BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £5, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!
 |
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] | |
United Kingdom Journals |
||
|
You are here: BAILII >> Databases >> United Kingdom Journals >> JILT 1997 (2) - AustLII Paper 4 URL: http://www.bailii.org/uk/other/journals/JILT/1997/greenleaf_2(austlii4).html |
||
[New search] [Help]
We always say that AustLII is 'approximately 50% politics', by which we mean that innovative techniques of computerising legal information count for little unless you can obtain the data that you wish to publish in some useable form, permission to publish it, and the necessary funds to continue doing so. When you propose to provide free public access, all of these elements require the creation of a constituency which will support this goal, or at least reluctantly acquiesce.
In this paper, we will give an informal outline of various aspects of 'the politics of AustLII', and attempt to answer some of these questions:
The computerisation of legal information in Australia has been a very political matter since its inception. It had two very different starting points.
Following a 1973 initiative of then Attorney-General Lionel Murphy, the Commonwealth Attorney-General's Department established the SCALE system in 1977, using the STATUS retrieval software to search Commonwealth legislation (1977) and High Court decisions (1980) See Greenleaf et al (1988, Ch 4) for a history of developments to 1988, from where the following paragraphs are derived.. Other Commonwealth and A.C.T. databases were added slowly, but the original plans to add State and Territory materials were overtaken by events for a time. Another Labour Attorney-General, Gareth Evans, decided that Commonwealth materials would not be allowed go on any system other than SCALE.
SCALE used much the same version of the STATUS software for twenty years until it migrated to the Internet as SCALEplus. It never received a great deal of use outside the public sector and some University law schools, and so did relatively little to put access to the law in public hands. However, it performed the inestimable public service of collecting Commonwealth and A.C.T. legislation and case law (and some State and Territory materials) in computerised form (and a very consistent one at that), and keeping it in public hands, for those twenty years until more sophisticated delivery systems could deliver what it had long only promised.
When AustLII (and eventually SCALEplus) arrived with new technologies, the availability of the 'old' SCALE's data only required a political decision in order to make free public access a reality. Even before those events, the availability of SCALE data to the creators of the DiskROM CD-ROMs in the early 1990s had created the first convincing demonstration of the real potential of computerised legal information - and showed that the monopolistic alternative to which we now turn, was a dead end.
It is easy to forget that throughout the 1980s all States and Territories (except Queensland, which did nothing effective at all) used their claimed control over copyright in legislation and case law [23] to grant the company originally called CLIRS (Computerised Legal Information Retrieval System), (owned by Computer Power Pty Ltd) a monopoly over the computerisation of this information for varying periods of years extending into the early nineties.
The monopolistic approach of the '80s was compounded by the 'STATUS Standard' by which the Standing Committee of Attorneys-General (SCAG) decided that no-one would be allowed to tender to computerise any Australian cases or statutes unless they agreed to provide retrieval features identical to those of STATUS, in addition to a list of other requirements. So the CLIRS/SCALE oligopoly was created.
After complex legal and commercial disputes the CLIRS system changed owners numerous times, changed its name to Info-One, and was a considerable commercial failure, redeemed somewhat in its final stages by a successful LawPac CD-ROM series of case-law. The CLIRS/Info-One decade could be summarised by saying that the data was very extensive, access was unacceptably expensive (at one point ludicrously so), the legal profession largely hated or ignored it, and few ever used it (except some University law schools with free accounts). The system probably never had more than ten simultaneous commercial users - AustLII already regularly receives nearly 200.
Finally, in 1995 Info-One was sold to Butterworths, and has now migrated to the Internet as a substantial part of the case-law content of Butterworths Online. On the one hand we are fortunate that extensive case law databases are at least available to those who can afford to pay for them - and in the hands of a publisher with expertise in this area.
On the other hand, it means that there was a whole decade when the computerisation of legal information was illegal by anyone other than the CLIRS/Info-One monopoly, and this is a monopoly benefit that Butterworths has now acquired. This is Butterworths' good fortune, but is the result of flawed public policy (over which they had no control). In addition, during this period the development of competitive - or even slightly innovative - approaches to legal information systems was largely frozen in Australia.
A small light at the end of the tunnel was created by the NSW government's decision in 1994 to grant a general public licence in relation to the reproduction of legislation. It also provided copies of part or all of the NSW legislation in computerised form, updated regularly, with no restrictions on the commercial re-use of the information. It even sold legislation on demand on floppy disks over the counter. This innovative approach helped create a proliferation of disk-based products which added value in various ways to NSW legislation. More important, in doing so, the NSW government demonstrated that different public policies were possible, and its approach illustrated most of the elements that AustLII has since argued are part of a sound policy for public legal information.
This was the environment that made AustLII necessary: no effective or affordable public access to legal information; a lack of competition in the provision of commercial products; and such products as did exist were largely the recycling of primary legal materials with little value-adding but very high prices. Electronic legal publishing equalled on-line services with few users, and it was easier to make a modest profit by selling some hundreds of copies of a CD-ROM of primary materials at relatively high prices than to provide pervasive commercial access to the profession, let alone the public.
Only two and a half years ago, at the beginning of 1995, there was no reason to expect that any free public access to legislation or case law whatsoever would ever be possible in Australia. The philosophy of 'user pays' predominated in government thinking. There was no guarantee that SCALE would reach the Internet as a free service, as discussed below. No legislature had published its own legislation free on the Internet as the Northern Territory has now done or Victoria will do soon. No court or tribunal in Australia had made its decisions available free, and many were still wedded to the notion of subsidising the court through photocopy sales. It was even difficult to obtain small quantities of data in computerised form for academic research or teaching purposes.
As has been described in other papers, we (Mowbray and Greenleaf) had already developed software and an approach to computerising legal information through our 'DataLex Project' from 1985-94. We decided to try to use the Internet to end the situation described above. We obtained an academic grant to create a 'research infrastructure', sufficient to purchase a server and to employ a staff member (King) for a year or so [24] . The application was supported by the Deans of all Australian law schools. Our stated aims in obtaining the funding were:
However, we were conscious that - with luck - this would give an opportunity to pursue a broader agenda:
These aims have been partly satisfied by AustLII's history to date, but nothing can be taken for granted, as the rest of this paper will illustrate.
Our most urgent problem in getting AustLII started was 'political' in the broad sense we use it here: we had no large-scale data in computerised form, nor permission to publish it if we obtained it. Fortunately, we had some experience in obtaining licences from the Australian Government Publishing Service (AGPS) to publish small collections of Commonwealth legislation on disk as part of our previous 'DataLex' work, and so we sent in a standard licence request to AGPS to publish Commonwealth legislation (generally) on the Internet. SCALE gave us a copy of the whole Commonwealth legislation to test the scaling-up of our mark-up techniques. To our surprise, the AGPS licence arrived. Holding both the data and a licence to publish on the Internet in our hands, we did so - perhaps, in our enthusiasm, slightly in advance of finalising negotiations. We were receiving 200 to 300 users a day within a week, most from the Commonwealth government, and the numbers snowballed from there. Fortunately, no one in Canberra then felt it was politic to tell us to take the data down.
AustLII's publishing of Commonwealth legislation came at a rather sensitive time, as the Commonwealth was still considering making the system that eventually became SCALEplus a 'user pays' service of some kind. This option was obviously less viable in light of what AustLII was now very publicly willing and able to do as a free service, and it was eventually shelved.
Having made significant progress with legislation, we immediately shifted focus to case law, and found a sympathetic reception from the Marshal of the High Court, who requested SCALE to provide us with a copy of the full text of the Court's decisions back to 1947. We published the decisions, with extensive hypertext links to Commonwealth legislation. Andrew Mowbray then wrote the SINO search engine, and the general model for AustLII's operation was in place.
In our view, free Internet access to a huge database of High Court decisions sent the right signal to every court and tribunal in Australia: that court decisions should be available to the public, and the Internet was an appropriate medium to achieve this. By 'leading from the top', our co-operative relationship with the Court has continued to demonstrate what can be achieved in Internet publishing of case law. We now publish decisions of the Court within a day of them being handed down, the bulletin of the Court's special leave and reserved decisions, and selected transcripts. AustLII also hosts the Court's home pages.
The next milestones in free access to law came from New South Wales. The New South Wales Law Foundation agreed to provide funding to AustLII to provide Internet access to Australia-wide primary legal materials. The Law Foundation's own 'Foundation Law' Internet service (which originally ran on AustLII's computers) encouraged the NSW legal profession to migrate to the Internet. The NSW Attorney-General agreed to provide complete NSW legislation. The Supreme Court, Land and Environment Court and Industrial Court all agreed that we could publish their decisions, with assistance from the Law Foundation and the NSW Attorney-General's Department.
Just as important, Law Foundation Chairman Justice Vince Bruce was instrumental in negotiations with the Commonwealth Attorney-General's Department which resulted in a general agreement that AustLII could publish databases on SCALE where the court or tribunal concerned gave its consent. This resulted in another 15 large databases on AustLII by early 1996. From that point on in late 1995 our relationship with SCALE and the Commonwealth Attorney-General's Department matured into the very extensive and co-operative one that has remained ever since. Our relationships with Commonwealth A-G and the NSW Law Foundation have been two cornerstones of AustLII's success, and the freeing of legal information in Australia.
The fact that AustLII now holds a copy of most of SCALE's data is significant beyond the fact that it now gives users a choice between two free Internet services. It helps to ensure that the data will remain on a free public access facility. In an era of outsourcing of the Commonwealth government's IT, privatisation of the Attorney-General's legal practice, and public tenders for functions of AGPS, there can be no certainty about free provision of legal information by government agencies.
Funding from public interest organisations like the Law Foundation, and academic sources, also played a political role. It enabled us to take the 'high moral ground' with the 'data holders' of public legal information, along the lines of 'we are willing to do all the hard work to provide this information free to the public, and organisation X is willing to fund it, if only you will licence it to us / release it into the public domain'. This approach is effective, but it depends on the necessary funding from public sources being maintained. That is discussed below.
AustLII's existence is based on the belief that it is in the public interest that public authorities should aim to maximise access to the 'public legal information' that they control. By 'public legal information' we include primary legal materials (legislation, case law, treaties, awards and administrative decisions); and at least those secondary legal materials created with public funding, under a duty to report, and with a purpose of public access (law reform reports, justice statistics etc.).
We argue that public authorities (such as courts, legislatures and Royal Commissions, and law reform commissions) should aim to maximise access to this 'public legal information' by removing all unnecessary impediments to its wide dissemination. In the first paper published about AustLII in 1995 ( Greenleaf et al; 1995a ), we argued that the effective provision of public legal information depended on public authorities (the sources of the data) providing that data according to six principles, which we set out unchanged in the sub-headings below. Our experience since then, reflected in the comments that follow, has reinforced our view that they are sound principles .
It is worth stressing that the six principles below have little to do with any special provision of data to AustLII: adoption of these principles by public authorities is for the benefit of all legal publishers, commercial and non-commercial alike.
AustLII's advocacy of these principles takes many forms: negotiations with data providers and governments, conference papers, public campaigns against recalcitrants , and (probably most important) the force of example provided by the existence of AustLII and the extent of use that it receives.
Each of these six principles should be read as 'public authorities should provide public legal information ...':
(i) ... in a completed form, including such additional information as is best provided at source';
(ii) ... in an authoritative form, including acceptable citations and numbering - medium-neutral and vendor-neutral citation;
(iii) ... in a form facilitating dissemination;
(iv) ... on a marginal-cost-recovery basis equally to anyone who wishes to obtain it;
(v) ... with no restriction on re-use for any purpose, and no licence fees ;
(vi) ... but while still preserving a copy in the care of the public authority.
It is necessary for cost-effective and reliable publishing of judgments by more than one publisher, that courts and tribunals accept the responsibility of producing one definitive 'unreported' version of a judgment. At least when it comes to computerised copies, the practice of courts is very variable: many are excellent, but some courts still have no effective methods of collecting all judgments centrally in computerised form, and sometimes supply them with judgments in the same case on different disks.
Other parties (such as publishers) should not have any role in assisting courts 'tidy up' their judgments prior to the official release by the court, because of the risk of copyright claims being asserted by them. AustLII's data has been affected by one claim by a legal publisher to copyright in aspects of judgements which made it impossible for us to disentangle the 'publisher's additions' from what would have been in the original judgment. However, this has not been a serious problem as yet.
The additional information provided should include catchwords nominated by the judge (as recommended by the Australian Institute of Judicial Administration in Olsson (1992)), and the consolidation of amending legislation by Parliamentary Counsel (a process not yet completed in some Australian jurisdictions). The inclusion of catchwords in judgments is still only sporadic but increasing.
It should be in an authoritative form, in the sense that includes citations and numbering such that it can be cited to any court in an acceptable way. This is not a problem with legislation, but is with case citations, discussed further below.
Where official bodies have created this data in computerised form as a by-product of their normal work, it is in the public interest that it should be available in that form. Two years ago, provision on disk in ASCII would have been the best that was obtainable in many cases. Now, our experience is that many courts and tribunals have no difficulty in e-mailing their judgments to multiple recipients (usually publishers) on a daily or weekly basis, sometimes in enhanced formats such as RTF.
Public policy should support maximising public access to the law. Its dissemination should not be regarded as a 'profit centre' supporting other aspects of the operation of the judicial system. The outputs of the legislative, judicial and law reform processes should not be subject to 'user pays' criteria, because maximum public access to this information is necessary for the proper operation of the judicial system and the democratic system.
Public policy should support the maximum dissemination of the law (including the creation of value-added products for resale), and in the forms to make it most understandable. The methods by which legal data is best disseminated are still unsettled and changing rapidly, and there are markets for the same source information with different features and at different prices. Fostering competition in the provision of different types of legal products seems to be the only way to meet the public interest.
Offices of Parliamentary Counsel need to preserve computerised copies of legislation archives, and courts and tribunals should keep computerised back-sets of their own decisions, irrespective of how widely they distribute them at the time of creation. Back sets need to be available from public authorities, so that new entrants into the computerisation of law are not effectively excluded by all electronic copies of the data being held in private hands. With legislation, archiving of copies of amended sections over time will enable the creation of far more valuable computerised legislation, able to provide the law as it stood at a specific date.
Courts and legislatures have been using computers to produce documents for many years, but some have made little effort to keep computerised archives in public hands. More State and Territory materials would have remained in public hands despite the CLIRS/Info-One monopoly, if this approach had been adopted more extensively.
This agenda is not unduly idealistic. The New South Wales Government's approach to the dissemination of legislation is a model implementation of at least elements (i) - (v), and the approach taken in many other jurisdictions is now fairly close (although licence fees of a percentage of income are charged in some cases). Many courts and tribunals also adopt principles (iii) - (v), but others still require very high licence fees.
AustLII is [25] ($110,000), and by its two host Universities ($25,000 each). In 1995/96 the Law Foundation of New South Wales made its first grant to AustLII (approximately $135,000 for 9 months), to support the inclusion of Australia-wide primary legal materials, and has renewed the grant at similar rate for the two successive financial years. Having been successful in gaining Australian Research Council competitive funding, and grants from other organisations for 1997, AustLII already has committed funding exceeding $800,000 for the three financial years 1997-2000 [26] . Taking into account reasonable expectations of other continuing annual funding, we are very confident that AustLII will receive approximately $400,000 per year for the next three financial years to July 2000. This funding is sufficient to maintain AustLII's current staff levels, and its equipment, but no more than that.
The point of summarising these matters is to demonstrate that AustLII will be able to at least carry out its core activities (including research and collection expansion on current projects) into the next century. In the short term AustLII is therefore 'sustainable' - it is not likely to disappear overnight.
We believe that AustLII is sustainable beyond the short term. One of the original ideas behind AustLII was that the provision of modest funding by a range of significant stakeholders in access to legal information could provide a long-term funding basis. The 'stakeholder model' is one that sees corporate representatives of classes of users, and data providers who wish to publish their data widely, being willing to contribute a share of the costs of running a free access service which includes the data in which they have a special interest, recognising that the system will also be used by 'free riders' who do not fall within any of the target groups of users. One of the attractions is that many stakeholders are interested in certain types of core data such as legislation or decisions of appeal courts (or at least parts thereof), and so each stakeholder only pays a small portion of the costs of making this data available. No doubt this is crude economics, but it makes sense to many 'stakeholders'. Of course, AustLII is not the only organisation that could offer to provide access to data on this model.
We envisage that these stakeholders will, over time, include a mix of organisations falling into at least these categories:
•business, trade union and professional organisations (including the legal profession);
•tribunals that have a strong interest in having their decisions more widely known, particularly small ones whose decisions are not often published by commercial publishers;
•government agencies with a strong interest in wider availability of a class of information (for example, the Department of Foreign Affairs and Trade in relation to treaties);
•community interest organisations (such as the Law Foundation);
•the academic legal community, for research and teaching purposes.
AustLII has to date been able to publish the decisions of every court or tribunal that wishes to have its decisions published. We request courts and tribunals to consider making a small contribution toward AustLII's overall running costs, but have not yet turned case law away because funding was not available.
It is much more difficult to manage secondary legal materials unless specific project funding is available. All AustLII 'core' operations must now be supported from project-specific funding, so all such projects must carry a portion of AustLII's core costs (management and infrastructure).
One alternative funding model now used by many web resources is advertising, but AustLII has no intention of adopting this model. Why should it be necessary to be bombarded by advertisements in order to do legal research on a University-based facility? Many Internet search engines now determine the advertisements that users see by the search terms they enter. It seems no coincidence that when you search one popular Australian search engine for 'sex discrimination', you get back lurid advertisements for sex aids. We have no interest in running AustLII on the basis of consumer manipulation.
Funding considerations are not complete without reiterating that we believe AustLII serves a role in prompting commercial legal publishers to add more value to the primary legal materials they publish, and to keep their prices moderate. Similarly, AustLII must cause governments and courts to think twice about charging for electronic access to public legal documents. In fact, whether AustLII has played a significant role in this or not, Australia is now better served by both commercial and free-to-air electronic legal resources than any county we know. Others think we are 'the lucky country' in this respect [27] - but sometimes you have to make your own luck.
A very encouraging example of cooperation emerging between Australian commercial and non-commercial legal publishers is occurring in the area of reforms to citation methods to accommodate the demands of electronic publishing. AustLII made its initial proposals on this topic at the Australian Institute of Judicial Administration Annual Conference in Wellington, New Zealand in September 1996 ( De la Fosse and Finaly, 1997 ), and in submission to the Council of Chief Justices in January 1997. Butterworths made its views public in early 1997. The Council of Chief Justices' Electronic Appeals Project ( De La Fosse and Finaly, 1997 ) is to cover the question of court-designated citations. The Councils of Law Reporting have also discussed the matter.
In June 1997 representatives of Butterworths, LBC Information Services, CCH Australia, Pink Ribbon (the citator publishers) and AustLII held the first of what promises to be a series of very constructive meetings. Our interpretation (other participants may have a different perspective) of the discussions is that there is a general acceptance of the inevitability of a court-designated method of citation emerging, in parallel with new forms of publisher-designated citations. There also seemed to be general agreement emerging that both forms of citation would have to accommodate four elements: year (of decision or publication), case number (court-issued or publisher-issued), court or series abbreviation, and paragraph number (uniform across citations, and stemming from court's numbering). However, many issues of detail arise, and will take considerable discussion to resolve, between publishers, the courts, court reporting bodies and other interested parties.
The following parts of this section of the paper reproduce the position that AustLII submitted to the Council of Chief Justices, and to the recent publishers meeting, as a summary of our thinking so far on this issue. The details of this issue are complex (even though the basic principles are clear), so we expect that our initial views will be modified as all the interested parties explore the issue.
One of the most pressing needs in the development of a policy for public legal information is for a method of citing the decisions of courts and tribunals that is independent of any particular publisher or particular medium of publication. These issues including details of US and Canadian developments are discussed at greater length in Greenleaf et al (1996). Such a 'medium-neutral' and 'vendor-neutral' citation would be designated by the court that made the decision. A court-designated case citation standard would have many advantages: writers would be able to cite other decisions without making assumptions about the particular publications available to their readers; readers would be able to find decisions cited in whatever 'court reports' they have at hand (print or electronic); the creation of automated hypertext links and searches would be enhanced greatly; potential copyright difficulties in citation use would be avoided; and the official citation for a case will be known as soon as a court or tribunal releases it. Most print publishers would also continue to use their own 'parallel' citations, to indicate their own selectivity, ordering and print volume location.
A citation standard can be put together in various ways using only court-designated citation elements, and that is susceptible to automated recognition and processing. The simplest approach, and the one that we consider is most desirable, would have three elements.
The basis of any such citation must be an authoritative unique decision number allocated by the court or tribunal, preferably with the number series recommencing each calendar year. Decision numbers should be continuous, and the sequence unalterable. Each court should only use one sequence.
In order to distinguish between courts, a new system of unique court descriptors needs to be developed - standard abbreviations for each court. In some senses, this is similar to the existing system for abbreviating the names of series of law reports (e.g. 'CLR', 'ALJR'), but is non-proprietary and court-specific. Any naming scheme should preferably not be limited (at least potentially) to indicating Australian jurisdictions, but be capable of application to any other jurisdiction and still give a world-wide unique identifier.
There appears to be developing agreement, at least in North America, that continuous paragraph numbering across a decision is the appropriate method of internally referring to parts of judgments ('pinpoint citation'). Paragraph numbering is needed because page numbering is print-medium-specific (and specific to particular publishers), irrelevant to computerised judgments, and too coarse. Some Australian courts already use paragraph numbering within individual judgments. However, as both Canadian and US reports have recognised, paragraph numbering needs to be made more absolute and should be continuous across all judgements in a decision.
Our interim conclusion, pending further research and experiment, is that a vendor and medium neutral citation standard something like the following is desirable:
[1996] 194 HCA 23
The brackets around the date are possibly unnecessary, but they make it appear more familiar as a case citation (at least in Anglo-Australian publishing), and they may assist automated recognition of citation commencement. The decision number allocated by the Court ('194') follows, then the Court abbreviation, then the paragraph number. We have concluded that, if a citation of the decision as a whole is intended, no paragraph number should be given [28] , so the citation for the above decision as a whole would be simply '[1996] 194 HCA' .
On an experimental basis, recent decisions in some AustLII caselaw databases are now given such a citation, where we have court-designated or tribunal-designated decision numbers available to us. This can be considered an 'AustLII citation' in the same fashion as commercial legal publishers provide their own citations for cases, but differs in that anyone else is welcome to adopt it.
The initial approach we have taken to court designators is to use the normal abbreviation for a court or tribunal (which is usually obvious), followed by 'A' for 'Australia' where a Commonwealth court or tribunal is concerned (e.g. HCA, FCA, FamCA, IRCA). Where a State or Territory court or tribunal is concerned, we are considering placing the standard abbreviation for the jurisdiction before the court or tribunal abbreviation, simply because this feels more consistent with previous citation patterns (e.g. NSWSC, VICAAT).
Development of the most appropriate Australia-wide abbreviation scheme, particularly if possible international compatibility is considered, will require discussion, research and experiment. Any decision on a permanent standard series of court and tribunal abbreviations would, in our view, be a matter principally for the courts and tribunals concerned. Our experimental court abbreviations can be readily altered across the whole of the AustLII databases.
For example, the decision described as Industrial Relations Commission Decision 2704/1996 is also given the AustLII citation '[1996] 2704 IRCommA', reflecting AustLII's file naming conventions, which for this case results in the URL http://www.austlii.edu.au/au/cases/cth/IRCommA/1996/2704.html
Similarly, for the recent Federal court decision listed in AustLII as Terrence Golby & Anor v Commonwealth Bank of Australia [1996] 1136 FCA (24 December 1996), the citation '[1996] 1136 FCA' has been provided by AustLII. The citation reflects the URL [29] http://www.austlii.edu.au/au/cases/cth/federal_ct/1996/1136.html
These AustLII citations will provide a convenient method for anyone to cite a case appearing in an AustLII database without having to recite a whole URL, as the case can be easily found on AustLII using the citation. It will also allow the creation of automated links to these cases. Experimental use of this approach to citation by AustLII will give us some experience in the practicalities of alternative citation methods, and may also help to encourage courts and tribunals to take an interest in the development of a uniform system of vendor neutral and medium neutral citation. Before any uniform system is adopted finally, there will obviously need to be considerable consultations with, and perhaps discussions between, courts, tribunals, and potential publishers and users of their decisions.
Publication of decisions on the Internet raises some difficult issues, one of which is the privacy of those whose personal affairs are reported in decisions. These decisions are now available to wider scrutiny than printed law reports or CD-ROMs, due to the level of public access to the Internet, and free availability.
In May 1996 AustLII found itself the centre of media controversy concerning its provision of decisions of the Family Court of Australia, with allegations of invasions of privacy and possible breach of publication restrictions in the Family Law Act 1975 (Cth) s221. The decisions were provided to AustLII by the Family Court via the Commonwealth Attorney-General's Department's SCALE system. AustLII decided to suspend the Family Court database until these issues were resolved, and the Commonwealth Attorney-General's Department did likewise on SCALE. On 19 June 1996 the Commonwealth Attorney-General and Minister for Justice, Darryl Williams QC, released a News Release headed 'Internet Material Cleared' [30] , stating that the publication on AustLII was not in breach of any privacy laws or the Family Law Act, and their availability was supported by the Family Court and the Family Law Council. The database has been restored to AustLII.
From AustLII's point of view, the episode is significant in a number of respects, which are of general relevance to other courts and tribunals. It reinforced our conviction that it must be the responsibility of the public bodies providing data containing personal information to make decisions concerning the appropriate balance between privacy interests and the public interest in publication. Publishers like AustLII cannot edit or censor the content of court decisions [31] . However, it is appropriate for courts and other public bodies to re-assess their publication practices concerning questions such as anonymisation, in light of the different accessibility of materials via Internet, as has been done by the Family Court in this instance. This also ensures that the decisions remain the same, irrespective of the medium of publication.
The initial apparent criticism of AustLII from some sections of the media caused a storm of e-mail from AustLII's supporters, and within a few days a number of newspapers and magazines had published editorials defending AustLII's role in publishing these decisions . Since then, there has been little criticism of the continuing publication, in the press or elsewhere.
AustLII has a responsibility not to encourage inappropriate use of the databases, so (for example) we exclude robots from indexing any of our case-law databases [32] , so that searches over general Internet search engines cannot inadvertently reveal that a person is involved (say) in Family Law proceedings. When we detect robots attempting to index our case law, we exclude them from access. If we became aware of other inappropriate access, we could also take steps to exclude it. In short, in order for anyone to find case law on AustLII, they have to come to AustLII - a rather boring looking non-sensational legal database - and search for it there.
