United Kingdom Journals

BAILII, Legal Education and Open Access to Law

Philip Leith [1], Cynthia Fellows [2]

Cite as: Leith, P., Fellows, C., 'BAILII, Legal Education and Open Access to Law', European Journal of Law and Technology, Vol. 4, No. 1, 2013

Abstract

The British and Irish Legal Information Institute (BAILII) entered the online legal information landscape in 2001 with charitable status as a provider of UK and European judgments, and has over the past decade or so moved from a system quickly put together with any materials which could be found, to a system which provides a core resource to professionals in law. [3] In this article we provide an overview for the law teacher of the system's first years and we then look at whether usage in law schools has matched that of the professional, how the JISC funded Open Law project enabled development for law students, and where we might go in the future as part of the Legal Information Institute collective which operates under the 'Free Access to Law' banner. [4]

As members of the Open Law team who sought funding, carried out the research and implemented the project, it seems to us that the project was generally successful. Our indications were that prior to Open Law the use of BAILII by students was low - it was not readily found or discussed by lecturers, was difficult to use, and generally less user friendly than it could have been. The changes implemented by Open Law appear to have changed that position considerably. However, our findings also indicate that there is much work to do to re-energise digital legal information as a legal education research field.

1. Where we were before BAILII

In simple terms - and primarily for readers outwith the common law jurisdiction - a law report is a document which provides an analysis of a court case, most often accompanied by text from a judge's decision. This law report will have been produced by a lawyer, but one who did not participate in the case itself. It is, in effect, a third party view but a view which arises from the legal perspective. Thus the general reporting of a case in a newspaper is reportage, but lacks the critical legal skills and qualities which comprise a law report. Law reports have been produced in England since the 13th century. Legal historians have suggested various reasons for their existence - precedent, educational, record keeping and entertainment value. [5] These reasons continue to be relevant, but the formal reason is that of precedent. Precedent in a common law system can be generally defined as the legal principles identified and developed through judicial decision-making - i.e. what has occurred in a prior case may have value in deciding a current case, and thus access to the reasoning and decision in prior cases is important (stare decisis). Current rules about precedent in the Court of Appeal were laid down in Young v. Bristol Aeroplane Co [1944] KB 718. In the UK the courts not only interpret and clarify statutory law, they also create law in areas not covered by statute, e.g. tort, civil liberties, etc. and take precedent into consideration as a base for this law creation process. Thus, access to the law reports is essential not only to lawyers and judges, but also to the teaching and study of the common law because in common law countries the real 'meaning of law' lies in the case law which interprets legislation.

It is important to remember the difference between a law report and a judgment. The latter is the product of the judge and the former combines (some of) the words of the judge with an analysis of what he or she said. Law reports can be a few hundred words or several thousand. This distinction means that law reports are the copyright of the law reporter (because they contain original work) whilst the judgment is the copyright of the judge (or, outwith the UK, his - or more rarely, her - employer). [6] [7] Historically a symbiotic relationship existed between the law reporter and the UK judge. The judge would produce a decision (often read out in court in full) and this might be in typed form. If it was, a copy would be stored at the court, and copies could be made available. For oral judgments, transcripts produced by reporters would be made available to the judge for correction. Since producing copies prior to the advent of word processing was expensive, there was little external call for copies to be made apart from by the law reporter who would produce a version which was suitable for legal public consumption. The law reporter effectively chose which judgments would be made public. The costs of dissemination of case law were thus met through the private enterprise of the law reporter and - amongst other agencies and publishers - the Incorporated Council of Law Reporting (ICLR), which publishes law reports of the decisions of the Superior and Appellate Courts in England and Wales. [8]

To the law school in the 1980s and 1990s, access to case law via any means apart from bound law reports on a library shelf would have been pointless: the technology needed to give widespread access to the online databases of law reports and judgments which were being produced by firms such as Lexis and Westlaw was not easily obtainable. Most law schools until the early 1990s had only one dedicated terminal which could access these databases. [9] The BILETA (British and Irish Law Education and Technology Association) reports from 1991, 1996, 2001 and 2003 on access to technology in law schools paint a dismal picture of technology improving but slightly. [10] It was relatively common, in such an information poor environment, for students to tear a headnote from a bound law report - an indication of the moral compass of some law students and the utility of a headnote as a synopsis of a case. Bound reports in university law libraries would routinely have photocopied replacement pages added.

From the 1990s, when word processing reached the judiciary, the situation in the courts changed. Copies of the original judgments could be accessed more easily by parties other than the law reporters. The founders of Cornell University's Legal Information Institute (Cornell LII) and the Australasia Legal Information Institute (AustLII) thus saw that law could be made more accessible, and - importantly - done so through a free-access mode. BAILII, a child of AustLII, was thus able to access these directly from the courts, but - since the law reports were the copyright of their producers - this meant that a very large quantity of cases which are central to UK law could not be accessed. [11] [12] The publishers have always viewed these law reports as valuable intellectual property and have been generally unwilling to make them available except on infrequent occasions. One such occasion was for the IOLIS project, where a computer assisted learning package containing a substantial body - some 2,000 - of undergraduate useful law reports was created. [13]

Digital judgments produced from word processing improved access to the law very substantially on a worldwide basis, though it is the common law countries where the full reasoning appears in the higher judgments. [14] The developments in the US at the Cornell LII [15] and in Australia under AustLII initiated a movement for access to the original judgments from the courts and to make this access freely available. With WorldLII providing a framework, there are now 123 jurisdictions which participate in this access movement. [16] Researchers can execute a single search across all databases covering these jurisdictions and with such open access to case law, the citizen can - without the need for an intermediary - glean the thinking of the judiciary and see how law is being interpreted, as well as picking up on the moral framework of the judge (given that 'who behaved well' is an important element in advocacy). Without it, they have only the legislation to consider and - as every legal educator knows - legislation can be interpreted in a variety of ways. [17]

There has been litigation in the US over access to judgments. The US Hyperlaw cases effectively removed the copyright protection in US Federal judgments being claimed by private publishers. [18] This meant that rivals could utilize judgments held by others. [19] In Europe, the Database Directive does not allow extraction of other's data and also the ECJ decision in Infopaq means that processing of printed reports to extract the judge's words and remove the copyright elements is unlawful. [20] [21] This means that only printed law reports prior to the 1890s are accessible to BAILII as sources of judgments. [22]

However, from 2000, BAILII has been receiving an increasing number of decisions from courts and tribunals - mainly as links have been built up or courts and tribunals have made efforts towards more open access; consequently the breadth and depth of content on BAILII has expanded significantly. It now hosts the most comprehensive set of British and Irish primary legal materials available for free and in one place on the internet. BAILII offers open access to not only British and Irish case law and legislation, but also European Union case law, Law Commission reports, and other law-related British and Irish material. Some of the documents on BAILII are freely available elsewhere on various government, court and tribunal websites. Uniquely however, BAILII has a large number of Privy Council judgments; Law Commission Reports; England and Wales Courts of Appeal and High Court judgments; and older tribunal decisions and court decisions from Northern Ireland, Scotland and the Republic of Ireland that are either not at all available on the internet or are not available for free. BAILII derives its data from a number of sources, relying heavily on direct and indirect feeds by relevant courts, government departments and other organization. Some of the data comes from existing free to air sites and some is based on published and unpublished CD-ROMs. [23] All of the data has been converted into a consistent format to which a generalised set of search and hypertext facilities have been added. [24]

Not everything useful is available: one example is whilst Northern Ireland first instance employment tribunal decisions are available, those from England and Wales are not. Requests to various courts and tribunals to supply judgments are sometimes successful and sometimes not: it sometimes appears that some individuals who control provision may be averse to providing BAILII with access. The situation at present is that for most courts and tribunals, it is not possible for BAILII to demand a stream of judgments. [25] Further, it should be noted that BAILII does not edit the materials it receives in any way. BAILII does not redact names from judgments, though some in the LII movement believe that this is a probable next step and something which should be planned for. [26] There is frequently a desire from parties - particularly in commercially sensitive disputes - to have judgments withheld or redacted. The decision on whether to do this is in the hands of the judge - sometimes the request is overturned and sometimes agreed to, even though the judge has reservations about this. [27] [28] Redaction of names of children etc. is common practice. [29]

2. Publication of Law Reports and Citation Practices

When printed law reports were the only format for judicial decisions there were limits as to how much could be reasonably published. But with digital formatting and electronic access, there is no physical necessity to restrict publication. Nevertheless, finding a balance between publishing too much and publishing not enough is a challenge in all common law countries. [30] Members of the legal profession in both the US and the UK have complained about having to deal with 'too much law' since the early seventeenth century. [31] The tension arises because of the doctrine of binding precedent in the common law: unreported judgments, opinions assigned no precedential value, and rules prohibiting the citation of either or both are the subject of much contemporary debate. In both England [32] and the US efficiency arguments in favour of no-citation rules and policy arguments opposed to no-citation rules have been advanced by lawyers and judges. Although each country has taken drastically different approaches to publication practices and no-citation rules, they share common problems and are perhaps moving in a common direction. [33]

In the UK, the perceived problem of access to a large number of 'unreported' judgments which have not been sifted and sorted by law reporters - the efficiency argument - is one which has been raised by the judiciary, notably by Lord Justice Diplock in 1983 and Laddie J. in 2001. [34] Various rules were issued in 1996 and 2001 limiting the rights of lawyers to cite unreported judgments and giving judges the power to prospectively declare the precedential value of their judgments. [35] [36] [37] These judicial opinions and the 2001 Practice Directions are explored at length in commentary by Munday in 2002. [38]

But there is a policy argument that a fundamental rule of law is that those governed by the law have the right to know what the law is. [39] Concerns about unreported judgments creating inequality of access to the law were substantiated in a 1992 study which concluded that English reporting practices gave affluent and repeat litigants an advantage because they were more likely to be aware of unreported judgments. [40] The judiciary is not insensitive to the need for a level playing field that provides 'access to all, free of charge, to our caselaw and statutes and other publicly available legal materials'. [41] Since access to the law is vital for litigants in person (LiPs) and to those assisting individuals seeking general advice:

'[Free access] would make a huge difference, for instance, to those who provide advice on housing or social security benefits, asylum, tax or employment law, if they could be sure they could find the latest judgments from the Administrative Court or the leading appeal tribunals on the BAILII site.' [42]

And there are advantages to the judges as well:

'[T]hose of us who are interpreting the law in the higher courts are being supplied with up to date copies of unreported judgments relevant to the points we are deciding. This has been particularly valuable in the early days of the Human Rights Act, where we have so much to learn from each other. The law would be in chaos if our early decisions were being made in ignorance of what another court had been saying on the same point. Another great blessing of modern technology is that when we post an important judgment on a website at the time we are handing it down in court, we know that it will then be communicated electronically throughout the country the same day to everyone who has a 'need to know', and this does not only mean lawyers.' [43]

In 1964, the Judicial Conference of the United States resolved that the federal appellate and district courts should only authorize the publication of precedential opinions [44] and a 1971 report of the Federal Judicial Center also recommended limited publication practices and a no-citation rule. [45] But the 'no citation' rules were challenged [46] and the issue eventually was placed on the agenda of the Advisory Committee on the Federal Rules of Appellate Procedure. US Supreme Court Justice Alito chaired the committee which conducted a lengthy study of the diverse circuit court rules governing the withholding of vast numbers of opinions from publication and limiting citation to such opinions. The committee wrote:

The disparity between litigants who are wealthy and those who are not is an unfortunate reality. Undoubtedly, some litigants have better access to unpublished opinions, just as some litigants have better access to published opinions, statutes, law review articles - or, for that matter, lawyers. The solution to these disparities is not to forbid all parties from citing unpublished opinions. After all, parties are not forbidden from citing published opinions, statutes, or law review articles - or from retaining lawyers. ... [T]he solution is found in measures such as the E-Government Act, which makes unpublished opinions widely available at little or no cost.' [47]

In the face of 'judicial resistance that raged from mild to fierce', [48] Federal Rule of Appellate Procedure, Rule 32.1 'Citing Judicial Dispositions' was subsequently adopted and took effect on December 1, 2006 . Part (a) of the Rule states:

(a) Citation Permitted. A court may not prohibit or restrict the citation of federal judicial opinions, orders, judgments, or other written dispositions that have been: (i) designated as 'unpublished,' 'not for publication,' 'non-precedential,' 'not precedent,' or the like; and (ii) issued on or after January 1, 2007.

This rule only addresses the citation of unpublished opinions issued after the effective date of the Rule and leaves a number of issues to the individual federal appellate courts, including whether to issue unpublished opinions and what precedential value to give unpublished opinions.

In the UK there is ongoing revision to the rules regarding citation of authorities. Practice Direction: Citation of Authorities (2012) was issued in March 2012 'in order to clarify the practice and procedure governing the citation of authorities'. This Practice Direction repeals several earlier Practice Directions and selected paragraphs of earlier Practice Directions. Two observations are relevant to this article's discussions. First, per paragraph 10.1.5 of Practice Direction [1999] 1 WLR 1027, it had been the practice of the Senior Courts in the UK to require permission to cite an 'unreported' case and usually not to grant permission 'unless advocates are able to assure the court that the transcript in question contains a relevant statement of legal principle not found in reported authority and that the authority is not cited because of the phraseology used or as an illustration of the application of an established legal principle'. This paragraph has been revoked by the 2012 Practice Direction. Second, with regard to 'unreported' judgments, the 2012 Practice Direction states in paragraph 10:

'Where a judgment has not been reported, reference may be made to the official transcript if that is available, not the handed-down text of the judgment, as this may have been subject to late revision after the text was handed down. Official transcripts may be obtained from, for instance, BAILLI ( http://www.BAILII.org/). An unreported case should not usually be cited unless it contains a relevant statement of legal principle not found in reported authority.' [49]

As the Incorporated Council of Law Reporting (ICLR) comments:

'It [Practice Direction: Citation of Authorities (2012)] recognises . . . the crucial role played by the British and Irish Legal Information Institute (BAILII) in collecting and making available the official transcripts of judgments delivered not only by the Senior Courts but a range of other courts and tribunals. Though these unreported judgments do not have the status of law reports, they provide an often essential and easily accessible resource for the study and practice of law, and as a charitable organisation ICLR is proud to support and co-operate with BAILII.' [50]

BAILII now has substantial support from its user community, uploaded 24,706 judgments in 2012, and is - as we show below - utilized by a large number of law students. It continues to subsist on a very small income through charitable contributions and while it employs a very small staff, it has had a major impact upon access to law over the past decade. That BAILII's impact has been substantial is well understood. Lord Neuberger, President of The Supreme Court said in the first annual BAILII lecture in 2012:

'The extent and the speed of the revolution represented by the BAILII website is hard to articulate. That is partly because, in such a short time, it has become an indispensable and comprehensive source of information, both generally as to what is going on in various courts, and specifically if one is trying to find relevant court decisions. ... As a quick, user-friendly, and reliable way of finding a particular case, cases on a particular topic, cases dealing involving a particular judge, or any more esoteric search, it is remarkably well organised, comprehensive, and practical.' [51]

What is surprising is that the open access model in legal information still appears to have opponents within the group who control access to judgments, and consequently BAILII is never completely successful in getting the judgments it requires to satisfy the needs of its users. For example, in a recent attempt to secure eight EWCA 2012 judgments that a government body wanted posted on BAILII so that they could be included in their training coursework materials, only one was obtained despite the fact that six of the remaining seven were obtained by Westlaw and Lexis (the commercial legal information providers). It appears that decisions about access are being made without consideration of the Public Sector Information (PSI) legislative framework, so those who control judgments are choosing to whom they wish to make these available, rather than providing equal access to all. This situation must surely change when the PSI regime becomes more properly implemented. [52]

It is not clear how BAILII has fully impacted upon the commercial sector. The SLS/BIALL Academic Law Library Survey 2010/2011 indicates that Westlaw and Lexis continue to dominate the market and to increase their pricing roughly 5% annually. That said, there are doubtless many educational institutions who do not need and cannot afford the comprehensive access to primary legal sources offered at high cost from commercial vendors; these institutions undoubtedly find in BAILII a useful and valuable resource for law. It should be noted as well that the view from within BAILII in its first three years was that commercial players found they had to improve their speed in uploading of judgments. [53] Other impacts include BAILII's first chairman, Henry Brooke, being responsible for introducing a neutral citation system to UK law reporting. [54] For legal educators, a major advantage of BAILII against all others should be that one is able to link directly to cases and paragraphs within cases: students can see a seamless interface between in-house teaching documents and materials held on BAILII without having to log in to a service provider. [55]

As to location, BAILII is hosted in the UK and Ireland by the Institute of Advanced Legal Studies (IALS), London and the Law Faculty, University College Cork. BAILII is legally constituted in the UK as a company limited by guarantee and as a charitable trust, supported by a number of major sponsors and assisted by many other organisations and individuals. [56] [57] BAILII celebrated its twelfth anniversary in November 2012 and now comprises 90 databases containing over 310,548 searchable documents making up 38 gigabytes of data. Three servers offer parallel data access to provide maximum up-time for the system, two being based at the IALS and the other at Queen's University of Belfast, all accessed via the UK academic network. Usage of BAILII is currently averaging over 58,616 users per week generating over 1,781,460 page requests per week.

The small staff means that BAILII must be technically efficient since carrying out large scale integration of materials into the database without automation would render the process too expensive to be feasible. Currently, in many cases BAILII 'grabs' material (often in batches) from various internet databases many of which have a table of judgments/decisions details and a link to the word or PDF file for each judgment/decision. BAILII must:

1. determine if they already have the data (the tables are ot always in order of when the judgment/decision was added/amended)
2. construct the BAILII file name (from the table details often not in the judgment/decision document itself)
3. retrieve or construct a citation (from the table details often not in the judgment/decision document itself)
4. create the <title> </title> for the PDF or word file that is being converted (from the table details often not in the judgment/decision document itself)

This is done through various programming tools. For example, a one pass program which:

1. renames the PDF or word to the desired BAILII file name
2. using COM object inserts the: <title> case name, (subject heading), citation (promulgation date) </title> into the judgment/decision document's title property then converts the judgment/decision document - leaving one with the html and rtf version of the judgment/decision both properly named and the html version with the correct <title> </title> information for the BAILII system.

Even with all the information in the judgment/decision word document, given the numerous types of formatting generated during MS Word document conversion to html it can take complex programs (based on regular expression programming techniques) to retrieve the same data from each of a batch of converted word documents consistently. Documents are added as they are made available - thus older judgments and decisions from prior years are consistently extending the database.

In terms of organisation, BAILII differs slightly from the other LIIs, since although it is closely connected with the academic community (with several academics serving as Trustees and servers housed in two University buildings) it has perhaps a closer linkage with the judiciary than those in other countries. Its chairmen have both been senior judges, and other senior judges have served as Trustees. Given the nature of the UK legal system and the general lack of technical expertise (or funding) within the Ministry of Justice and Courts Services, it would have been significantly more difficult to build BAILII if the senior judiciary had not been so involved and so supportive. More recently there has been an increase in practitioners serving as Trustees. There are currently no Trustees from a government background. Trustees are unpaid and do not make calls upon BAILII's income.

3. Researching Information Usage in Law

Those with an interest in exploring how technology has impacted the way legal information is being used over the past few decades have had very little information upon which to base their research. [58] Since most law was locked up through contractual agreements between courts, reporters, and legal publishers it was difficult to ascertain who was using the electronic databases which were commercially available, how often, and for what purpose. Such information was rarely made available given its commercial sensitivity to the legal publishers. The inception of Legal Information Institutes (LIIs) has hopefully changed this by manifesting a more open attitude to potential researchers of legal information use. For the LIIs anyway, usually funded as charities, it is an imperative to prove to those who provide funding that the community actively uses the service. Without that information, then funding will - and probably should - cease.

However, some of the tools which are available and commonly used by commercial entities wishing to understand browsing practices from users of their website can be relatively intrusive. Systems such as Google Analytics are powerful, yet demanding of information: they leave cookies, analyse search patterns, and utilise various analytical techniques which many object to, particularly in a context where legal confidentiality may be very important. [59] Since BAILII has consistently promised its users that it will not use any privacy undermining techniques and thus does not carry out this sort of background analysis, a user survey is the best means (also in terms of cost and effectiveness) to gauge information about BAILII's users, even though it is accepted that responses to online surveys are usually relatively low. Given the limitations of online surveys though, when BAILII posted an invitation on its website in early 2012 encouraging users to complete a survey, 3,274 users responded - a relatively high figure for this kind of research - around 6% of the weekly user figures.

4. The Open Law Project

BAILII had made a successful bid for funding in 2004 from JISC to carry out two tasks - development of the database and development of improvements to the interface. [60] At that point, the materials available on BAILII were substantial in quantity: 400,000 searchable documents with several million internal hypertext links but these were mostly related to the 1990s and early 2000s. [61] Despite the paucity of undergraduate useful cases on our database, we still had a large contingent of educational users as a survey of users during our first three years demonstrated with one third of respondents to that survey indicating they were involved in education. Of that group:

• 74% identified themselves as undergraduate, 12% postgraduate, 6% research, 4% CPE or LPC or BVC students and 4% as from secondary education.
• The frequency of use by students in our target area is shown by the 42% who use BAILII at least once a month, 29% once a week, 23% a few times a year, and 6% who use BAILII daily.
• Some 31% of students from our target area valued BAILII as very important, 28% said it is important, 33% useful and 8% not important.

BAILII had been able to gather up cases as they became available after 2000, but also was given access to materials from the Court of Appeal from the official transcribers. Some judges provided materials directly to BAILII as they were produced, but generally there were a very limited number of pre-2000 judgments from the High Court and other courts. This meant that BAILII was of more use to postgraduate students carrying out research on current legal developments, and of limited use to undergraduate students who wanted access to the case law which was discussed in their legal texts.

The Open Law project attempted to overcome that lack of utility to undergraduate students by building a list of 'important cases' and then trying to get as many of these from as diverse sources as required. The lists were compiled by making contact with the academic community (e.g. academics, librarians, and special interest groups such as the Society of Legal Scholars) and ordering the importance of judgments so that we tried to get those which were the most useful. The problem of 'ownership' of judgments arose but, in general, we were able to carry out the task which we set ourselves. [62] The Open Law project identified leading cases from the past for 16 selected subjects. [63] BAILII then sought permission (and sometimes the cases themselves) from various libraries and publishers to digitize the judgments and enabled access to over 2,380 leading cases searchable by subject. BAILII does have some small resources to add more cases of note to the leading cases lists, but those materials that have been identified as desirable additions often remain problematic in terms of copyright ownership.

The project assumed that judgments prior to around 1890 were in the public domain and viewed these as suitable for processing and digitising first. BAILII then sought various sources of copyright-free materials for the remainder. We successfully scanned and converted materials in the Supreme Court Library, judgments directly from one cooperative transcriber, from the House of Lords and Privy Council. BAILII negotiated permission from the ICLR and SCLR and from the Estates Gazette law reports for permission to publish these limited numbers of pre-2000 'leading cases'. The Bar Library in Belfast provided copies of Northern Ireland cases.

It should be noted that one of our aims in the Open Law project was to try to move students away from simply reading textbooks and regurgitating lists of cases: our primary aim to overcome the long standing problem of students rarely reading case materials. The UK does not have a history of basing law teaching upon the Socratic method (as used by most US Law Schools [64]) and there is a tendency for students to be concerned more with learning the name of a case and the important point raised (as described in the student texts) rather than reading the case itself. We were not intent upon bringing the Socratic method to the UK for a number of reasons, but partly because it is viewed by many as hidebound and formalised and also because some studies have found that women students can be intimidated by the Socratic method. [65] [66] However, we do feel that law students who had rarely read a full case during their law school education were not being fully educated.

The second part of the Open Law project was about improving the interface to the BAILII system. We had been aware that the original interface was sufficient for searching, but it was decidedly unfriendly. Pursuant to a report summarizing the findings of the initial analysis and recommending specific improvements to the website and search engine, BAILII implemented enhancements to the database on a test server in May 2006. The enhancements included visual improvement of the screen displays; more points of access for initiating searches and selecting databases; additional options for locating documents by title; additional options for results display; revised search help and some basic subject indexing of the leading cases in core areas of law. The website underwent continual modification over a period of several months. Comments on the usability of the website received via the feedback email form provided on the test server were taken into consideration. In November and December 2006, 30 law students and legal academics participated in a usability study of the test server interface. After each phase of testing, the website was revised in response to student comment and observed behaviour. The resulting interface replaced the original BAILII interface on 4 February 2007.

The Open Law project terminated with an updated database of undergraduate useful case law and an improved frontend, and was allowed to settle for a period of time. Having carried out that work, we were then interested in finding out whether the project had impacted - as we hoped - upon student usage of the system. In early 2012, we therefore added questions to our online survey which were particularly relevant to answering that question.

5. Who Uses BAILII Overview

The responses of the 3,274 participants in Bailli's January 2012 survey demonstrate substantial support for BAILII as an open access mechanism, a technically competent dissemination tool, and a useful resource for lawyers and non-lawyers alike. Such positive response, we suggest, indicates that BAILII's resources are now threaded through the fabric of UK digital legal information, strengthening the ability of all citizens to access and become better informed about the laws of the land. And because BAILII is freely accessible via the internet, it is a valuable resource for students, law practitioners, other professionals and the general public worldwide.

The 2012 survey results, although not necessarily reflecting the extent of BAILII's user group internationally or professionally, provide a useful snapshot of who uses BAILII. [67] 78% of the respondents were from the UK and the remaining 22% from elsewhere in Europe, North and South America, Asia, Australasia, and Africa. [68] 57% were directly involved in the legal profession (solicitors, barristers, judges, law librarians, etc.). 27% indicated employment in other professions and businesses that need access to UK/Ireland court and tribunal decisions. Law students and academic lawyers comprised 16% of respondents. [69] Self-represented litigants and members of the general public comprised the remaining six percent. Our earlier survey indicated a higher percentage of educational users; we consider that the change in percentages has been due to the increase in professional users rather than a fall in educational users. What is striking is the large group who we can describe as 'other':

BAILII came into being partially because of a perception that the professional lawyer was unable to access the law of land at a reasonable cost and that publishers were maximizing profit by fragmenting access to the law. Certainly, with 57% of survey respondents involved with the legal profession, that original need appears to have been met. But it was envisioned early on that BAILII would provide access to the law not just for lawyers, but for ordinary citizens who have no meaningful access to the laws of the land. [70] In a paper from BIALL to the Library and Information Services Council in 1985, it was noted that 'the standards of collections and level of legal reference service in public libraries is so abysmally low (with one or two honourable exceptions) as to be useless.' [71] The survey results show clearly that BAILII has changed this landscape and multiple sectors of the community are taking advantage of BAILII's resources.

The survey responses reflected that BAILII attracts a 'new generation' of users of legal information whose need for access to UK legal information was largely invisible and unmet prior to BAILII's inception. A short extract from the survey responses illustrates the wide range of BAILII users who are not primarily or directly involved in the legal profession:

• Local Authority Public Rights of Way Officer
• Freelance law costs draftsperson
• In House Tax Manager, covering both VAT and Corporate taxes throughout Europe
• Registered Nurse working in a hospice. I have an MA in Medical Law and Ethics from King's College London.
• Environmental Health Officer
• House keeper. I enjoy reading the cases.
• Academic teacher not based in UK or Ireland
• Management Consultant
• Support Immigration Detainees and keep records of cases which their solicitor may overlook
• Scottish Law Commissioner, Queen's Counsel
• Health student
• Criminal Justice Inspector
• Information service, including immigration
• Legal anthropologist
• Trade Union Official
• PhD student, School of History and Archives, Dublin
• Home Office Presenting Officer
• Cost Draftsman/Director of Cost Drafting Firm
• Former jurisdiction adviser in the Office of the Parliamentary and Health Service Ombudsman (and specialist in public administration and administrative justice)
• Senior Home Office Presenting Officer

It is not surprising that the user base is so wide given that law has entered so many fields in terms of, for example, commercial regulation, human rights, immigration, and employment.

Users who took the survey were asked how frequently they used BAILII. 42 % were 'medium users' (several times a month), 32% were 'heavy users' (several times a week) and 26 % were light users (every now and then). Case law databases containing decisions of the courts and tribunals are the most heavily accessed resource. 89 % said they usually turn to BAILII when they know about a case and want to read the judgment. Roughly 50 % use BAILII to search for cases relevant to a legal issue and 41 % consult BAILII to update their knowledge of the status of cases (whether a case has been cited in a subsequent judgment or decision been issued on appeal). 39 % rely on BAILII to keep abreast of the most recent judgments as BAILII is the most up-to-date source for recent judgments which are usually available within hours of receipt. Only 18 % of respondents said they use the legislation databases. No poll was taken to ascertain usage of the secondary sources materials offered on BAILII (including Law Commission Reports; Scottish Law Commission reports, consultation papers, and discussion papers; Northern Ireland Law Reform Committee publications; and Treaties) although several respondents voluntarily mentioned they found these extremely useful.

6. Educational Usage

Given that several major changes were made to BAILII during the Open Law project, we were interested in whether it - as a project specifically directed towards use by students, rather than lawyers - had made the impact which the application for funding had promised. An important element of this question is 'are we integrated with legal education' and how might we improve this. Of those who responded:

• 5.3 % were academic lawyers based in UK or Ireland (n= 173).
• 1.9 % were academic lawyers not based in UK or Ireland (n= 62).
• 9.2 % were law students (n=300).

This gave us quite a substantial sample of almost 500 UK respondents, with one third staff and two thirds students. In the following sections we report on the responses to our questions.

Generally our student sample reflected positive evaluations of Bailli's resources. One third of students responded to the query requesting comments. Overall happiness at system was high:

'I think BAILII is a wonderful resource that has served me well through both undergraduate and postgraduate legal education. It is a particularly useful resource for my hobby as a legal blogger because it provides resources which are not locked away behind a paywall, and are therefore suitable for linking in blog posts intended for a wider audience.'

And they gave advice:

'It could be immensely popular with law students as they will be able to use citation managers such as Zotero that are forbidden with the commercial services (Nexis / Westlaw). I suggest that if BAILII moves fast, it could gain an incredible advantage over other services. What this means is that I do not have to spend hours typing in references.'

But with the usual student grumbles:

'The cases need to be presented on screen in a more user-friendly manner. E.g. there should be gaps between paragraphs. The line spacing should be similar to that of a good newspaper website. I venture to suggest that the way in which news websites such as the NY Times would be a better model of how to display a case, etc.'

In the next sections we look in more detail at student responses.

7. Do you add links?

One of the advantages of BAILII is that links can go directly to a case or even a paragraph in a case without a 'login' process. We wanted to know if this was used in teaching and why/why not.

There were a number of reasons given for why users did not add links to their teaching materials which pointed directly to a judgment or section of a judgment. The included:

1. Not responsible for preparation of teaching materials.
2. Provide written materials in other way so students don't have to link.
3. Prefer to get student to find cases themselves for research purposes - 'I never insert links into teaching documents for cases - looking cases up is an important skill for students to learn.'
4. Other providers have headnotes and analysis.
5. 'Hyperlinks are more trouble than they are worth.'
6. 'Students only 16-18 and need guidance to determine importance/relevance of decisions'

We can see here a number of diverse perspectives - teaching materials are being provided (probably in a bundle) to students rather than encouraging digital access; a view that students should research cases themselves rather than be spoon fed. Interesting is that view that the added value of the law report is a useful facility for law students to utilise. It is possible, of course, for BAILII to provide some kind of contextual introduction to fields of law (much as Cornell do with their WEX dictionary and encyclopaedia [72]) but this could not be from its own resources. Rather, some wiki or blog-approach might be suitable as a joint venture with law schools.

And finally, the point is made in (f) that not all students using the system are at University level. We had, with Open Law, stated that we would attempt to fill a need if it came from schools. However, our feeling is that the teaching materials produced could be used as a way to provide the 'guidance' our respondent suggested was missing from BAILII.

The responses, in some ways then, make us consider that we are still at the early stages of linking BAILII and legal education together, given that such a large group of our respondents feel that linking is not for them.

8. Is BAILII your first choice for student use?

We might expect some bias towards BAILII (given that non-users will not be completing the survey), but clearly BAILII is now being seen as important resource even though our database remains limited in terms of historic case law.

One question of interest is whether students go to BAILII without explicitly being directed there. We also asked whether students went to BAILII because they were told to go there, or whether they went by themselves. 64% say that they took themselves to the BAILII site, indicating that their usage is not simply being pushed by teaching staff; in fact 38% said that BAILII was not generally mentioned during their law courses, [73] yet that 38% were on BAILII and responding to the survey. The majority of students were heavy (47.3%) and medium users (41.2%) of BAILII. In addition to using BAILII for access to cases, students reported use of Westlaw (66%), Lexis (55%) and court and tribunal websites (40%). Clearly, from the 64% who arrived by themselves, students now see BAILII as a useful resource:

We were also interested in knowing whether students had easy access to BAILII from outwith the University. 94% said they use BAILII from place of work or study, with 74% using BAILII from home. We are not sure whether this tends to demonstrate that the system is usable for part-time students and that access from home is non-problematic, or that with 29% not using this from home, there are issues of internet accessibility. The general figures for internet home usage in the UK is 80% of all homes in 2012 [74] having access, but students who are in rented accommodation may have less access than is available to the less peripatetic student.

9. What materials on BAILII are most used?

There is a tendency to think of BAILII as only providing case law. In fact there is a large corpus of legislation on the system, which appears in as enacted format. The Statute Law Database project - now transferred to the National Archives [75] under the 'legislation.gov.uk' rubric - provides legislation as required at any specified date, which should be more useful to any user of legislation. However, there have been problems with the project which have arisen through the complexity of UK law, its updating 'by reference', and the cost and effort involved in providing a completely updated system. While the system is certainly useful to the professional lawyer who understands the nature of legislation, it appears to be a difficult tool for the student to utilise. We were interested then whether law students were using this, accessing legislation from BAILII or other resources, and also whether they used other commercial providers.

In answers to a question about what resources students use for access to cases, decisions, and legislation , responses reflected that 98% used BAILII for access to cases; 28% BAILII for legislation; 66% Westlaw for cases; 55% Lexis for cases; and 40% court & tribunal websites for cases and decisions.

For the significant number who use BAILII for legislation, they will almost always be working with out of date law (unless they manage to put all Statutory Instruments into their mix) but perhaps for most student usage, the most important element they need is an overview of the legislation and its general intent, rather than a detailed description of every amendment. What was surprising to us was the continued use of court and tribunal websites, given that many of these are difficult to search, and are often PDF based (but students may prefer that). What was not surprising was that other commercial providers are used. The main reasons for the use of these other commercial suppliers were not difficult to understand given that although Open Law provided a significant number of cases, it did not cover every need of the student. Some representative student comments:

• Older materials available elsewhere - 'Other databases give access to law reports.'
• Others have materials and citators allowing a more integrated view of law - 'Westlaw is easier to link to related articles.'
• Tradition - 'It is when a decision is unreported and it is on BAILII - if it is reported, I suggest the reported version of the case.'
• System set up - 'Often not your fault. The student may already be logged into Westlaw or Lexis and they/we use a version of a case sourced from there.'
• Some prefer other search engines - 'The search engines are better, it is readily accessible, and easier to use.'

As we discuss below, there are projects underway which will significantly improve the integration of BAILII with other materials, so we would expect some change in attitude towards BAILII re: how it provides a more coherent entry to legal searching.

10. Usability and Search

The second strand of the Open Law project was to improve the original front end of BAILII so that it was more user-friendly. We did not query our 2012 respondents against the older version, because we presume that current law students would not have seen the original. Instead, we simply asked them whether they found the interface easy to access:

The results indicated:

• 57% - Easy to use;
• 39% - Moderately easy;
• 4% - Difficult.

Students reported they found their searches mostly satisfactory in that they found the case they wanted. But - perhaps not surprising with law students who can be seen as the perfect exemplars of the Google generation - advanced techniques were rarely used, and only 25% of students bothered to use the 'limitation by court' facility. BAILII offers several search options and filters that are easy to use and, if used, can substantially reduce the number of irrelevant cases which are presented to the user. Because a recurring lament was that it took time to sort through search results (which can be ordered by relevance, jurisdiction or date) it is somewhat surprising that few students take the time to do searches using these useful and basic tools to help sieve the useful from the not useful. Luckily we found that very few ever tried to contact BAILII for support, since there is no 'user help desk' provided. It may be that law librarians are the first port of call for students who have search strategy difficulties.

The ways in which digital information is presented and accessed are evolving at a breakneck pace and patterns of information use within the legal community are in a state of flux. [76] The BAILII survey reflected the not so surprising fact that comfort levels with search engine technology in general vary greatly both within and without the legal community. Users - not just law students - who expect all search engines to perform like Google are at a distinct disadvantage as the thrust of what Google does actually removes the searcher from the search process by interjecting search algorithms between the user and the materials. Google-like search query processing is expected by 'digital natives' - that generation born during or after the general introduction of digital technology who make up the bulk of students. [77] It is also often expected by those who have no experience performing legal research in databases such as Westlaw and Lexis, which until fairly recently required careful attention to syntax, field context, truncation, and Boolean operators. [78] As Palfrey notes: 'The Google-search practice of digital natives translates poorly into searching on Westlaw and LexisNexis most of the time'. [79]

To retrieve a known case on BAILII, you can search by citation and by title. [80] You can also browse alphabetical title lists and case lists (by court and date). BAILII's case law databases contain only raw judgements without any headnotes, indexing or associated case history. Consequently constructing a search query does not involve the complex legal knowledge often needed for successful searching and navigation in commercial legal databases. Nonetheless BAILII's search engine requires some attention to basic syntax. It has no same sentence or same paragraph operators, but it will process numerical proximity connectors that can effectively focus search results. Users with more advanced skills can take advantage of BAILII's Boolean, truncation and wild card operators. When a judgment cites a case or other document - such as a Law Commission Report - that is in the BAILII database, there is a hyperlink to the cited material.

You can limit a case law search on BAILII to specific jurisdictions and courts and you can apply date limits. [81] When a search is complete, the results are presented in a list that the user can sort and re-sort by relevance, jurisdiction, date (oldest last and oldest first) or title. To browse the text of any given judgment returned in a search result, users can either (1) click on the red context arrows which mark-up the case to show where the search terms occur and/or (2) use FIND (alt-F) to search for words or phrases within the case. From the survey comments, it appears another Google-driven habit interferes with search efforts: some users will simply scroll through a large number of hits rather than put in the effort to comprehend the reason their query has returned an unmanageably long list of hits and appropriately refine the search query accordingly.

Keyword text searches on BAILII can be more problematic, especially for the inexperienced user. The 2012 BAILII Survey results contained laments to the effect that 'it [the search engine] can never find what I ask for'. This comment reflects a classic misunderstanding on the part of users who fail to realize that their search results give them precisely what they have asked for in their search string which is not necessarily the same thing as what they want. It is the weakness of the query, not, as many users presume, primarily the fault of the search engine. That said, certain searches will require somewhat complex query formulation to be successful on BAILII due to the nature of legal research in general and because the database contains only the text of raw judgments. Even a skilled researcher can have difficulty composing a syntactically complex question due to the nature of legal language and concepts. This factor, combined with the fact that BAILII's case law content is not comprehensive, can result in frustrations that strain user tolerance and lead to diminishing returns.

11. User Evaluations Generally

The survey posed a number of questions about what search options were used, how successful users were in finding the cases they wanted, and how user-friendly they found the interface. Overall responses (that is, not restricted to educational users) were extremely positive with 61% evaluating BAILII as easy to use, 37% as moderately easy to use, and only two percent as difficult to use. When asked 'Do you usually find the case you want?' 70% said almost always and 30% said sometimes. There was no appreciable difference among the various user groups (lawyers, non-lawyers, academics) in how they ranked BAILII's ease of use. The survey elicited user comments asking the open-ended question: Do you have any difficulty navigating BAILII's website to find what you want, or any general comments on BAILII's website or BAILII itself? This question generated 1,434 responses that were very revealing as to overall attitudes toward BAILII, how much is understood about the nature of BAILII, and usability issues. The comments were overwhelmingly enthusiastic, describing BAILII in superlatives: brilliant, excellent, invaluable, superb, marvellous, fantastic, vital, incredibly useful. Negative comments comprised only a small fraction (2.8%).

Predictably, many suggestions for improving usability involve data enrichment that is beyond BAILII's ability for the most part: expanding the scope of case law coverage (primarily adding older cases), adding headnotes/indexing, and adding citation features (case history, subsequent citations to case, authorities cited in case). There was also a demand for printable PDF files. [82] Browsers such as Google Chrome are now integrating PDF printing into their tools, so the demand for a propriety document format may be met by others without BAILII having to do this. [83] Interestingly enough, these desires mirror the most popular demands for improvement reported in the recent LEXUM survey of the uses of its Supreme Court of Canada decisions website which also revealed that CanLII, BAILII's counterpart in Canada, 'tops the market when it comes to accessing primary [Canadian] legal information'. [84]

BAILII has studied the survey comments regarding usability and is in the process of identifying improvements to the interface that can be accomplished within BAILII's existing resources. Perhaps not surprisingly, what BAILII needs most is a 'facelift', so to speak. The developments from Open Law improved the original substantially, but we see a constant need to re-engineer the front end to keep up to date with current fashion and use. BAILII is a content heavy website and it is apparent from the comments that many users do not actually see all of the available BAILII resources, search options, and search help. BAILII aims to make revisions that will visually streamline the website and provide tutorial guidance.

12. BAILII as part of the information landscape

The commercial legal database business model is designed to meet the needs of consumers in the legal sector who can afford the cost. Commercial legal information providers, supported by their subscribers, have vast sums to invest in state of the art programming, editorial enhancements, training, public relations and advertising. Employees of legal practices, government entities and academic institutions that can afford the high cost of access flourish within these 'walled gardens' of richly enhanced legal information. [85] But commercial databases are outside the reach of many legal practitioners who cannot afford the cost. And even putting cost aside, these models do not necessarily address the needs of other consumers, including non law trained professionals, academia in general and members of the public. The open access model eliminates both price and permission barriers. In the tradition of a public library, BAILII's content is free of charge, open to all, and available worldwide in any location with internet access. BAILII's presence on the internet has thus significantly improved access to UK and Irish case law for jurisdictions both within and without the UK and Ireland and as such is a unique and readily accessible resource for users of legal information who would otherwise have only very limited access or none at all. The 2012 BAILII survey results demonstrate that:

• Lawyers and non-lawyers alike have need and desire for open access to UK/Ireland law; and
• BAILII's content and search utility is manageable for law trained users and those without legal education.

Although some users of BAILII have no other access to UK law, many lawyers, judges, and academics actually have access to commercial legal databases but nonetheless are frequent users of BAILII. One user comments: 'I think BAILII is a fantastic resource and research tool which puts the likes of Lexis-Nexis and Sweet & Maxwell to shame. Despite subscribing to these I find BAILII is my normal first point of call as it saves a massive amount of time for case searching'. This is especially true of the student respondents - 75% said that BAILII was their first choice over Westlaw and Lexis.

BAILII welcomes organisations and individuals to link to its documents with the caveat that subscriber or pay per view sites make it apparent to users that BAILII content is in fact free and that the origin of the content is clear. There are around 3,000 domains currently linking to BAILII. BAILII is the backbone for many popular legal blogs and commercial publications in the areas of commercial law, employment law, criminal law, human rights law and media law. [86] A hyperlink to BAILII is stable - the linked document will neither change nor vanish so long as BAILII is alive. Anyone - lawyers and non-lawyers alike - can incorporate hyperlinks to BAILII and its documents when preparing digital information: blogs, email communications, teaching materials, news articles, etc., and URLs for BAILII's documents can be noted in print publications.

13. Funding Open Access

BAILII arrived at a time when the law business was doing relatively well - funding from legal aid was accessible, students on leaving law school managed to enter practice. The general sense was that the provision of a legal service to the population was good and might be getting better. Even at that period, though, law was seen as being a too expensive commodity when provided by the private sector. Indeed the cost of access to law was the prime initiator of BAILII as critics noted the tendency of publishers to fragment their products to maximize profit rather than access. [87] We now see a considerable change in environment: legal aid is difficult to access; students (and indeed lawyers) are finding the job market problematic; solicitors are under pressure to reduce costs; self-represented litigants are increasing; legal advice centres are expected to provide a high standard of advice but at the same time to operate with low overheads; and there are reductions in funding for national and local governmental departments. If it could be said that in 2000 the primary target user group for BAILII was the lawyer in private practice, this can no longer be the case: the need for an effective and low cost method to access up to date law for all sectors of society has risen with the current economic climate.

Within the legal information landscape BAILII is becoming more robust in terms of content, links to BAILII are becoming increasingly ubiquitous, and thus BAILII is becoming ever more woven into the fabric of legal information and communication. BAILII relies on charitable funding to maintain and enhance the content and service it provides and has launched an Appeal for Funds directed towards sets of Chambers and solicitors' firms and welcoming contributions. [88]

There has been some criticism of BAILII for allowing Google to access case citations and titles but not the text of judgments. [89] There are two primary reasons for this, one of which is tied directly to funding:

• BAILII needs protection from having its data mined; and
• BAILII has a responsibility to maintain and protect the veracity of its data.

In order to continue its funding as a charity, BAILII needs to demonstrate that it is well used and functions as its goals suggest. With regard to protection, if BAILII usage decreases as a result of judgments and other materials being replicated and accessed on other sites, BAILII will fail and BAILII will fold. With regard to veracity, if it is necessary to remove a judgment from the BAILII database - for example if BAILII has been provided with a copy of a judgment which discloses confidential information such as the name of a child in a sexual assault case - BAILII has no control over that judgment if it has been harvested by another provider. Things are made more difficult once judgments have been indexed by a search engine. And if the text of a judgment is amended by the judge, the unamended version should not be available in the cache of a search engine.

Respondents to the 2012 BAILII Survey volunteered views regarding the provision and funding of open access law. A fee paid judge who uses BAILII when sitting in courts with limited resources and as a practicing solicitor said:

'I find it almost incomprehensible that the UK could find itself without free to access case reports, particularly where the number of litigants in person is rising, and they need access to the law, without fee. In days gone by, lawyers only relied on reported decisions, meaning that most decisions of the courts were not recorded, or available in future cases. Now it has become the norm for courts to be referred to all sorts of previous decisions. Lawyers and LiPs must have a way of obtaining these quickly.'

Many expressed their opinion that open access law should be provided/supported by the government:

'I would have no objection to public funds being used to help provide such a service, and can see good arguments for their being provided.'

'I think it is a great service, and given number of decisions being produced by the court, and the expense (particularly for small law firms and sole practitioners, never mind students and the public) of 'knowing the law', it really should be a state funded service.'

'I would support a campaign to release the Law Reports from ICLR and have them on BAILII.'

'It is appalling to think how much of tax payers money goes to the big two ... I would much rather see BAILII get direct financial support from the government than see the amounts spent on the big two providers.'

'Bailli's function ought to be provided the the UK Govt. It is a shame, nay, a Disgrace in the information Age, it is not.'

'My general feeling is that it is totally taken for granted and that the profession, which pays huge sums to Lexis Nexis and others, should be much more willing to supervise it.'

'I think it is an essential part of English life in the 21st Century, if it was not there, Govt would have to provide it and we all know what Govt IT is like!'

'BAILII should be publicly funded. I would in principle be willing to contribute but would want the government to say why it won't fund BAILII first, before committing myself or my organization'.

A view which we have articulated elsewhere is that BAILII has followed the model of the 'National Law Library' [90] which was proposed in the 1970s by the Society for Computers and Law. We see in these comments support for that view suggesting that government funding should be available for BAILII since it is a core democratic resource. Currently there are discussions with the Ministry of Justice about whether BAILII will become a national repository of case law, which would meet the needs of the harmonized access to European case law which is currently being set up. [91] Unfortunately, there appears to be no linking of the various parts of the legal system (courts, ministry of justice, government users of information) to come together and calculate whether cost savings could be made through utilizing BAILII as a central and government funded resource.

BAILII is clearly used by law students, yet funding from law schools has been minimal with only a handful of law schools providing support. There is a greater awareness of open access in the academic environment, with research councils articulating a model whereby they will provide funding to enable publicly funded work to be made publicly available. [92] It is not clear whether or how such a model would impact upon BAILII: it is a resource upon which publicly funded legal studies are often based, and thus is part of the open access environment. If BAILII was to terminate due to lack of funding, a possible result could be increased charges from the commercial providers who no longer have an open access competitor to consider in their charging regimes. This would likely impact in a major way upon the smaller law schools or universities and colleges (and in secondary education) where law was not a prime field of study - just the situation which existed prior to BAILII coming into existence.

Publishers of digital materials have always complained that no matter how much is made available to customers in a product, the customers always demand more. [93] BAILII appears to be being treated in exactly the same way, here, as our 2012 respondents pointed to the fact that BAILII did not contain all the law they wanted:

'Usually no problems. But seem not to be able to find the older cases 1700's etc. I do not know how far back the cases go and it would help if I did. it is my first place to search for a case when it is not found, I then turn to Westlaw, but would rather use BAILII.'

'Would be a most useful exercise if all cases were available on the website as opposed to just a few. All cases of all years.'

The Open Law project provided a significant boost to BAILII by allowing it to build up its database of historic materials, yet there are still significant gaps. A survey carried out in 2006 [94] indicated that there was desire to have House of Lords materials added to the BAILII database. For judgments the responses were:

And for 'other' materials [95] while the response was not so positive it still indicated a substantial desire to have access to these materials:

Funding for such a project would probably not be of interest to those who currently support BAILII most (that is, the professions) since the advantage to them is the cost, speed and ease of access to current law. Given the low level of funding from law schools to BAILII, it is unlikely that we would have sufficient resources to carry out such a useful task without funding from some central resource such as JISC or some other organization which supports basic research material provision.

14. Does BAILII Help Overseas Law Student Recruitment?

One of the pressures on law schools is to recruit overseas students. [96] It can be difficult to know why law students from overseas actually choose one possible location against another, but perhaps one reason is that they have access to the law of their destination from home. This might be important for several reasons: first, they can see the breadth of topics which reach the courts and the standard of analysis which the judges apply; second, they may feel that having a UK legal experience would be wasted if they were unable to get access to judgments easily after their degree and keep up to date with developments. But no matter where the study of law is undertaken, there is a growing global component that needs to be addressed in law school syllabi. [97] The general open access model appears to be appropriate here: accessibility substantially improves use of materials. [98]

It is clear that BAILII - as a free to user service - is well used by overseas lawyers. For example, there are indications from IP addresses attached to our site's Feedback Form that overseas access is frequent and positive feedback is the norm:

'In addition to being a practitioner (specialising in the field of Intellectual Property and Information Technology) I am also a visiting lecturer teaching in Comparative Intellectual Property at the Swiss Federal Institute of Technology (ETHZ) in Zurich. The BAILII site is of course excellent and one that I find very useful in both my teaching and practitioner roles.'

'I just thought I'd drop you a note from a grateful Canadian who found BAILII extremely useful. It's a marvellous world when I can go from shaking my head at 'Fyfe v. Glasgow Hospital, how will I ever find that?' to printing out a copy of the decision less than two minutes later. After enjoying the versatility of CANLII for so long, it's a delight to extend our reach even further. Thanks for your excellent service. (Professional)'

'I am very impressed with this site that I now have it listed on my favourite website list. Here in Papua New Guinea (south pacific islands) it costs an arm and a leg just to locate English Case laws. In the whole country we have only two main law libraries. One at our national University and the other at our Supreme Court. So gaining access to the English law cases is difficult. Thank You. (Professional)'

Users of BALII, though, are frequently naïve about the commercial context of legal information provision. For example, one Dutch student suggested to us that commercial publishers would not object to copying reports they had published - a situation far from the truth:

'Please include more cases! Especially the older cases, before the Law Reports, are very hard to find on the continent. In the Netherlands, only the Peace Palace has a complete set. Moreover, making cases more than 20 years old freely available will not raise objections from commercial publishers.'

An interesting question, therefore, to ask students from overseas would be whether they had used BAILII prior to entering UK legal education, and also to ask whether they used it after completing their legal studies and had returned home.

15. Future Educational Developments for BAILII

We should note that since legal information systems are no longer stand alone, but can communicate with other systems, there is the important aspect of interconnecting with other information sources. Most recently, the ICLR, [99] publisher of official law reports, has partnered with BAILLI, linking their two services online to provide a seamless delivery of content to the legal community. This service was launched late 2012 and will provide benefits to users of both services. ICLR users will now be able to access the huge database of reported and unreported judgments on BAILII, and BAILII users will be able to access at the ICLR summary of a judgment (where one exits) and purchase the authorized case report, or go directly to the authorized case report if already an ICLR subscriber. Users can move from one database to the other whatever their point of entry. BAILII is also happy to discuss this kind of linkage with the other publishers.

There have been one or two attempts to utilise App technology with BAILII. Crimeline, for example, which aims to ease the searchability of BAILII on mobile devies, [100] and most recentlyLawSauce, an electronic resource locator for over 7,000 records covering case law and legislative sources for multiple jurisdictions worldwide. [101] BAILII considered setting a 'Student App' competition in 2012, but the feeling of the Trustees (decided narrowly) was that there could be difficulties in supporting an App should it be problematic or not be updated: the concern being that the BAILII brand was important and had to be protected from situations where users could blame BAILII for something which was not within their control. However, no doubt such projects will come to fruition in the future.

BAILII also has a link to LawCite, an international legal case and journal article citator that is a collaborative project of the Free Access to Law Movement and was awarded a high commendation in the Innovation category at the Australia and New Zealand Internet Awards presented in Canberra in 2012. [102] This case citator allows easy visual information on which cases cite with others, etc. and offers to become a powerful tool for the student researcher. It is not currently in a 'finished' state, [103] and knowledge of its existence in the legal education world seems to be relatively low. As a research tool in its present state, even though it cannot claim completeness, it offers benefit to the student and academic. BAILII has also provided a location for 'other materials' to be made accessible: books which are now out of print, open access journals requiring mirrored sites, etc. We perceive that this function will increase in future years as the LII movement builds upon its success from over the past decade or so.

Overall, the survey appears to have shown that the Open Law project was worthwhile in that it broadened the usability of the interface and also encouraged more student use. There are clearly questions of how we further develop BAILII to integrate more with legal education (how, for example, to integrate 'other' materials/citators) but that requires input from law teachers. What is the best way to build the community of legal educators who can make best use of, and mould, BAILII to ensure that students read cases? These are issues, perhaps, for the next decade.

16. Conclusion: Legal Education and Open Access

It seems to us that there three points which are worthy of emphasis in conclusion. First, there is the general openness to law which BAILII has encouraged whereby the citizen and the non-legal professional have been able to access law in its fullest for the first time in centuries. This is a point of education in the very widest sense: the building of social and intellectual capital within the community. The evidence gathered from our survey surely supports the overwhelming success of BAILII's role in building that capital. BAILII's accessibility and hyperlink capabilities contribute to what Schauer and Wise - using the model from sociology and political science - characterize as 'social capital'. [104] Schauer and Wise hypothesized that increasing access to primary law for non-lawyers and ordinary citizens can serve socially valuable purposes, encouraging citizens to see themselves more as part of a common and public-value-producing enterprise and less as partisan adherents to one or another warring faction. [105] Such a perspective echoes the underlying ethos of the EU's Information Society Programme and the Montreal Declaration on Free Access to Law. [106]

This is not to suggest that, standing alone, merely making primary law freely available empowers the average citizen or suffices to meet needs of lawyers. Leiter makes the point that free materials, even if they comprise the sum total of all primary law in a country at every level and jurisdiction, will amount to only a minor portion of the materials that lawyers need in order to practice law, and the public needs in order to understand it. [107] The law for lawyers and non-lawyers alike needs 'curating'. One such curating effort is freelegalweb.org, a project 'designed to deliver a web service that joins up and makes sense of the law and legal community and analysis on the web, providing a substantially more reliable, useful and efficient service to both lawyers and the community at large than is currently available'. [108]

There is much discussion both within and without the academic legal community about the open law movement. In particular there is a great deal of debate about who can and should take on the responsibility for providing open access to law and legal scholarship such as Armstrong's, 'Crowdsourcing and Open Access'; Plotin's, 'Legal Scholarship, Electronic Publishing and Open Access'; Gallacher's, ''Aux Armes, Citoyens!:' Time for Law Schools to Lead the Movement for Free and Open Access to the Law'; Arewa's, 'Open Access in a Closed Universe'; and Carroll's, 'The Movement for Open Access Law'. [109] [110] [111] [112] [113] The research councils grouping in the UK (RCUK) has also recently taken the decision that publicly funded research should be available through open access channels. [114]

Second is the particular role of BAILII in legal education. Here we found a significant effect upon use from the student community after the Open Law project. What was clear is that BAILII is viewed as a solid resource, and one which - apart from its lack of historical materials - is viewed as being of as least as much value as the commercial enterprises. However, on a more negative view, the hope that we had in the Open Law project of increasing the integration between teaching materials and BAILII does not appear to have been as successful as we would have wished. For example, students are not being enabled to utilise the full extent of BAILII's search engine and seem often to prefer the Google method of hoping that the required document appears on the first page of the search results. From the point of view of integration of teaching materials with BAILII, there does not appear to be significant usage by law teachers of even the most basic linking to paragraphs etc. which is a relatively simple technical task yet enables deep linking to sections of long judgments where the important arguments may be found. [115]

It is not clear just why there has been this failure (though failure is probably not the correct word). Certainly other attempts to move technology into law schools have been just as limited in success and we wonder whether the cultural context of the law school is one which is difficult to develop. Certainly those who have tried to do so have rarely felt encouraged: we remember the outcry in the late 1980s when staff were encouraged to do their own word processing rather than use secretaries to retype handwritten notes. Similar experiences are now being met as law staff are encouraged to use electronic submission of essays rather than paper-based ones, and the level of whingeing about having to use technology rises yet again. [116]

BILETA, as the organisation which was most concerned with technology in the UK law school in the 1980s and 1990s moved, to an extent, away from that role and towards supporting information technology law and suchlike. It seems, perhaps, that there is still a very substantial role for BILETA in encouraging the development of staff expertise and interest in using technology in the law school in parallel with the UK Centre for Legal Education. [117] This is not to say that there is no research or interest in legal education technology - that is not the case with a variety of approaches in the UK such as Bromby. [118] Maharg, too, has been one of those who has been fruitful in his experimentation in legal educational technology yet almost seems to be in a parallel universe from the bulk of the law school establishment. [119]

Third, we can point to legal information research as a research field in the 1970s and 1980s with various researchers interested in how to find that single judgment in the haystack of a legal database. Researchers such as Tapper, Bing and Sprowl, for example, were recognised for their legal scholarship as well as their interest in the technology of searching law; they were intimately interested in the relevance of search engine results and saw it as a fascinating and important area in which to do research. [120] We have certainly not arrived at a position where there is no need for more research on how to search, how to build search engines, how to make usable interfaces, etc. etc. which are relevant to law, but the field - at least to us - currently appears lacking in vitality, despite the clear evidence that it is an important topic for lawyers. [121] Systems such as Google have shown that research on search algorithms can produce significant development - so why, we ask ourselves, is there so little interest in finding algorithms to search law? There seems to us to be little current connection between the law school and departments of computer science and librarianship and questions such as usability of legal research systems receive little focus, despite BAILII existing as a resource which might help enable such research work. We would hope that having BAILII as a tool and a data resource might encourage younger researchers to consider a research career in the field. We can only hope that law schools see the utility of such research in building tools to support teaching and learning in law. Overall, though, BAILII and the law school - we might say - have begun their marriage and hopefully over the next decade or two, legal education technology might revisit the questions raised about how to better enable better legal research through online tools.

Appendix: BAILII's Search Engine

A legal information system is composed of four parts - the data, the method of structuring that data within the database, the search engine and the user interface. All four need to be effective for the system as a whole to be effective. BAILII's data is structured in libraries which match the differing courts and jurisdictions but which are accessible via a publication neutral citation. BAILII uses the search engine SINO ('size is no object') which was originally developed by the AustLII project in 1995 but has during its usage by BAILII been added to and amended by the BAILII team. Andrew Mowbray, who designed and programmed SINO, has pointed to various elements which make it suitable for running a legal information provision accessed by large numbers of users, but supported by relatively small technical teams:

• Sino is very fast both in retrieval and indexing times
• Sino is easy to interface with
• Sino is relatively small and easy to understand
• Sino has been in use on a number of major web sites answering many millions of requests for the past 10 years and so is robust and reliable.
• Sino is open source and is licensed under the GNU General Public Licence [122]

These advantages make it suitable for usage in BAILII. However, they do not particularly emphasise the user interface for the search engine, and this is usually the weak part of any technical system: if untrained users find the user interface unfriendly or difficult to use then they will go elsewhere. SINO is certainly powerful and can be used to carry out complicated searches, but system designers who utilise it need to insert a usable interface between the user and the core search system. The ultimate route towards simplicity is to have one single text box on the user interface, but this is not desirable in a legal information system because searching all the various types of legal source material (court judgments, legislation, treaties, journals, etc.) with the same search is often unproductive due to the varying nature of the materials of each type. The original interface for BAILII suffered various usability flaws and part of the Open Law project involved carrying out user testing of the interface and producing improvements. User studies revealed that if it was clear how to limit a search to case law, legislation, or other resource materials (secondary source material such as Law Commission Reports), the chances for conducting a successful search for the sought after material was greatly improved. It is simply a matter of context. In addition to offering search screens specific to case law, legislation and other research materials, enhancements to BAILII included visual improvement of the screen displays; more points of access for initiating searches and selecting databases; additional options for locating documents by title; additional options for results display; revised search help and some basic subject indexing of the leading cases in core areas of law. These have been available since March 2007 and the 2012 survey project was undertaken with the view to seeking feedback that would provide guidance for developing further improvements.