University of New South Wales Faculty of Law Research Series
Last Updated: 17 November 2011
AustLII: Thinking locally, acting globally
Graham Greenleaf, University of New South Wales
Andrew Mowbray, University of Technology, Sydney
Philip Chung, University of Technology, Sydney
This paper is initially presented at the ALLA & NZLAA Cross Currents Joint Conference, Melbourne, 2010 and published in (2011) Australian Law Librarian, pgs 101-116. This article may also be referenced as  UNSWLRS 41.
The Australasian Legal Information Institute’s (AustLII's) Australasian service is the testbed for new Australasian service is the testbed for new developments which are then implemented in AustLII's international services (WorldLII, AsianLII and CommonLII), and often also adopted by other LIIs with which AustLII collaborates most closely. This presentation focuses on the most important new developments taking place on AustLII in 2010, including redevelopment of the results interface to integrate it with LawCite; expansion of LawCite to include law journals, law reform reports and treaties; large scale expansion of AustLII content both horizontally (comprehensive current caselaw and legislation sources) and vertically (historical collections of legislation, case law and legal scholarship); versions of legislation at different times; and content-specific Libraries involving virtual databases. Other new developments in the pipeline will be mentioned including RSS and other feeds, and user-generated content/contributions. How these developments are then adopted by AustLII's international projects and collaborating LIIs is then explained, in the context of the overall aim of a global network of free access legal information. Some new international projects are previewed.
The Australasian Legal Information Institute (AustLII) is best known for operating a free access service for Australian and New Zealand law, also know as AustLII <http://www.austlii.edu.au/> . Over the last two years, development of the LawCite citator <http://www.austlii.edu.au/lawcite/> has also become an important part of what AustLII does with Australian legal data, but LawCite has remained separate from AustLII until now.
AustLII's Australasian service is the test bed for new developments which are then first implemented in the three international portals operated by AustLII in cooperation with other LIIs: the World Legal Information Institute (WorldLII), the Commonwealth Legal Information Institute (CommonLII) and the Asian Legal Information Institute (AsianLII). New developments are also often simultaneously introduced on the New Zealand Legal Information Institute (NZLII), for which AustLII operates the technical infrastructure for its New Zealand partners who run NZLII.
Technical innovations are often subsequently adopted by other LIIs with which AustLII collaborates most closely, particularly BAILII in the UK, PacLII in the Pacific Islands, HKLII in Hong Kong, SAFLII in southern and eastern Africa, and CyLaw in Cyprus, all of which use AustLII’s open source search engine, Sino. Which innovations are adopted by each of these independent LIIs, and the rate of adoption, depends on the local LII.
This paper focuses on new developments in which AustLII is involved in 2010, how they are applied in AustLII’s work both in Australia and internationally, and a redefinition of the goals of AustLII’s work, both in Australia and internationally.
A summary of AustLII’s development of its Australasian resources over the 15 years since its formation in 1995 is in ‘AustLII in 2010: A snapshot’. Its global involvements are summarised in the second half of that document. Readers who are not familiar with the context of some of the new developments discussed here should refer to those annexures.
Thinking locally ... AustLII
Redefining AustLII: Aiming to be comprehensive
AustLII was the first legal information institute to aim to provide free access national coverage, to the extent to providing at least the consolidated legislation from every jurisdiction in the country, and the current decisions of every superior court. We achieved this in 1999 after four years efforts to ‘free the law’. Since then, free access has been the default condition under which public authorities have provided legal information in Australia.
Last year, AustLII consciously adopted a more ambitious goal, though it is one we had been progressing toward consistently since 1999. We now aim to provide comprehensive free access to Australian ‘public legal information’, to the extent that this can be achieved and maintained within AustLII’s budgetary constraints. This involves an expansion of AustLII’s content both ‘horizontally’ and ‘vertically’. It is an aspiration which will no doubt take some years to achieve, but progress has started. Perhaps it can be achieved by AustLII’s 20th birthday in 2015.
Expanding horizontally (comprehensive current law)
AustLII now publishes decisions of over 150 Australian courts and tribunals, and we are aiming to publish the decisions of all courts and tribunals which are of legal interest and relevance. In the case of lower courts and some tribunals, this means that only a small selection of decisions will be released by the court for publication, and they will often be anonymised. With funding assistant from the Australian National Data Service (a national research infrastructure fund), AustLII is aiming to publish the decisions of all remaining courts and tribunals in Australia and New Zealand by the end of 2010, which should result in it hosting the decisions of over 200 courts and tribunals by the end of 2010. For many smaller courts and tribunals, AustLII is the only publisher of any of their decisions, but these decisions are of considerable importance to some practitioners and community groups. Among the new categories of case law databases are medical tribunals, and coroners, magistrates and ombudsmen’s reports (not quite case law, not quite law reform). Decisions of small tribunals and lower courts are often of great importance to very specialised audiences, and AustLII tries to assist those audiences as well as those whose only interest is in superior court decisions.
AustLII provides 76 databases of legislative materials, comprising 1,189,000 searchable documents (the number is large because each section of an Act is separately searchable). There are 30M hypertext links within the legislation, including to text items such as definitions, other Acts and other sections. For each of the nine Australian jurisdictions these databases include consolidated Acts (ie including all amendments), Acts as enacted each year, Bills, Explanatory Memoranda accompanying Bills, consolidated Regulations and annual Regulations. For some jurisdictions there are also databases of Repealed Acts, detailed analyses of Bills by Parliamentary Libraries, reports on Bills by Parliamentary Committees, and Explanatory Statements accompanying Regulations. On average, these databases go back about twenty years, except for the consolidated databases which are comprehensive. Many of these databases are new in the last two years.
Two major challenges remain with legislation:
● (1) To obtain all the delegated legislation in a jurisdiction, other than Regulations. The first of which (in 2011) will be a ‘Federal Register of Legislative Instruments’ (FRLI) database of all Commonwealth delegated legislation in force;
● (2) To create either ‘point-in-time’ or ‘versions’ databases for all current legislation, showing each successive amended version of a section of an Act. AustLII’s aim is to do this with consistent presentation and allowing side-by-side comparison with previous versions. AustLII did this some years ago for NSW Acts since 2002, and for South Australian Acts since 2004. In 2011 a similar approach (depending on the quality of date information) will be extended to Victorian Acts and Commonwealth Acts.
AustLII has funding from the Australian Research Council (LIEF fund) and the Victorian Legal Services Board to undertake these developments.
AustLII’s Australian Treaties Library is already comprehensive of new treaty information, including treaties as signed, those ratified (in the ATS database), and seven other databases of explanatory materials. AustLII also has databases of all current Australasian law reform commissions back to their inception, and some are currently up-to-date (eg ALRC, NSWLRC), but others have not been updated for some years due to lack of staff.
A major expansion of AustLII content has been in legal scholarship. AustLII has included some law journals from inception and in 2008-10, due to an academic grant, expanded these into the Australasian Legal Scholarship Library. The Library now contains over 30,000 searchable items of scholarship and may be the second largest free access collection on the Internet (at least in English), after the US-based Legal Scholarship Network (LSN/SSRN). The Library contains the full texts of 60 Australian and New Zealand law journals (mainly from Law Schools; there are four repositories of current academic scholarship (including research not yet published); the abstracts of all law-related Australian PhD theses; the full texts of legal books published by Sydney University Press; and two databases of ‘judicial scholarship’ (speeches and articles by judges). All this content is also searchable via Google Scholar and is highly visible there.
Expanding vertically (developing historical collections)
AustLII’s long term aim is to make all reported decisions, and important unreported decisions, of all Australian courts available for free online access. This will take some years. AustLII’s coverage of the High Court and the various federal courts is already historically comprehensive, but this is not so for State courts. AustLII is at present scanning every reported decision of Victoria’s Supreme Court from 1875-1956 (the Victorian Law Reports (VLRs)), and earlier decisions from 1861, and making them searchable. We will then exchange the digitised copies for a digital set of the Victorian Reports (VRs) from LexisNexis. A complete set of reported Victorian case law for 150 years will then be available for free access from AustLII, and also available from LexisNexis. AustLII has also received the scanned text of over 30,000 unreported Victorian Supreme Court decisions from the Court, and will also make them searchable. When all of this is completed, which we expect will be by the end of 2010, and LawCite has digested the resulting citation information, Victoria will be the ‘model jurisdiction’ for free access to case law. We hope to have similar projects in subsequent years for the other States and Territories.
The first jurisdiction for which AustLII has comprehensive legislation is Victoria, the Victorian Historical Acts database including all legislation from 1851 to 1995 (when the current legislation databases commence). The Victorian Gazette will also be added soon. We aim to develop similar historical databases from other States, and the Commonwealth.
AustLII’s Australian Treaties Library already includes all treaties to which Australia has become a party since Federation. There may be some scope to include colonial treaties which applied to Australia.
In the Legal Scholarship Library, most back-sets go back to the inception of law journals, and for some of the more important journals go back to their inception in the 1950s. AustLII is discussing with legal historians and others what forms of Australian legal scholarship in the colonial period and from 1900-1950 would be it be most valuable to make available. Law Society journals, textbooks and other materials are under consideration, subject to funding becoming available.
LawCite as the glue within AustLII
LawCite is a free access citator, developed by AustLII two year ago, which currently contains citation records for 3.5M judgments, journal articles (about 250,000), law reform reports, and treaties. Citators do not normally include citation histories for treaties and law reform reports, so this is a valuable extra feature of LawCite.
Because there are less than 0.5M Australasian cases on AustLII, this means that LawCite knows about more than 2.5M other cases, either Australian cases not yet on AustLII, or international cases. LawCite’s breadth and depth of coverage compares very well with any other Australian or international citators.
Now that AustLII holds over 30,000 Australian journal articles (and knows about a lot of others, including Australian journals on Hein Online), the depth of LawCite’s coverage for citations between Australian journal articles is very high, and its coverage of where journal articles are cited in cases or law reform reports is unmatched.
Australian legal documents are the largest component of LawCite’s records (because of the number of cases etc on AustLII), but it is an international citator based on collaboration between a group of LIIs. It has very good coverage for most common law countries. It is particularly useful for tracking wherever a cases has been cited in other cases in common law countries, and in tracking citations of Australian and New Zealand law journal articles LawCite’s citator software is developed by AustLII and uses heuristics to recognise and extract citation information from the content made accessible by the collaborating LIIs and other sources. LawCite records are updated daily (where possible), because it is fully automated.
LawCite is of crucial importance to AustLII in the following ways:
● 1. LawCite metrics and links in AustLII search results The most important change to AustLII for some years is the incorporation of information from LawCite in AustLII search results. In a new version of AustLII’s results interface released in September 2010, each item in a list of search results also include an indication (1 to 5 stars) of how often it is cited, and a link to its full citation record in LawCite. Search results can also now be sorted ‘By Citation Frequency’. The relevant parts of LawCite are therefore available from any AustLII search.
● 2. Augmenting AustLII cases with parallel citations The parallel citation information identified by LawCite’s data mining is used to add these extra citations to cases on AustLII every time a case database is rebuilt. As part of this, neutral/AustLII citations (ie those of the form ‘  HCA 1’) are added to cases wherever LawCite can identify an existing citation in a case with its equivalent neutral citation. This means that AustLII can then link from the neutral citation to the case on AustLII, thus continually improving the hypertext linking between cases on AustLII. The same applies to references to journal articles, treaties and law reform reports. AustLII’s data is therefore not static, it is constantly becoming richer and more interlinked.
● 3. LawCite’s own interface allows new types of searches For example, all treaties between the two countries can be found by putting ‘Australia’ and ‘Thailand’ in the ‘Parties’ field. All reference to an Act and section can be found using the ‘Legislation Considered’ and ‘Section’ fields.
● 4. Law Journal metrics LawCite’s coverage of Australian journal articles, and information about citation of Australian law journal articles in other journals, is allowing AustLII to develop metrics for citations of Australian journal articles which do not exist anywhere else. We can combine this with information about access rates to copies of journal articles on AustLII.
Making more of what we have
So far, we have seen that AustLII is developing its already very large collection of legal databases further in toward being comprehensive, and is using the LawCite citator as its principal method of exploiting semantic connections between its different types of content.
The other challenge AustLII is addressing are the means by which it can get more value for its users out of the information AustLII already has, other than through LawCite. There are four main ways we are exploring doing this at present: useful statistics
Providers of databases to AustLII want to know how often their databases are accessed, and users may also find this information of interest and value, so statistics will soon be provided for individual courts, journals etc on the home pages of each database. Comparative tables, updated monthly, of at least the most accessed courts and journals will also be provided. Derivative statistics, such as the average number of accesses in a month per database item, many be more indicative of the significance of particular types of content.
Information concerning the number of accesses to individual cases or journal articles, perhaps also coupled with derivative information about the average number of times per month an item is accessed, might be useful. This could even provide another method of ranking retrieved items. However, the processing implications, and utility, of such statistics, are yet to be explored.
Users’ search terms give AustLII very valuable information in aggregate by which it is possible to predict what other users are likely to be searching for, and therefore anticipate their most likely search options and offer them before the user has finished typing. The LawCite database can be used to supplement search results to suggest material that is not directly available on AustLII (such as international cases and other materials). Such anticipation of searches is already built into LawCite: if you type ‘ma’ in the ‘Parties’ field, it suggests you want ‘Mabo v Queensland (No 2)’, followed by ‘Maher, Waltons Stores (Interstate) Ltd v’, and a mere ‘c’ in the ‘Legislation considered’ field results in the suggestion ‘Crimes Act 1900’. A similar approach will soon be implemented on AustLII, replacing the current ‘Autosearch’ facility. ‘Work-in-progress’ can be seen on AustLII’s beta server <http://beta.austlii.edu.au/> where new approaches are tested.
More convenient access mechanisms
There are continuing benefits to be gained by developing better access mechanisms for users. Here are three examples of recent access improvements:
● (i)The improved ‘Victorian Law Resources’ page <http://www.austlii.edu.au/au/vic/> allows users to conveniently choose any combinations of AustLII’s Victorian databases, and other information relevant to Victoria such as ‘Victorian Law on Google’. The same approach will now be taken to the pages for other States and Territories.
● (ii) A link to a case’s LawCite record from the header of the case has been added to all High Court decisions (see for example Parsons v R  HCA 1) , and for some other courts and tribunals added recently. LawCite links are being added progressively to all other court, tribunal and journal databases.
● (iii) Metadata has been developed for each database on AustLII, to be part of the repository operated by the Australian National Data Service (ANDS), and also provided by AustLII, so that search engines more easily find databases on AustLII.
These are all relatively simple improvements, no rocket science involved, but they are the types of accessibility improvements any information service needs to keep introducing to make better use of its underlying content.
We are also exploring the introduction of alert services including RSS feeds for identifiable new content such as new Bills or Acts, or amendments to existing legislation, new judgments, and journal articles newly available on AustLII. In some cases there are challenges in identifying what is ‘new’. With case law there is the very considerable problem of preventing RSS feeds being used to feed republication without the essential controls of take-down notices sent by courts and tribunals to those publishers receiving data directly from them.
Content-specific Libraries and virtual databases
LIIs like AustLII have an increasing wealth and diversity of legal data, as explained above. This profusion of content leads to problems in precision of searches. One way commercial publishers have dealt with this, and added value to their content, is by creating subject-specific research facilities on topics such as environmental law, IP, criminal procedure etc. The challenge for free-access LIIs is that any such value-adding cannot involve the high costs of constant editorial intervention, nor the commissioning of subject-specific commentaries.
AustLII is attempting to deal with this by another variant of ‘data mining’: creating useful subject-specific 'Libraries' in subject-areas of Australian law such as indigenous law, taxation and industrial law, by means which are almost entirely automated once the initial parameters of the Library are defined. The Library is then automatically updated daily, from all content coming into any database on AustLII. The approach involves methods of identifying and isolating within databases of general content that which is on specified subjects, by largely automated and repeatable means, particularly the use of approximating searches. Methods of testing the searches used to construct Libraries are then undertaken.
So far, this method has been used to develop the Australian Taxation Law Library with ATAX and ATO support, the Indigenous Law Library, and the Privacy and Surveillance Law Library. An Australian Aviation Law Library is now under development.
How can AustLII use UGC & SNS?
As yet, there are few successful law-specific examples of services that rely upon user communities and wiki-like co-authoring of material. There is a range of legal content on generalist services such as Wikipedia and also some law-specific projects (eg Jurispedia) but there is considerably greater potential for exploiting such approaches. Other aspects of ‘crowd-sourcing’ of useful information, such as user ‘tagging’ of photographs and other documents (user-generated metadata), or the National Library of Australia’s use of user correction of text errors in scanned/OCR’d newspapers, are different but potentially valuable examples of utilizing user input.
AustLII is well placed to take advantage of these technologies as it has such large usage from a range of potential content contributors offering a variety of complementary perspectives, but has not yet sought to leverage the legal knowledge and experience of its user community. ARC funding has been sought, for a 2011-12 project by which AustLII will support user communities and enable user generated content. Given the authoritative nature of its legal content, there are considerable constraints on how AustLII can allow user interaction with that content. AustLII-user-generated-content (AUGC) will probably be located on an ‘AustLII community’ website with an address separate from AustLII. Potential projects may include a user-developed Australian legal dictionary, suggestion of additional citations to documents and citations unrecognised in documents (which will improve LawCite), and some forms of commentary. These projects will be seeded with existing material that can be automatically generated from AustLII content (including LawCite data). For example, the dictionary will include definitions from legislative sources and “words and phrases” from cases. AustLII content may also be framed for users to develop content around it without altering the framed content.
May websites now allow users to connect their content to social networking services (SNS), allowing users to develop delivery mechanisms for sharing notifications of content among communities of interest. The proposed project will consider various such mechanisms, ranging from email updates to Twitter feeds.
Acting globally ... building capacity for free access to law
How are these Australian developments then adopted by AustLII's international projects and collaborating LIIs? It is necessary to discuss this in the context of the overall aim of a global network of free access legal information.
Re-defining WorldLII: A global network of LIIs and a global LII?
WorldLII is the shared portal of the Free Access to Law Movement, which now has 34 members. But how global a portal is WorldLII? The following table shows which LIIs are the sources of its databases (and those on the other two portals operated by AustLII – CommonLII and AsianLII) as at the end of 2009. There were then 1,190 databases searchable via WorldLII at that time and there are approximately 200 more databases searchable via WorldLII by now (an annual count is done for AustLII’s Annual Report).
Table 1. Searchable databases in multi-LII systems (as at 19 November 2009).
The Declaration on Free Access to Law (2002) to which the members of the Free Access to Law Movement adhere, recognises both recognises ‘the primary role of local initiatives in free access publishing of their own national legal information’ and that ‘All legal information institutes are encouraged to participate in regional or global free access to law networks’.
So WorldLII has been seen primarily as a means of ‘networking’ or providing a common search mechanism for as many national (or broader) LIIs as wished to collaborate in relation to it. In practice. CommonLII is also primarily such a networking exercise, because (as seen from the above table) only 69 of its 755 databases had their home on the CommonLII server itself: the others were all located on other collaborating LIIs (including AustLII).
Is WorldLII yet anything like a comprehensive global free access research facility? No, it is not. It provides searches over databases from 165 countries at best, but really only 125 because about 40 on WorldLII only have constitutions. This is about half of the countries in the world. There is almost no content from Europe (UK and Ireland excepted), only federal case law from the USA (but 800,000 cases nevertheless), only GLIN summaries from Latin America, and nothing from the middle east or central Asia. Francophone North Africa is largely missing, though GLIN has some.
Free access providers are no worse than commercial legal publishers in this respect. The LexisNexis and WestLaw international platforms cover only a small number of countries, and Kluwer/CCH does not have a global platform. WorldLII covers many more countries than they do, but usually in less depth.
However, from the majority of these countries ‘missing’ from WorldLII free access content, already in digital form, is available in significant quantities from government websites. But unless a researcher is familiar with all of these sites – and that is a non-trivial task - doing comparative law research is nearly impossible.
Given sufficient resources, it is possible to republish legislation, case law and treaties from most of these government websites. Most countries’ copyright laws – unlike Australia – do not claim any form of government copyright in these materials, and often not in government-provided translations either. In some cases, the government website will place a robot exclusion against any automated copying of certain categories of content on the site, in effect prohibiting republication whatever copyright law has to say on the question. Well-behaved publishers (including AustLII, and (usually) Google) respect the robot exclusion, as they should, and ‘keep out’. But as a broad generalisation, most primary legal materials located on government websites around the world are available for republication by other publishers, without the necessity to seek a specific licence to republish from the relevant government or court.
This situation poses a key question for AustLII in its development of WorldLII, which is essentially a choice between two approaches to WorldLII’s development:
● (1) WorldLII solely as a portal for content from other collaborating LIIs (national like NZLII or regional like PacLII). This means that if a country/region has no LIIs, or none that wish to collaborate in WorldLII, then that part of the world has no content searchable via WorldLII.;
● (2) WorldLII as a combination of databases provided by other LIIs, and databases developed by AustLII on WorldLII from other countries where there is no collaborating LII. Such development of other databases would always of course be subject to the financial resources that AustLII could obtain to develop, and maintain, databases from other countries. But it might develop slowly in the direction of a free access LII with genuinely global coverage.
Ten strategies toward a global free access network & LII
The approach explicitly chosen by AustLII since 2009 (though implied in some of its earlier projects is a variant of (ii), the ‘global’ approach. It involves the following ten elements, which apply equally to AsianLII and CommonLII (the content of which contribute to WorldLII in any event):
(a) Assisting and encouraging development of new LIIs, particularly but not exclusively those willing to be involved in LII portals, wherever possible. Legal information is always best understood and managed locally. From the point of view of the operation of portal sites, long-term sustainability depends on local partners. In this sense, the long-term aim is still the ‘network’ model.
(b) Wherever possible, encouraging major existing free access sources of data to actively collaborate with WorldLII by making their data searchable via WorldLII and, where possible, for them to become members of the Free Access to Law Movement.
(c) Giving priority to adding databases to WorldLII (or AsianLII or CommonLII) where this is done in cooperation with local partners and has good prospects of eventually resulting in the development of a locally-operated LII.
(d) Developing common tools (such as the Sino search engine) and data sets (such as the LawCite database) that can be used by collaborating LIIs, both the improve their own services, and to make their systems more interconnected.
(e) Cooperating wherever possible with commercial publishers in mutually advantageous exchanges of texts and citation data, so as to broaden the scope and quality of the data in the free access domain.
(f) Finding large sources of republishable legal data which give global coverage of certain subject areas or types of documents, so that the content available through WorldLII steadily expands in relation to all countries, not only those within (a) and (b).
(g) Adding key databases from more significant countries that currently do not have their own collaborating LIIs, where permissible according to copyright law and robot exclusions.
(h) Adding databases in languages other than English, but with English language translations wherever possible so as to improve the comparative law value of the system.
(i) Adding databases which cannot be represented in single byte encodings (eg many Asian languages) only when AustLII’s Sino search engine has been developed to search those languages simultaneous with English and other languages.
(j) As a practical matter, to give priority to any of the above where necessary to complete a project for which AustLII has funding, and to seek funding for such projects, jointly with other partners where possible.
Depending on the partnerships developed and the funding obtained, WorldLII should be able to continue moving forward each year, further in the direction of a free access network with global coverage.
Expanding free access content toward global coverage
Here are some examples of the above approaches, from current or recent AustLII projects:
Expanding accessibility: A global search engine for law?
WorldLII is likely to retain a bias toward the English language, with as many databases as possible being obtained in English to assist WorldLII having as broad a scope as possible for comparative law research by English-speakers. However, many databases in other languages are being added and will be added in future, particularly as part of assisting development toward local LIIs in countries in which English is not the principal legal language. If this is to be done, it is most useful if the Sino search engine can search materials in that language, so that multi-lingual searches can be done. This is only a problem with languages that use double-byte encodings, as is the case with many Asian languages such as Chinese, Vietnamese and Korean. Sino has been developed so that it can search each of these languages, by a method which can be adapted for any language, but requires some customisation for each language. For example, AsianLII’s Chinese Language Databases page allows numerous databases in Chinese from the People’s Republic of China (PRC), Hong Kong SAR, Macau SAR and Taiwan all to be searched together, using both simplified and traditional Chinese.
At present, searching multi-lingual databases requires the user to know how to construct alternative search terms in multiple languages. There is considerable research going on, in Asia as well as in Europe, on the development of legal thesauri in multiple languages, to assist a search in one language being able to automatically find relevant documents in other languages as well. Much of this research is government-commissioned, and the outcomes are likely to be available for free access LIIs to use.
LawCite’s international dimension: Gluing LIIs together
LawCite is an international citator, recognising citations of cases, articles, and treaties in cases primarily from common law Australasia, the Pacific, Africa, South Asia, Singapore and Malaysia, Hong Kong, the UK, Ireland, the USA and Canada. LawCite is used by some of the collaborating LIIs whose data is used to develop it, as their own citator. AustLII encourages other collaborating LIIs to integrate it fully into their search result displays, as it is now used on AustLII. It is a shared resource. LawCite is now the primary means by which different LIIs are inter-connected, other than the fact that their content can be searched together via WorldLII.
LawCite has some citation information from civil law countries in Asia, but the amount of citation of other cases, or articles, is comparatively slight. Once a significant quantity of cases from European civil law countries is accessible through WorldLII, either through new collaborating LIIs in Europe, or databases developed on WorldLII, it will be possible to test just how much citation information there is in these cases, and therefore how useful LawCite is to all LIIs, not only those in common law jurisdictions.
Creating other means of access across LII content: Global law libraries
The same methods involving use of virtual databases that have been used on AustLII to create subject specific ‘Libraries’ such as the Australian Taxation Law Library are now being used to develop Libraries which draw their content from multiple countries and multiple collaborating LIIs.
For example the Commonwealth Criminal Law Library on CommonLII is built primarily from five virtual databases (case law, legislation, journal articles, treaties and law reform) with data drawn from all Commonwealth countries. The Privacy Law Library on WorldLII is similar but also contains many specialised databases from data protection authorities and specialised journals.
A virtual database has also been built for each country in the world, comprised of content concerning that country from law journal articles, treaties, and international court decisions. These ‘country virtual databases’ have been located on the pages on WorldLII, AsianLII and CommonLII for each country, and have provided valuable content particularly for countries for which few separate databases were held. However, they have been withdrawn temporarily until the problem of duplication of some search results when the whole portal is searched is resolved.
Conclusion: Where might AustLII ‘thinking globally’ lead?
In conclusion is may be worth summarising the different types of contributions by which AustLII assists in the globalising of free access to legal information. The LII portals that AustLII operates allow users to do comparative law research not otherwise available. Their value for this purpose is expanded by AustLII’s development of new databases on the portals themselves. The portals provide additional users for the LIIs who collaborate in their development, because search resulss send users back to the original LIIs from which the data came. Contributing LIIs can use the LawCite database as their citator, and can use it to improve their own data by adding parallel citations. AustLII does the development work to maintain and improve the LawCite database and its interface. AustLII’s open source search engine, Sino, is used by seven other LIIs, and AustLII keeps on improving it and making the improvements available to other LIIs. It is now progressively expanding the scope of the languages that Sino can search, which now include Chinese and other Asian languages, and these are available for other LIIs to use. AustLII provides technical assistance and advice to organisations starting new LIIs, and continues to provide this to some established LIIs. Finally the daily synchronisation of data from other LIIs to AustLII’s servers, for purposes of updating the portals and LawCite, also provides another back-up copy for each LII.
Where might all this lead? AustLII has taken the position that, in the long term, those who support free access to legal information should try to develop one or more global platforms or portals, so that comparative research with global coverage is made easier, or possible at all. There may be more than one platform, because one approach might be for major languages each to have their own separate ‘WorldLII-like’ portals. They could be operated by different providers, with interconnections between their data such as are being developed by LIIs at present. Alternatively, there might only be one global platform, with interfaces in different languages.
The major commercial legal publishers have not yet been successful in developing such global platforms. Their international portals still only have coverage of more than a small number of countries, even though their local operations in many countries hold much more data. However, in the long run we can assume they will develop more global coverage, as the globalisation of legal work and the interconnections of the world’s legal systems grow.
Free access legal information providers, through collaboration, are already doing as well as the commercial publishers in providing portals with multi-national coverage. AustLII is committed to expanding that global coverage, and is developing a variety of methods to do so. However, in the long run, the success of those initiatives will depend on the extent to which that objective and commitment is shared by other free-access providers around the world.
 Graham Greenleaf AM is Professor of Law at the University of New South Wales; Andrew Mowbray is Professor of Law at the University of Technology, Sydney; and Philip Chung is Lecturer in Law at the University of Technology, Sydney and Executive Director, AustLII. The authors are the Co-Directors of AustLII and therefore of CommonLII, WorldLII, AsianLII and LawCite. Thanks to the many staff members of AustLII who have worked on the developments described in this paper.
 Available at <http://www.austlii.edu.au/austlii/publications/2010/1.pdf>
 The Declaration on Free Access to Law (2002) <http://www.worldlii.org/worldlii/declaration/> says ‘Public legal information means legal information produced by public bodies that have a duty to produce law and make it public. It includes primary sources of law, such as legislation, case law and treaties, as well as various secondary (interpretative) public sources, such as reports on preparatory work and law reform, and resulting from boards of inquiry. It also includes legal documents created as a result of public funding.’
 See Point-in-Time Legislation Project at <http://portsea.austlii.edu.au/pit/>
 Australasian Legal Scholarship Library at http://www.austlii.edu.au/au/journals/s
 For example, Sydney Law Review 1953-, University of Tasmania Law Journal 1958-, Monash University Law Review 1974- , and University of New South Wales Law Journal 1975-.
 LawCite citator <http://www.worldlii.org/LawCite/>
 AustLII, BAILII, Cylaw, HKLII, IRLII, NZLII, PacLII, SAFLII, MalawiLII, ULII, AsianLII, CommonLII and WorldLII.
 Graham Greenleaf "Subject libraries in free access law services"  ALRS 19; in Helmut Rüßmann (Ed) Festschrift für Gerhard Käfer, pgs 75-94, republished in  ALRS 19 at <http://www.austlii.edu.au/au/journals/ALRS/2009/19.html>
 <http://www.austlii.edu.au/au/special/tax/> and see
 See Jurispedia <http: www.jurispedia.org>
 Australian Newspapers Digitisation Project <http://www.nla.gov.au/ndp/>
 The LII (Cornell) has had some success in such a development with its WEX user-generated legal dictionary, but it is of course for American law. See <http://topics.law.cornell.edu/wex> .
 http://www.worldlii.org/worldlii/declaration/; See above n 2.
 For some indications of the range of content available, and some of the difficulties in finding it, see Greenleaf G ‘Free access to legal information, LIIs, and the Free Access to Law Movement’, Chapter in Danner, R and Winterton, J (eds.) IALL International Handbook of Legal Information Management. Aldershot, Burlington VT: Ashgate, 2011
 Found in the robots.txt file at the root of the site; see A Protocol for Robot Exclusion
 <http://www.asianlii.org/chi/> for example, search for ‘"软件" or software or "軟件"’- the results 'By Database' shows 12 databases - first 5 from PRC, 6-8 from Macau, 9-12 from Hong Kong.