary interpretative layer that surrounds statutes and regulations. What they want is legislation, regulations, and material that interprets those things.” Posting of Thomas R. Bruce to O'Reilly Radar, http://radar.oreilly.com/archives/2007/08/carl_malamud_ta.html (August 24, 2007, 05:20 EST).Google Scholar

56 See, supra, n. 17.Google Scholar

57 The LII is a leading provider of statutory information. While a Google search for “United States Code” and “US Code” places the LII third, behind the Government Printing Office site that publishes the Code on the internet, a search for “Federal Statutes” places the LII site first. Search performed September 18, 2007.Google Scholar

58 At the time of writing, Altlaw's coverage is limited to relatively recent federal appellate court decisions. The oldest decisions on the site are from the Supreme Court, from 1991. http://altlaw.org/vl/about/coverage (accessed September 19, 2007). Most circuit court decisions are from the mid 1990s. Id. The LII's coverage is mostly limited to statutory and regulatory materials. See, supra, n. 54. Coverage is, of course, something that can be added over time, but in order to be truly helpful, an open access site must have comprehensive coverage of all federal and state opinions from the earliest to the most recent.Google Scholar

59 Obvious though the use of an index might be to some, it is by no means as obvious to contemporary law students as we might think. Every spring semester, in both law schools where I have taught full-time, I assign an in-class research exercise to demonstrate the continued vitality of book-based legal research. In particular, my goal is to show the students that there is little practical difference in the amount of time it takes to take to perform a simple research task in a book and on the internet. The exercise consists of one student being asked to find a recipe for Welsh Rabbit (or Rarebit) in a cookbook I bring to class and another student being asked to find a recipe online. A third student is assigned to be the timekeeper, making a note of how long it takes each student to find the recipe. The student researching online usually finds the recipe first, usually within 30 seconds. The student using the book usually takes about twice as long, finding the recipe in about one minute. Then I ask the class who was faster, and they all take pleasure in telling me that the internet was faster – something they were already confident they knew and were confident I would be unhappy about. I tell them they are wrong and ask again. This usually results in a short but lively conversation about my sanity, and once that is over I ask the class how long each student took to find the recipe. The timekeeper confirms the times, and again I tell the students that they have the wrong answer and tell them that the assignment has been conducted in law time instead of chronological time. If I have a student who has worked in a law firm, that student often recognizes at this point that both students took 0.1 of an hour to complete the assignment, and that gets the class into a discussion of how, when working in terms of billable hours, there is often no difference in speed between researching on Westlaw or Lexis and the books. For our present purposes, though, the significance of this assignment is the reason the student using the cookbook takes so much longer than the student using the internet to find the recipe. In fact, when I first used this exercise, I anticipated that the student using the cookbook would beat the student using the internet, because the internet student would have to type “Welsh Rarebit,” which would take longer than merely looking up the words in the cookbook's index. But in six years of conducting this exercise, no student using the cookbook has ever gone directly to the index. Students have used a variety of means to access the information, including looking at the table of contents and flipping through the pages, but the index has been the second or third choice for finding information. One year I even had to suggest that the student use the index because the student was on the point of giving up. The student responded with a rolling of the eyes that I took, charitably to myself, to mean that the student was exasperated with his failure to remember such an obvious finding aid. Although anecdotal in nature, the consistency of the students’ failure to first use the index suggests that some profound changes are occurring in the way our students learn and think about information acquisition.Google Scholar

60 This list is called a thesaurus by indexers. Daniel P. Dabney, The Curse of Thamos: An Analysis of Full-Text Legal Document Retrieval 78 Law Libr. J. 5, 11 (1986). As Dabney notes, the thesaurus is a powerful document that itself can have a “substantive development on the subject of the collection.” Id. at 11–12, n.8. As illustration, Dabney uses the development of the topic heading “Bastards” to “Illegitimate Persons” to “Children-out-of-Wedlock.” Id. at 12, n.8. While these changes reflect a societal shift from antipathy to agnosticism towards children born of unmarried parents, Dabney's point that the choice of language to describe a legal concept can appear to be a subtle, or not-so-subtle, commentary on the subject of the concept, that could have an effect on the way a researcher thinks about the legal issue, is an important one and reflects an inherent defect in the pre-indexing approach to legal research.Google Scholar

61 For a more comprehensive overview of the West digesting process, see, Morris L. Cohen, Robert C. Berring, & Kent C. Olson, How to Find the Law, 83–110 (West Publ'n. 9th ed. 1989).Google Scholar

62 Dabney, supra n. 60, at 12.Google Scholar

63 Matryoshka dolls are Russian nesting dolls, in which progressively smaller dolls are enclosed within a large primary doll.Google Scholar

64 46 F.SuPp. 969 (S.D. Cal. 1942). This case was selected at random as the first opinion I found in my files when I began looking for a case to illustrate this point.Google Scholar

65 Id. at 972.Google Scholar

66 Id. at 969. In fact, the West indexer back in 1942 did not follow assign these designations and sub designations to the case, since the West thesaurus encoded legal issues differently when the case was originally decided. I have used the contemporary designations for the legal issue described here for simplicity's sake, but the reordering of designations from time to time is another complicating feature inherent in a pre-indexed research process.Google Scholar

67 Id.Google Scholar

68 Id.Google Scholar

69 Id.Google Scholar

70 Id.Google Scholar

71 Id. Again, these are the designations and sub designations currently in use for pretrial procedure issues, not those in effect in 1942.Google Scholar

72 Id. at 969–970. The case identifies a total of twenty legal issues, but there are thirty separate key numbered headnotes, indicating a number of issues that can be categorized in more than one way.Google Scholar

73 West has even anticipated the possibility that the relevant digest will not be available, as sometimes happens in law schools during the intensive legal research instruction most first year law students undergo. Decennial and General Digests gather up the digest entries from the other digests and publish them. Although these publications are so compendious as to be somewhat unwieldy when searching for a specific issue and jurisdiction, they can be invaluable when other, more targeted, digests are unavailable. For a more complete description of these digests, see, Cohen supra n. 61, at 94–99.Google Scholar

74 Dabney, supra n. 60, at 8.Google Scholar

75 Id. (“Manual indexing is only as good as the ability of the indexer to anticipate questions to which the indexed document might be found relevant.”)Google Scholar

76 Berring, supra n. 54 at 310.Google Scholar

77 Id., at 310–11.Google Scholar

78 “Members of the Ohio Bar who worked to develop [LexisNexis] defined what they wanted as a ‘nonindexed, full-text, online, interactive, computer-assisted legal research service.'” Jo McDermott, Another Analysis of Full-Text Legal Document Retrieval, 78 Law Libr. J. 337, 338 (1986)(quoting William G. Harrington, A Brief History of Computer-Assisted Legal Research, 77 Law Libr. J. 543 (1984–85)).Google Scholar

79 Legal Research and the Computer (1975), quoted in, Dabney, supra n. 60, at 14, n. 13. Dabney describes this as “early promotional material from LEXIS.”Google Scholar

80 “Boolean-logic searching, in effect, would allow each researcher to create an ad hoc index specific to the problem at hand. Harrington, supra n. 78, at 546.Google Scholar

81 “It is amusing today to recall the furor this proposition [of a nonindexed database] engendered when it was released for discussion. Self-anointed experts pronounced a nonindexed system a major error. Many law librarians were appalled to learn that the new concept of computer-assisted research would operate free of their dearly beloved, elaborate structures of digests and indexes. Some of them were intemperate in their scorn.” Id., at 546.Google Scholar

82 The LEXIS project started work in 1965 and was introduced to the legal community in 1973. For a discussion of the process that led to the creation of LEXIS, see Harrington, supra n. 78. Westlaw was unveiled by the West Company in 1975 as a competitor to LEXIS. Id. at 553.Google Scholar

83 This describes the traditional Boolean-type searching offered by both Lexis/Nexis and Westlaw. Both services now also offer a form of “natural language” search in which the researcher, in essence, types a question and asks the computer to return cases related to that question. Although this type of search offers a more pleasing front to researchers who might be uncomfortable with constructing Boolean searches, they are, in effect, Boolean searches constructed by the computer. Using an algorithmic process, these natural language programs remove irrelevant words from the search (“the” and “and,” for example), translate the remaining search terms into a Boolean search, calculates the frequency of the results in each document, and ranks the results in descending order of frequency. See, Walter, supra n. 4, at 569, n.l.Google Scholar

84 Research has shown that both the Lexis and Westlaw databases have numerous cases where terms are misspelled. See John Doyle, Misspellings in LEXIS and WESTLAW: A Statistical Test, 1 Trends L. Libr. Mgmt. & Tech. 5 (1989)(testing 350,000 cases in which a test word was used revealed 556 cases on Lexis in which word was misspelled and 276 cases on Westlaw). See also, Thomas Woxland, More on Misspellings in CALR Databases, 3 Trends L. Libr. Mgmt. & Tech. 1, 2 (1990)(suggesting that misspellings result in researcher missing up to 10 percent of relevant cases).Google Scholar

85 This is why it is important for legal researchers, particularly those who are researching an area of the law with which they are unfamiliar, to use secondary sources to gain an understanding of the concepts and vocabulary used by the courts to analyze an issue before they attempt to search in primary sources. See, e.g., Sloan, supra n. 6, at 29 (“Secondary sources can give you the necessary background to generate search terms [when researching an area of law with which you are unfamiliar.]”); Roy M. Mersky & Donald J. Dunn, Fundamentals of Legal Research, 16 (8th ed. 2002)(“To assist in formulating issues, it is useful to consult general secondary sources for an overview of relevant subject areas”).Google Scholar

86 “A computer search for terms such as ‘negligence’ or ‘equal protection’ will probably retrieve too many documents to be useful because the computer will retrieve every document in the database that contains those terms.” Sloan, supra n. 6, at 342.Google Scholar

87 As Dabney notes, the concept of “relevance” in the context of a legal search can be tricky to nail down and is largely dependent on the subjective determination of the researcher. Dabney, supra n. 60, at 15. For our present purposes, I will use the term to mean a document the researcher believes – rightly or wrongly – to be helpful in resolving the legal question at issue.Google Scholar

88 “The performance of a single research tool can be assessed according to its ability to select a greater or lesser percentage of relevant documents from the total number of relevant documents available ([the] recall)….” Scott F. Burson, A Reconstruction Of Thamos: Comments on the Evaluation of Legal Information Retrieval Systems, 79 Law Lbr. J. 133, 134 (1987).Google Scholar

89 Dabney, supra n. 60, at 16.Google Scholar

90 Jon Bing, Performance of Legal Text Retrieval Systems: The Curse of Boole, 79 Law Lbr. J. 187, 196 (1987).Google Scholar

91 Christopher G. Wren and Jill Robinson Wren, Using Computers in Legal Research: A Guide to Lexis and Westlaw 767 (1994).Google Scholar

92 See, e.g., Dabney, supra n. 60, at 26–31; Burson, supra n. 88, at 136–140; Bing, supra n. 90, at 196–197.Google Scholar

93 David C. Blair & M.E. Maron, An Evaluation of Retrieval Effectiveness for a Full-Text Document Retrieval System, 20 Comm. Ass'n. Computing Machinery 289 (1985).Google Scholar

94 Id. at 289.Google Scholar

95 Id. at 290.Google Scholar

96 Id. at 291. The researchers were, therefore, thoroughly familiar with the subject matter of the case and with the scope and content of the database, placing them on at an advantage over most legal researchers, who usually interrogate a legal information database with, at best, only a general understanding of the contents of the database that might be relevant to their inquiry.Google Scholar

97 Id.Google Scholar

98 Id.Google Scholar

99 The conclusions drawn from the Blair and Maron study by Dabney and others were hotly disputed by both Lexis (Jo McDermott, Another Analysis of Full-Text Legal Document Retrieval 78 Law Libr. J. 337 (1986)) and Westlaw (Craig E. Runde & William H. Linberg, The Curse of Thamos: A Response 78 Law Libr. J. 345 (1986)). Dabney then responded to the responses. Daniel Dabney, A Reply to West Publishing Company and Mead Data Central on The Curse of Thamos, 78 Law Libr. J. 349 (1986).Google Scholar

100 Molly Lien, Technocentrism and the Soul of the Common Law Lawyer, 48 Am. U. L. Rev. 85, 118 (1998).Google Scholar

101 Gene Koo, New Skills, New Learnings: Legal Education and the Promise of Technology, 6 (Berkman Center for Internet & Society at Harvard Law School 2007).Google Scholar

102 Id. Troubling though this finding is, its opposite – that newer lawyers favored precision over recall – would be equally problematic.Google Scholar

103 Burson, supra n. 88, at 137.Google Scholar

104 Anecdotal evidence certainly suggests that law students are (or are becoming) unwilling or unable to use print resources to aid them in their research. See, e.g., Theodor Potter, A New Twist on an Old Plot: Legal Research is a Strategy, Not a Format, 92 Law Libr. J. 287, 287 (2000)(quoting student, on being assigned to use a legal encyclopedia, as saying “I can't make this work to complete my assignment; I'm a computer person.”); Erica V. Wayne & J. Paul Lomio, Book Lovers Beware: A Survey of Online Research Habits of Stanford Law Students, 14–15 (Robert Crown Law Library Legal Research Paper Series, Research Paper No. 2)(2005)(when assigned to use “library resources” to find statute of limitations for fraud in California, one group of students went directly to computers housed in library and “Googled” their way to the answer.)Google Scholar

105 There can be no question that West's key number digesting system is proprietary and copyrighted.Google Scholar

106 Measured both in terms of developing the thesaurus, and therefore a new way of dividing the law into searchable segments, and also in terms of the salaries of the army of indexers necessary to analyze every opinion written by the federal and state courts.Google Scholar

107 Berring, supra n. 54 at 315.Google Scholar

108 Id.Google Scholar

109 Available at http://www.facebook.com/ (accessed October 2, 2007).Google Scholar

110 Available at http://www.myspace.com/ (accessed October2, 2007).Google Scholar

111 All diagrams in this article were created using Microsoft Viseo and are for intended only as a demonstration of the possibilities of this approach. They are limited by the author's profound lack of graphic skill and should in no way be taken as a definitive representation of what a final version of legal social network mapping might look like.Google Scholar

112 Network analysts refer to “nodes” as representing information within a network and “links” as the connections between nodes. See, e.g., James H. Fowler, et al, Network Analysis and the Law: Measuring the Importance of Supreme Court Precedents, available at http://ssrn.com/abstract=906827 (accessed September 4, 2007).Google Scholar

113 Legal research is an area blessed with many texts. Aspen Publishers lists eleven legal research textbooks, including workbooks and specialist texts to international students and paralegals seeking to conduct legal research (http://www.aspenpublishers.com/search.asp?Mode=SEARCH&keyword=research&ISBN=&Author=&Sort=DEFAULT&profitcenter=30); Foundation Press lists ten research textbooks (http://www.westacademic.com/Professors/ProductSearchResults.aspx?tab=2&publisher=5&subject=98&searchtypeasstring=ADVANCED-SEARCH); West (which also owns Foundation, but which publishes its own series of law school textbooks separately) lists seven research books and one set of DVDs in its catalog (http://www.westacademic.com/pdf/wls_2007_catalog.pdf); and Lexis lists five research books (http://bookstore.lexis.com/bookstore/catalog?action=prodlist&cat_id=T&pcat_id=89 (all accessed September 26, 2007)).Google Scholar

114 Legal research has been identified as a fundamental lawyering skill by the MacCrate Report (ABA Section on Legal Education and Admissions to the Bar, Legal Education and Professional Development: An Educational Continuum 157 (1992)), and all law schools in the country have some form of legal research education for their students.Google Scholar

115 It is a sign of the times that this is only true of people of a certain age. Telephone books are familiar to those of us who grew up in a pre-cell phone world, where telephones were devices tethered to walls and all telephone users were registered in telephone books. As cell phones become increasingly the standard method of distance communication, the notion of a book in which contact information is stored will likely become increasingly anachronistic and, perhaps more ominously, the techniques used for extracting this information will become increasingly unfamiliar.Google Scholar

116 There is no one national telephone book, of course, that is the equivalent of West's General, Decennial, or Century Digests, but the geographical division of the telephone directory, albeit in significantly narrower slices, mirrors the geographical divisions in the regional digests published by West.Google Scholar

117 The stacking in the telephone book is so obvious, and so familiar, as to be rendered almost invisible. My local telephone book, for example, contains the names of those who live within a discrete geographical area, who have purchased telephone service, and who have agreed to have their contact information published. All these are precoordinated elements that any user of the telephone book will assume without reflection. The information selected for inclusion in the telephone book is then organized in alphabetical order, last name first, and each identical last name is then organized alphabetically by initial letter of first name, each identical first letter is organized alphabetically by second initial, and so on.Google Scholar

118 At least, more quickly when calculated in chronological time. As we have seen, however, a more relevant measure of time from a lawyer's perspective is the “law time” unit of billable time. See, infra, n. 59.Google Scholar

119 Although the internet can return accurate and relevant information very quickly, it can also return accurate but irrelevant information just as quickly. A Google search for my name suggests that I could be a law professor at Syracuse University, (http://www.law.syr.edu/faculty/facultymember.asp?fac=136), a jeweler and diamond merchant in Stirling, Scotland, (http://www.iangallacher.com/), a computer game designer (http://www.mobygames.com/developer/sheet/view/developerId,33440/), a pastry chef (http://www.georgebrown.ca/Marketing/FTCal-Jan/chefschool/H411_Ian_Gallacher.aspx), or an actor in the movie “Knightriders” (http://www.showbizdata.com/credits/410797/Ian-Gallacher). http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4DMUS_enUS207US207&q=percent22ian+gallacherpercent22. And someone who entered a common misspelling of my name (believing it to be spelled “Gallagher” instead of “Gallacher”), might be surprised to learn that I am either a New Jersey-based performer of “Irish, Oldies, Standards, Rock&Roll, [and] Disco” who is “not your typical Irish singer” but is rather “a crooner with a clear jazz sensibility” (http://www.iangallagher.com/BAND.html), or a fictional character in the Channel 4 series “Shameless.” http://en.wikipedia.org/wiki/Ian_Gallagher. http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4DMUS_enUS207US207&q=percent22ian+gallagherpercent22 (all accessed September 26, 2007). Any researcher would be confused by the plethora of possibilities uncovered by these two quick searches and, without more information, would find it difficult to separate the mundane me from my other possible, and substantially more interesting, personas – my “Googlegangers,” in a contemporary, although perhaps transitory, coinage. For the record, I am two of the “Ian Gallacher” or “Ian Gallagher's described above.Google Scholar

120 Or, in these technological times, the individual's Blackberry, cell phone, or “Facebook” account.Google Scholar

121 It is more correct to say that all published opinions, or those existing in the netherworld between published and unpublished occupied by those opinions printed in West's “Federal Appendix” reporter are digested.Google Scholar

122 Thomas E. Baker, Intramural Reforms: How the U.S. Courts of Appeals Have Helped Themselves, 22 Fla. St. U. L. Rev. 913, 927 (1995).Google Scholar

123 See, e.g., Donna D. Adler, A Conversational Approach to Statutory Analysis: Say What You Mean & Mean What You Say, 66 Miss. L.J. 37, 54 (1996)(“Opinions are drafted not only to display the courts’ logic but also to set out the evidence of legal authority that led to the decision”).Google Scholar

124 See, e.g., Laurel Currie Oates & Anne Enquist, Just Research, 22 (2005)(identifying cases in chart of “finding tools” because of the “[r]eferences to cases in other cases …”).Google Scholar

125 Frank Shepard's “annotation pasters” – the forerunner to the present Shepard's citator – first appeared in 1873. Lynn Foster & Bruce Kennedy, The Evolution of Research: Technological Developments in Legal Research, 2 J. App. Prac. & Process 275, 277, n.7 (2000), citing Thomas A. Woxland & Patti J. Ogden, Landmarks in American Legal Publishing 43 (1990). Reed Elsevier, the owner of the LexisNexis service, bought Shepard's in 1996. Svengalis, supra n. 10, at 88.Google Scholar

126 West's KeyCite service is West's citator service, apparently created in response to Shepard's purchase by Reed Elsevier. Id.Google Scholar

127 A citator's function is to “catalog cases and secondary sources, analyzing what they say about the authorities they cite.” Sloan, supra n. 6, at 129.Google Scholar

128 It should go without saying that having found a case on which a lawyer might wish to rely, it is crucial for that lawyer to then check the development of that case using a citator. “You must check every case on which you rely to answer a legal question to make sure it is still good law. In general, you will want to use Shepard's or another citator early in your research, after you have identified what appear to be a few key cases, to make sure you do not build your analysis on authority that is no longer valid. Using a citator at this stage will also help direct you to other relevant authorities.” Sloan, supra n. 6, at 130.Google Scholar

129 Missing, of course, from this representation is the editorial content that makes a citator like Shepards or KeyCite so crucial for lawyers. Put in social networking terms, the diagram can indicate when and where A's friends were when A met them, but it cannot show whether A and C, for example, remain friends today or whether they had a falling-out some time after they met. Whether or not a case remains good law is vital information for lawyers seeking to cite a case as authority for a position. Yet most research experts will caution against over-reliance on the editorial features of citator services and will recommend that the prudent attorney read and evaluate subsequent authority before citing to the target case. See, e.g., Sloan, supra n. 6, at 142 (“Always research the Shepard's entry and review the citing sources carefully to satisfy yourself about the status of a case.”)Google Scholar

130 It would likely be necessary to research doctrinal issues rather than cases in their entirety. The sheer volume of information generated by an entire case would be so great as to swamp any useful information the researcher might derive from the search. Limiting searches to specific issues would minimize (although perhaps not completely eliminate) the danger of information overload.Google Scholar

131 “In general,… you will know that you have come full circle in your research when, after following a comprehensive research path through a variety of sources, the authorities you locate start to refer back to each other and the new sources you consult fail to reveal significant information.” Sloan, supra, n. 6, at 345–6.Google Scholar

132 The numbering of the various hypothetical opinions is not related to the chronological order of the opinions.Google Scholar

133 The researcher could do this by finding the most recent case in the relevant jurisdiction and reading the opinion. This is not the same as determining whether the case is still “good law,” something fledgling legal researchers are instructed to do. See, e.g., Sloan, supra, n. 6, at 130 (“You must check every case on which you rely to answer a legal question to make sure that it is still good law.”) This advice is, of course, both absolutely correct and slightly misleading: while it is important to know if the case itself is still “good law”, it is even more important for the researcher to know that the opinion's discussion of a particular doctrine is still valid. As Sloan notes, “a case with a negative Shepard's Signal… may no longer be good law for one of its points, but it may continue to be authoritative on other points.” Id., at 141–42. The network map of an opinion, therefore, can provide the essential information necessary to a researcher seeking to rely on that opinion, but will not provide the depth of analysis offered by a citator like Shepards or KeyCite. Were a network map such as the one described above to be offered as part of a legal information database, it is possible that some enhancements could be provided that would make the interpretation of subsequent citing cases somewhat simpler for the researcher. Loislaw, for example, offers “GlobalCite,” a listing of all documents indexed in the Loislaw databases that cite to the target opinion. http://www.loislaw.com/snp/fpopwind.htm (accessed September 27, 2007). Loislaw's promotional material suggests, somewhat optimistically perhaps, that, to a large extent, this service “replaces the annotations that often accompany printed editions of primary law.” Id. The search engine used by Loislaw to generate the “GlobalCite” list marks the text of the found documents in yellow whenever they refer to the target case and “renders ‘case treatment terms [such as ‘reverse,’ ‘remand,’ ‘affirm,’ etc.] in blue font.” Id. Similar enhancements could doubtless be added to a search engine that could generate a network map for court opinions.Google Scholar

134 46 F.SuPp. 969 (S.D. Cal. 1942). Supra, notes. 64–71 and accompanying text.Google Scholar

135 46 F.Supp. at 972.Google Scholar

136 36 F.Supp 168 (D.Pa. 1940).Google Scholar

137 109 F.2d 88 (10th Cir. 1940).Google Scholar

138 108 F.2d 509 (19th Cir. 1939).Google Scholar

139 Interestingly, the network relationship approach to legal research reveals a connection between these four cases that might not be immediately apparent to researchers using the more traditional pre-indexed approach. As noted supra. at n. 71, the Abrams court's decision on this issue was assigned two key numbers – 170Ak1835 and 307Ak687 – by West's digest editors. The same issue was assigned three different key numbers in the three cases relied on by the Abrams court to reach its decision: 302k214(1) in Butler; 302k214 (2) in Weeks; and 302k360 in Berger. While a careful and diligent legal researcher would doubtless have uncovered these additional key numbers, and would therefore have expanded the search to include them, it is salutary to note the degree to which the same or very similar issue can be assigned different key numbers even by careful editors. The network mapping approach would disclose the relationship between the cases, but a researcher using more traditional methods would have to be more assiduous in following up the various research options than experience suggests most legal researchers are.Google Scholar

140 For the sake of simplicity and space, I have limited the search to federal cases decided in the twentieth century. Without this restriction, the researcher would learn that the Weeks opinion cited to five state court opinions as well: Los Animas Consol. Canal Co. v. Hinderlieder, 100 Colo. 508, 68 P.2d 564 (1937); People v. Hadfield's Estate, 98 Colo. 206, 56 P.2d 25 (1936); Slusser v. First Nat'l Bank of Denver, 93 Colo. 219, 25 P.2d 183 (1933); Armstrong v. Johnson Storage & Moving Co., 84 Colo. 142, 268 P. 978 (1928); and International State Bank of Trinidad v. Trinidad Bean & Elevator Co., 79 Colo. 286, 245 P. 489 (1926).Google Scholar

141 81 F.2d 452 (10th Cir. 1936).Google Scholar

142 78 F.2d 881 (10th Cir. 1935).Google Scholar

143 In fact, the case cites to six Supreme Court opinions but one of them – Pullman-Palace Car Co. v. Missouri Pacific R. Co., 115 U.S. 587 (1885) – falls outside the restriction of federal cases decided in the twentieth century.Google Scholar

144 Although, as we have seen, the pre-indexed digest approach is limited by the decisions of the editor reviewing the case, and the assignment of a different Key number would mean that some of these cases would not be revealed in at least an initial search of this topic.Google Scholar

145 Another, though rarer, problem would arise where a court, by accident or on purpose, misstates the holding of a previous case in order to support its own decision. But even this could provide relevant information for a researcher, and would suggest that the target case is not the strongest on which to base an argument.Google Scholar

146 See, discussion supra at notes. 122–123 and accompanying text.Google Scholar

147 A conscientious researcher who researched all key numbers assigned in these cases to this topic would discover the following: a search for key number “170Ak1835” in Westlaw's “AllFeds” database on October 1, 2007 returned 3,682 cases; a search for “307Ak687” returned 6 cases; a search for “302k214(1) returned 59 cases; a search for 302k214(2) returned 40 cases; and a search for “302k360” returned 188 cases.Google Scholar

148 The examples of mapping used to illustrate this article have no such refinements, in part because they were created by an author with no skill in graphic design, and in part because they were designed to accompany an article that will be reproduced in black and white, not color.Google Scholar

149 Available at http://mapquest.com (accessed October 2, 2007). A search for “Syracuse New York” on this program resulted in a map locating Syracuse in the central region of New York state. http://www.mapquest.com/maps/map.adp?fo0&latitude=&longitude=&name=&phone=&level=city-syracuse&state=ny&zipcode= (accessed October 2, 2007). In the corner of the map is an indicator with the four cardinal compass points and a “plus” and “minus” indicator. By clicking on “north, “ “south,” “east,” or “west,” the user can reorient the map, always keeping the original target location in view. Id. By clicking on the “minus” button, the view expands incrementally by one of five levels: at first, the map broadens out to display substantially more of New York state; the next level shows a substantial portion of the east coast of the United States, with Washington D.C. in the south and Cleveland in the west and a large part of the eastern portion of Canada; the next level shows even more of both countries, with Savannah, Georgia in the south and Cedar Rapids, Iowa in the west; the next level shows Syracuse in relation to the entire United States and Canada; and the last level shows the location of Syracuse in relation to the North and South American continents, Europe and Africa. Id. Going up the scale from the initial setting provides increasingly detailed information until one is looking at a street map of the city. Id.Google Scholar

150 This is one area where a community-based approach to legal information might prove helpful. While some automatic features, such as detecting key words like “overruled,” “superseded,” “reversed,” or “vacated” in subsequent opinions, might provide some hints to the researcher about a case's viability, more sophisticated analysis would be reliant on human analysis, something that would be cost-prohibitive were editors to be hired especially for that purpose. But were it possible for a researcher who reads a case and concludes that it overrules, or affirms, or in some other way affects, a previous case, to annotate that result, such information could be used to build up a more complete citator-like service over time. Such an approach could either work automatically, as, for instance, does the reCAPTCHA method of deciphering text (see, http://news.bbc.co.uk/2/hi/technology/7023627.stm (accessed October 2, 2007)(image of indistinct word sent to two different people and results compared in order to determine accuracy of response)), or through some form of mediated Wikipedia, in which members of the community could submit case analysis but an editor would vet the submissions before attaching them to cases on the website. Such an approach would inevitably leave gaps in treatment, but would allow the analysis of at least some of the cases on the site to become deeper and richer over time.Google Scholar

151 See, Sloan, supra. n. 6, at 89–90. Sloan also notes, however, that “[t]he easiest way to find relevant topics and key numbers is to use the headnotes in a case that you have already determined is relevant to your research” Id. at 89.Google Scholar

152 See, e.g., Oates & Enquist, supra. n. 124, at 188 (“Step 1: Spend 30 to 60 minutes doing background reading in a state practice manual or book, a hornbook, or Nutshell, or a legal encyclopedia.”)Google Scholar

153 Id. (“Step 3: Cite check the cases that you plan to use to make sure that they are still good law.”)Google Scholar

154 In fact, of course, the network map provides the researcher with more information, because it explores the development of an issue retroactively as well as prospectively.Google Scholar

155 See, Sloan infra n. 6, at 340 (combination of print and electronic resources necessary “for accurate and efficient research.”)Google Scholar

156 This approach would also allow the researcher to refocus the research on another case when that appeared to be a more profitable approach.Google Scholar

157 “[H]eadnotes are … merely finding aids and should not be cited or relied upon as authoritative.” Cohen et al, supra, n. 61 at 25. As the Eastern District of New York has noted, headnotes and annotations in legal research are “where research starts, not where it ends.” Amorgianos v. Nat'l R.R. Passenger Corp., 137 F.Supp. 2d 147, 189 (E.D.N.Y. 2001). Over-reliance on headnotes has caused attorneys to get into trouble with courts. See, e.g., Pileri Indus., Inc. v. Consolidated Indus., Inc., 740 So. 2d 1108, 1109–10 (Ala. Civ.App. 1999)(“[T]here are deficiencies in Pileri's brief. Throughout, Pileri's brief cited headnotes from West's Southern Reporter. A headnote is not legal authority; rather, it is a publisher's interpretation of what the particular court stated, and it should not be relied upon to convey the precise case holding. A headnote is intended solely for the convenience of the public and the Bar as a research and indexing aid.”); Barber v. Motor Vessel “Blue Cat,” 372 F.2d 626, 629 n.7 (5th Cir. 1967)(quotation of a headnote as though it were the words of a court is “abomination of appellate advocacy”).Google Scholar