=Paper=
{{Paper
|id=Vol-2309/08
|storemode=property
|title=Spent Convictions and the Architecture for Establishing Legal Semantic Workflows
|pdfUrl=https://ceur-ws.org/Vol-2309/08.pdf
|volume=Vol-2309
|authors=Pompeu Casanovas,Louis de Koker,Markus Stumptner,Wolfgang Mayer,Jeffrey Barnes,Stammers Mira
|dblpUrl=https://dblp.org/rec/conf/jurix/CasanovasKSMBM18
}}
==Spent Convictions and the Architecture for Establishing Legal Semantic Workflows ==
https://ceur-ws.org/Vol-2309/08.pdf
Spent Convictions and the Architecture for
Establishing Legal Semantic Workflows
Pompeu CASANOVAS1 2,3,4, Louis DE KOKER 2,3, Markus STUMPTNER1, Wolfgang
MAYER1, Jeffrey BARNES2,3, Mira STAMMERS2,3
1
University of South Australia, Adelaide
2
La Trobe Law School, La Trobe University, Melbourne, Australia
3
Data to Decisions Cooperative Research Centre
4
Autonomous University of Barcelona (IDT), Spain
Abstract. Operating within the Data to Decision Cooperative Research Centre
(D2D CRC), the authors are currently involved in the Integrated Law Enforcement
program and the Compliance through Design project. These have the goal of
developing a federated data platform for law enforcement agencies that will enable
the execution of integrated analytics on data accessed from different external and
internal sources, thereby providing effective support to an investigator or analyst
working to evaluate evidence and manage lines of inquiries in an investigation.
Technical solutions should also operate ethically, in compliance with the law and
subject to good governance principles. This paper is focused on the Australian spent
convictions scheme, which provide use cases to test the platform.
Keywords. Legal natural language processing of legal texts, law enforcement
investigation management, spent convictions, Compliance through Design
1. Introduction
This paper presents ongoing research of the Australian government-funded Data to
Decisions Cooperative Research Centre (D2D CRC). 2 It focuses on specific spent
convictions use cases selected by the Australian Criminal Intelligence Commission
(ACIC) to produce a proof of concept on Compliance through Design (CtD) modelling.
This will be developed jointly with Guido Governatori (Data61, LegalRuleML), Mustafa
Hashmi (survey on CtD) and Víctor Rodríguez-Doncel (Polytechnic University of
Madrid, Natural Language Processing).
We have introduced the subject in previous works [1] [2]. This paper, which aims to
share further insights and research, is structured as follows: The first section outlines the
Australian spent convictions scheme and describes some features and conceptual
problems. The second section offers an overview of the platform, legal information
workflows, compliance services and the hub of knowledge. The closing section contains
1 Corresponding Author. P.CasanovasRomeu@latrobe.edu.au
2 http://www.d2dcrc.com.au/
85
some trends for our immediate work on compliance and the spent convictions regulatory
model.
2. Australian spent conviction schemes
Generally, information about a criminal conviction in a court of law will be recorded and
the criminal conviction remains part of a person’s record unless the conviction is
overturned, quashed or annulled. Not allowing a person to escape the burden of
convictions in the past was deemed unfair by Australian authorities and as impacting
negatively on the rehabilitation of criminals [3]. As a result, Australian jurisdictions, like
many other countries [4], introduced spent conviction schemes, regulated by different
laws and rules in the different Australian States and Territories and at a Commonwealth
(federal) level.
In general, a “spent conviction” is a conviction which becomes hidden from public
view after a set period of time. Once a conviction has become spent, the convicted person
may generally legally answer “No” when asked whether he or she was previously
convicted for that offence.
A blanket deletion of criminal records would cause problems as the conviction
information may have relevance later or in specific contexts. As a result, the information
remains on the record but shielded from publication or use, except in a small number of
defined cases where authorities or other interested parties would need access to the
complete criminal record of a person. Australia has, for example, a Working With
Children scheme that requires all youth workers to apply for and undergo a Working
With Children check before they are allowed to work with young people. 3 As part of the
background checks on an applicant, access to spent conviction data is required to ensure
that someone with historic child molestation convictions are not approved under this
scheme without appropriate consideration of the facts [14] [15].
The Commonwealth and each State and Territory have their own spent conviction
legislation [5] [6] [7] [8] [9] [10]. The exception is Victoria that regulates general spent
convictions through a Victoria Police information release policy rather than a law [13]
[15].4 This means in practice that a high degree of discretion is exercised in Victoria. 5
3 Working With Children Check (WWCC) is a background check requirement, assessing the
criminal record of those working or volunteering in child-related work. See
https://aifs.gov.au/cfca/publications/pre-employment-screening-working-children-checks-and-
police-checks.
4 Victoria Police Information Release Policy (September 2017). See also the specific scheme for
historical homosexual convictions (expungement) in Part 8 of the Sentencing Act 1991 (Vic).
5 “Victoria is the only Australian jurisdiction without legislation that provides for convictions to
be spent. Instead, the information on a person’s criminal record is governed by a Victoria Police
information release policy. Under this policy, if an adult has been found guilty of an offence within
the past 10 years, Victoria Police will disclose all prior findings of guilt as part of a criminal history
check. This means that any crimes that a person has been found guilty of, even where that person
did not receive a conviction, will still show up on their record. Victoria Police will also release
information on pending charges where a person has not yet been found guilty.” [13].
86
A significant implementational challenge lies in the differences in the rules that
apply in the spent convictions schemes of the States, Territories and the Commonwealth.
For example, South Australia excludes serious sex offences from becoming
automatically spent and their Spent Convictions Act 2009 have detailed provisions
distinguishing between different types of sex offences. 6 Under the Commonwealth
scheme however the conviction of a person convicted of a federal sex offence becomes
spent if the person was not sentenced to imprisonment for the offence, or was not
sentenced to imprisonment for the offence for more than 30 months, and the prescribed
period after conviction has ended. 7 Most jurisdictions use whether a term of
imprisonment was imposed and the length of any such term as proxies for the seriousness
of the offence. The terms of imprisonment applied, however, differ. At the
Commonwealth level it is 30 months while, for example, in New South Wales, it is 6
months or less, subject to a number of exceptions. 8 A prescribed crime-free period –
generally 10 years for adult offenders - must have expired after the conviction, though
in the Australian Capital Territory that period only commences after the person’s release
from prison.9
Even the ambits of the schemes differ. In terms of their information release policy,
Victoria Police releases criminal history information on the basis of findings of guilt as
well as details of matters currently under investigation or awaiting court hearing. The
spent conviction scheme of Western Australia, however, focuses on convictions and not
on findings of guilt. “Conviction” is defined as a conviction incurred by a natural person
for an offence against the law of this State or of a foreign country” by section 3 of the
Spent Convictions Act 1988 (WA). The Commonwealth scheme, on the other hand, is
broader. According to section 85ZM of the Crimes Act 1914 (Cth) a person shall be taken
to have been convicted of an offence if:
(a) the person has been convicted, whether summarily or on indictment, of the
offence;
(b) the person has been charged with, and found guilty of, the offence but
discharged without conviction; or
(c) the person has not been found guilty of the offence, but a court has taken it
into account in passing sentence on the person for another offence.
In Victoria, only the first two of the three parts of the Commonwealth definition would
be covered by its spent conviction scheme while only the first part (an actual conviction)
would be covered by the Western Australian scheme.
In practice, tools such as comparative tables and flow charts, are essential to navigate
the complexity of the Australian spent convictions landscape [12] [14].
6 See s 3, 5(2) and 8A of the Spent Convictions Act 2009 (SA). See similarly s 7(1) and (4) of the
Criminal Records Act 1991 (NSW) for a list of sexual offences that are excluded from the spent
conviction scheme of New South Wales.
7 S 85ZM(2)(b) of the Crimes Act 1914 (Cth).
8 S 7(1)(a) of the Criminal Records Act 1991 (NSW).
9 S 13(2)(c) of the Spent Convictions Act 2000 (ACT).
87
2.1. National Police Checking Service
Exchanges of criminal records data among the jurisdictions in Australia are coordinated
by and through the Australian Criminal Intelligence Commission (ACIC). It manages the
processes and provides the system through which Australian police agencies and
accredited bodies submit nationally coordinated criminal history checks.
The ACIC operates the National Police Checking Service that assists organisations
to screen and make informed decisions for example about prospective employees and
volunteers, visa and citizenship applications and work-related due diligence relating to
national security. The service is used by 244 accredited agencies and bodies. During the
period 2016–17 4.75 million checks were processed, and 1.42 million checks were
referred to police agencies for further assessment to determine whether the information
may be disclosed in accordance with their spent convictions legislation and/or
information release policies [11]. The extensive number of checks referred to police
agencies is directly linked to the complexity of the regime and inconsistencies among
the different jurisdictional schemes, discussed in 2 above [11].
2.2. Automated solution desirable
Figure 1 shows a decision flow chart for the Commonwealth spent convictions scheme
produced by the Office of the Australian Information Commissioner to guide decisions
of individuals.10 The flow chart, which embodies an interpretation of the law but is not a
legal instrument in its own right, indicates the types of key determinations required to
reach a correct conclusion on the reach of the Commonwealth scheme.
It is clear that an automated solution would be ideal to support compliance with
spent conviction rules. The volume and complexity cannot be handled efficiently by
humans only, especially where the majority of questions would not necessarily be
complex. Such a solution will require access to a range of sources, including for example
relevant: (i) Legislation, (ii) Regulations detailing legislative requirements, (iii) Policy
documents, (iv) Judgments of courts as well as decisions of the Administrative Appeal
Tribunal, (v) and prior spent conviction release decisions.
10
https://www.oaic.gov.au/individuals/privacy-fact-sheets/general/privacy-fact-sheet-
41-commonwealth-spent-convictions-scheme
88
Fig.1 Spent Convictions scheme. Source: Office of the Australian Information
Commissioner.
2.3. Focus of the Compliance through Design project
The use cases at the heart of the Compliance through Design project are centred around
the Commonwealth spent conviction scheme.
89
“Scheme” is used in this context to refer to the totality of the hard and soft law rules,
structures and cultures that govern, enable, operate and shape the decisions regarding the
denial or release of information regarding spent convictions. The scheme itself is not law
and is not confined to the law but embodies the whole framework of rules, cultures and
structures relevant to spent convictions. Such schemes are identified to cluster, simplify,
convey, discuss, interpret and analyse policy and implementational frameworks,
including legal frameworks.
The project focuses on the Commonwealth spent conviction scheme embodied in Part
VIIIC of the Crimes Act 1900 (Cth). The complex definition of a spent conviction under
this scheme is summarised as follows by the Office of the Australian Information
Commissioner:
A “spent conviction” is a conviction of a Commonwealth, Territory, State or foreign offence
that satisfies all of the following conditions: (i) it is 10 years since the date of the conviction
(or 5 years for juvenile offenders); AND (ii) the individual was not sentenced to imprisonment
or was not sentenced to imprisonment for more than 30 months; (iii) AND the individual has
not re-offended during the 10 years (5 years for juvenile offenders) waiting period; (iv) AND
a statutory or prescribed exclusion does not apply. (A full list of exclusions is available from
the Office of the Australian Information Commissioner).11
The scope of rights and obligations under the scheme varies depending on factors
such as: (i) whether the conviction is for a Commonwealth, state or foreign offence, (ii)
who requires the information and for what purpose, (iii) and where the person requiring
the information is located.
A a practical level a range of questions arise in relation to the spent conviction
scheme, for example: Has a person been convicted for purposes of the scheme? Has that
conviction become quashed or has the person been pardoned? Has the necessary waiting
period expired? Have any further convictions been handed down since the initial
conviction? Do any exclusions apply? What are the privacy rules that apply and the
ethical and legal implications relating to privacy and other rights of an individual if
information that should not be released, is released?
To extract or elicit a full set of modelling requirements or constraints can become a
complex task, because interpretation and human decision-making enter into several
stages of the information workflow. This is the reason why the approach of legal
Compliance through Design (CtD) can be appropriate at the implementation level to
counter-balance and complement its semantic processing.
In addition to the rules of the scheme, there is a body of scholarly criticism of spent
convictions schemes, especially their contents and their operation [15] [16] [17] [18] [20]
[21]. It is important for the project to engage the scholarly comment, but such
engagement falls outside the scope of this paper. It will however be discussed in a
following paper that will provide a more detailed analysis of the scheme.
11
https://www.nationalcrimecheck.com.au/resources/spent_convictions_information
90
3. Overview of the architecture
3.1. Purpose
The project aims to develop a platform where Compliance by Design (CbD) and
Compliance through Design (CtD) principles can jointly guide work processes and
decision-making. For this purpose, an explicit compliance element complements the
information management and process management element that are commonly found in
human-in-the-loop information systems to ensure that relevant policies, rules, and
associated legal constructs are available and enforced by the system where that is
appropriate.
The conceptual architecture for the system is depicted in Figure 1. The user interface
layer implements a consistent user-facing portal for searching and accessing information
and for interacting with the workflows embodied within the system. Automation Services
drive the execution of business processes, whereas Compliance Services assess the
compliance of process executions with relevant rules. The process automation and
compliance mechanisms build upon the Knowledge Hub Services and the Search
Services that provide data access and discovery. Platform Services offer overarching
functions for logging, monitoring, and security complete the architecture. The role of the
major elements is described in the following paragraphs. Details about the technical
implementation of data stores and data processing pipelines are beyond the scope of this
paper and are presented in [1] [2].
Fig. 2. Architecture overview
91
3.2. Knowledge Hub and Search Services
The Knowledge Hub Services aim to serve as a single point of access to the data
maintained by the system, whereby services for ingesting, accessing, processing, and
linking data support the data management lifecycle of the applications built upon the
platform. The pool of data can be exposed in the form of a domain-specific knowledge
graph which comprises entities, their attributes, links, augmented with provenance and
meta-data.
The organization of the knowledge graph is governed by an explicit ontology which
describes the types of elements, links, and meta-data that may occur in the knowledge
graph. In addition to supporting search and data organization, the semantic models
embodied in the ontology enables the platform to associate data with the relevant process
elements in the Automation Services element and provide an anchor for identifying
relevant rules, policies, and related information within the Compliance element of the
architecture. The Search Services element provides functions that support the discovery
and access of information provided by the knowledge hub and includes functions for
browsing, keyword search, and semantic search expansion using the ontology.
3.3. Process Automation Services
The Automation Services address requirements related to workflow definition, execution,
and automation. The technology underlying this element rests on process templates that
are instantiated in the context of a specific workflow scenario. Our current
implementation rests on a Business Process and Notation (BPMN) workflow engine for
workflow execution with embedded explicit decision models (DMN), and an event
notification mechanism that relays relevant business events to the process engine. The
automation services use mediators to External Services that invoke and access to external
systems.
A library of workflows, tasks, and information objects complemented with rules that
govern process execution can be created and used to support the execution of the system.
Configurable process templates specify the dependencies between activities, whereby
process parameters determine the fillers for placeholder roles, data elements, and
concrete sub-processes that implement hierarchical process steps. For example, business
rules embedded in process templates select appropriate sub-processes tailored for
communicating with different external organizations (to address variety in required
information and technical submission procedures) and determine decisionmakers for
manual steps. This configuration step is based on information in the knowledge graph
capturing the context of the process, organization structure, and external parties’ systems.
Information that cannot be acquired automatically is supplied by the user.
Although process configuration and rule-based execution can accommodate defined
processes and variations, manual intervention may be required if exceptions arise, if
conflicting or ambiguous business rules apply, or if human interpretation is desired. Our
approach to automation aims to detect this and fall back to human intervention. This
hybrid strategy simplifies the approach as exceptional cases do not need to be modelled
in detail for each process. In the context of law enforcement investigations, we hold that
a semi-automated approach is sufficient, provided that all actions and responses are duly
captured on a timeline in a log. However, simply abandoning all process governance
92
would be inappropriate in most cases. Instead, a CtD approach can be employed to
inform the decision makers and help them arrive at compliant decisions.
3.4. Compliance Services
The Compliance Services provide services related to checking and enforcing compliance
of workflow executions with set policies, rules, and the law. It interacts with the
Automation Services to inform the configuration of process templates, to detect
compliance violations and enforce compliance where possible based on the events that
occur in workflows and the related data in the knowledge hub.
The compliance module includes a Compliance Knowledge Graph that links
processes, actions and events, and data elements in the Knowledge Hub to the relevant
compliance-related knowledge. In the simplest form this Compliance Knowledge Graph
captures the compliance rules that apply to events and data elements in each process.
These rules can be interpreted by the Compliance Services to assess compliance and
impose decisions on the process configuration and execution by the Automation Services
for CbD purposes. Furthermore, the provenance of each entry in the compliance module
is maintained to enable users to discover and review relevant information for CtD
purposes.
Although proactive compliance-enforcing measures could in principle be
implemented by embedding decision rules within each process template, separation of
the compliance mechanism from the business process execution model is advantageous,
as a simple process model that is loosely coupled with a separate compliance element
simplifies both elements. Moreover, advanced compliance mechanism that include for
example, reasoning about norms, detecting anomalous events from using artificial
intelligence techniques, and compliance models that may have been acquired from
natural language text can be integrated and validated more easily in a modular
architecture.
The architecture as described so far rests on the assumption that the relevant
procedures and policies are known, well understood, and that they have been expressed
in the form of semantic models that the machine can interpret. Although semantic models
may be devised using natural language processing techniques [6], challenges remain in
the disparity between rigid formal representations (e.g., formal modal logics) and the
often context-dependent interpretation of legal texts. Here, a Compliance through Design
approach is needed. The Compliance module enables users to browse and retrieve
compliance-related information, such as rules and legal texts, that can inform their
decision-making. This could be done encompassing CtD requirements to improve the
quality and relevance of legal information.
To demonstrate that compliance by/through design is indeed possible for complex
legal scenarios, such as the spent convictions scheme outlined in this paper, the workflow
steps and associated and compliance monitoring functions will be implemented in an
automated system. The tasks required to accomplish this include the analysis of relevant
legal texts and translation of the natural language text into formal models; the definition
of the workflow for spent conviction request processing and the variation points within
the workflow as well as events related to compliance verification; the implementation of
the workflow in an automated workflow management engine; and implementation of a
reasoning engine that verifies the workflow execution with respect to the formal
compliance model.
93
At the time of writing the legal texts pertaining to spent convictions are being
analyzed and translated into formal models in that will enable automated reasoning about
the rules that govern the spent conviction disclosure process. Our formalism of choice is
LegalRuleML [22] as this language is formally precise and it includes advanced ideas,
such as defeasible reasoning, which are essential for representing legal rules in the
presence of conflicting and incomplete information.
Concurrently, models of the process and data necessary to assess the compliance
conditions are being defined. The process model captures the workflow the users of the
system would follow when ascertaining SC disclosure requirements, whereas the data
model specifies the data elements (and type information) that describe the concrete case
that is being assessed. The process model will be implemented in an automated workflow
management system (such as Camunda12) which will be linked to a data store in the
Knowledge Hub component for information access. The process implementation will
rely on a separate compliance verification module to verify that the workflow execution
is compliant with the rules that apply to the specific case. Although LegalRuleML is a
powerful language that can represent complex legal rules, it is a relatively new language
that is not yet well supported by automated reasoning systems.
Although formal semantics and translations to other formal systems (such as
Defeasible Logic) have been devised [23], extensions to existing reasoning systems may
need to be made to be able to correctly interpret the entire set of compliance rules for SC
formulated in LegalRuleML. A suitable implementation technology will be chosen for
the compliance module based on the nature and complexity of the compliance rules
pertaining to the SC scenario, and the reasoning engine will be linked with the process
engine to support reasoning about compliance.
The prototype system will serve as a demonstrator for automated legal compliance
verification. Experiments with the demonstrator will be conducted to assess the
effectiveness and efficiency of the automated compliance mechanism and identify areas
where future research may be necessary to further improve compliance monitoring and
verification in the legal context.
12
https://camunda.com/
94
4. Future work: Compliance processing and SC regulatory model
We are conducting at the moment several subprojects with coordinated objectives,
reflected on CRC Deliverables: (i) the creation of a minimal portal for the search of legal
information related to spent convictions through heterogeneous publication formats [27],
(iii) Legal RuleML [28] modelling, (iii) a survey on the differences between Compliance
by and through Design [29 ] (iv) steps for the interpretation of spent conviction schemes
[30], (v) potential interpretations of elements of Part VIIC of the Crimes Act 1900 (Cth)
[31], and (vi) Australian case law on spent convictions [32].
This will inform a clearer definition of what legal compliance may consist of,
compared to regulatory compliance [30].
Acknowledgements. This research was partially funded by the Data to Decisions
Cooperative Research Centre (D2D CRC, Australia), and Meta-Rule of Law (DER2016-
78108-P, Spain). Views expressed herein are however not necessarily representative of
the views held by the funders.
References
[1] Mayer, W., Stumptner, M., Casanovas, P., de Koker, L. Towards a Linked Information
Architecture for Integrated Law Enforcement. In M. Poblet, E. Plaza, P. Casanovas (ed.)
Linked Democracy: Artificial Intelligence for Democratic Innovation, IJCAI-2017 Workshop,
August, Melbourne, CEUR, pp 15-37 (2017), http://ceur-ws.org/Vol-1897 /
[2] Stumptner, M., Mayer, W., Grossman, G., Liu, J., Li, W., Casanovas, P., De Koker, L.,
Mendelson, D., Watts, D., & Bainbridge, B. An Architecture for Establishing Legal Semantic
Workflows in the Context of Integrated Law Enforcement. In AI Approaches to the
Complexity of Legal Systems: AICOL International Workshops 2015–2017: AICOL-VI@
JURIX 2015, AICOL-VII@ EKAW 2016, AICOL-VIII@ JURIX 2016, AICOL-IX@ ICAIL
2017, and AICOL-X@ JURIX 2017, Revised Selected Papers (p. 124-139). LNAI 10791
Cham: Springer (2018)
[3] Australian Human Rights Commission, Discrimination in Employment on Basis of Criminal
Record (December 2004)
[4] Knowler, J. ‘Living down the past – spent convictions schemes in Australia’ [1994] Privacy
Law and Policy Reporter 80 (1994)
http://www.austlii.edu.au/au/journals/PLPR/1994/80.html
[5] Commonwealth: Part VIIC of the Crimes Act 1914 (Cth); New South Wales: Criminal Records
Act 1991 (NSW); Queensland: Criminal Law (Rehabilitation of Offenders) Act 1986 (Qld)
[6] Australian Capital Territory: Spent Convictions Act 2000 (ACT; Tasmania: Annulled
Convictions Act 2003 (Tas); Western Australia: Spent Convictions Act 1988 (WA).
[7] Northern Territory: Criminal Records (Spent Convictions) Act 1992 (NT)
[8] South Australia: Spent Convictions Act 2009 (SA)
[9] Tasmania: Annulled Convictions Act 2003 (Tas)
[10] Western Australia: Spent Convictions Act 1988 (WA)
[11] Australian Criminal Intelligence Commission (ACIC), Submission to Productivity
Commission’s Inquiry into Data Availability and Use (2016) ACIC Criminal History Checks
https://www.acic.gov.au/criminal-history-checks
[12] Darby, P., Kretzenbacher, J., Simmonds, S. Liberty Victoria’s Rights Advocacy Project, A
Legislated Spent Convictions Scheme for Victoria: Recommendations for Reform (May
2017): https://libertyvictoria.org.au/sites/default/files/RAP-report-A-Legislated-Spent-
Convictions-Scheme-for-Victoria-Recommendations-for-Reform-20170517.pdf
95
[13] Office of the Australian Information Commissioner, Privacy Fact Sheet 42: Commonwealth
spent convictions scheme (May 2014): https://www.oaic.gov.au/individuals/privacy-fact-
sheets/general/privacy-fact-sheet-41-commonwealth-spent-convictions-scheme
[14] Australian Human Rights Commission, Discrimination in Employment on Basis of Criminal
Record (December 2004): https://www.humanrights.gov.au/human-rights-discrimination-
employment-basis-criminal-record-0
[15] Lenton, S., 2004. Pot, politics and the press—reflections on cannabis law reform in Western
Australia. Drug and alcohol review, 23(2), pp.223-233 (2004)
[16] Weathered, L., Pardon me: Current avenues for the correction of wrongful conviction in
Australia. Current Issues Crim. Just., 17, p.203-216. (2005)
[17] Hamer, D., Wrongful convictions, appeals, and the finality principle: The need for a criminal
cases review commission. UNSWLJ, 37, p.270-311 (2014)
[18] Dioso-Villa, R., A repository of wrongful convictions in Australia: first steps toward
estimating prevalence and causal contributing factors. Flinders LJ, 17, p.163 (2015)
[19] Naylor, B., Criminal records and rehabilitation in Australia. European Journal of Probation,
3(1), pp.79-96 (2011)
[20] Bagaric, M., The disunity of employment law and sentencing. The Journal of Criminal Law,
68(4), pp.329-355 (2004)
[21] Hufnagel, S., 2012. Harmonising police cooperation laws in Australia and the European
Union: the tension between local/national and national/supranational interests. Australian
Journal of Forensic Sciences, 44(1), pp.45-62 (2012)
[22] Athan, T., Governatori, G., Palmirani, M., Paschke, A. and Wyner, A. LegalRuleML: Design
principles and foundations. In Reasoning Web International Summer School (pp. 151-188).
Springer (2015)
[23] Lam HP., Hashmi M., Scofield B. Enabling Reasoning with LegalRuleML. In: Alferes J.,
Bertossi L., Governatori G., Fodor P., Roman D. (eds) Rule Technologies. Research, Tools,
and Applications. RuleML 2016. Lecture Notes in Computer Science, vol 9718. Springer
(2016)
[24] Casanovas, P., Palmirani, M., Peroni, S., van Engers, T., Vitali, F. Special Issue on the
Semantic Web for the Legal Domain, Guest Editors Editorial: The Next Step, Semantic Web
Journal 7 (2), pp. 1-13 (2016)
[25] Watts, D., Casanovas, P. "Privacy and Data Protection in Australia: a Critical overview."
W3C (2018) https://www.w3.org/2018/vocabws/papers/watts-casanovas.pdf
[26] Athan, T., Governatori, G., Palmirani, M., Paschke, A. and Wyner, A. LegalRuleML: Design
principles and foundations. In Reasoning Web International Summer School (pp. 151-188).
Springer (2015)
[27] Rodríguez-Doncel, V. Deliverable D.C3.3. Natural Language Processing and Semantic
Tools: A Roadmap towards publishing law as data. CRC Programme, La Trobe CRC D2D,
30 January 2019.
[28] Governatori, G. Deliverable D.C3.7 LegalRuleML for the Australian Spent Convictions
model. DC25008: Compliance by Design (CbD) and Compliance through Design (CtD). CRC
Programme, La Trobe CRC D2D, 30 January 2019.
[29] Casanovas, P. Hashmi, M. Deliverable D.C3.2. Survey on Regulatory and Legal Compliance:
Legal Quadrant and Conceptual Clustering. CRC Programme, La Trobe CRC D2D, 30
January 2019.
[30] De Koker, L., Casanovas, P. Law and Policy Deliverable D.C3.1 Spent Convictions Use Case.
Preliminary Notes. CRC Programme, La Trobe CRC D2D, 20 January 2019.
[31] Stammers, M. Deliverable D.C3.4. Spent Convictions Scheme: Steps of the Legal Analysis.
CRC Programme, La Trobe CRC D2D, 30 January 2019.
[32] Barnes, J. Deliverable D.C3.5. Crimes Act 1900 (Cth) – Part VIIC – Division 3: Sections
85ZV, 85ZW and Associated Definitions. Potential Interpretative Issues. La Trobe CRC D2D,
30 January 2019.
[33] O’Toole, S., Keyzer, P. Deliverable D.C3.6 Case law on Spent Convictions: Set of Sentences
and Summaries. La Trobe CRC D2D, 30 January 2019.
96