=Paper=
{{Paper
|id=Vol-1283/paper24
|storemode=property
|title=
Open Source Intelligence, Open Social Intelligence and Privacy by Design
|pdfUrl=https://ceur-ws.org/Vol-1283/paper_24.pdf
|volume=Vol-1283
|dblpUrl=https://dblp.org/rec/conf/ecsi/Casanovas14
}}
==
Open Source Intelligence, Open Social Intelligence and Privacy by Design ==
https://ceur-ws.org/Vol-1283/paper_24.pdf
Open Source Intelligence, Open Social Intelligence and
Privacy by Design
Pompeu Casanovas 1,2
1
Institute of Law and Technology, Autonomous University of Barcelona,
08193-Barcelona, Spain, pompeu.casanovas@uab.cat
2
Centre for Applied Social Research, Royal Melbourne Institute of Technology, Australia,
pompeu.casanovas@rmit.edu.au
Abstract. OSINT stands for Open Source Intelligence, (O)SI for (Open) Social
Intelligence, PbD for Privacy by Design. The CAPER project has built an
OSINT solution oriented to the prevention of organized crime. How to balance
freedom and security? This position paper describes a way to embed the legal
and ethical issues raised by the General Data Reform Package (GDRP) in
Europe into this kind of surveillance platforms. It focuses on the indirect
strategy to flesh out Privacy by Design principles (PbD) through Semantic Web
Regulatory Models (SWRM). Institutional design, self-regulatory systems, and
the possibility to build up a meta-level rule of law are discussed.
Keywords: CAPER, Open Source Intelligence (OSINT), Social Intelligence
(SI), Semantic Web, Institutions, Semantic Web Regulatory Models (SWRM),
Privacy by Design (PbD),
1 Preliminaries: the legal and political problem
There are many ways to face and conceptualize intelligence —individual, collective,
swarm. etc. —to describe and research how it works or how to use it in courses of
action. Since 2001 onwards this practical side has received a strong boost. The
explosion of Internet, the wide use of protocols HTML, the speeding of the Semantic
Web through W3C standards, are related to it. But the entrance in the new century
through what it was called the global terrorist threat after September 11th was key to
fund and promote new research programs for military use. Some of these programs
are focused on Open Source Intelligence (OSINT).
The word is a bit misleading, because it is well known that Open Source (OS) has a
venerable history in computing. There is also an ongoing discussion in legal theory on
the role that OS plays in intellectual property, licenses, publishing and patents. But
when applied to intelligence, OSINT does not refer only to the origin of the digital
outcome but to the legal and political sphere of the community where the "intelligent"
outcome is encapsulated, distributed, reused and transformed. What does it mean for a
document, an image, a video to be qualified as OSINT? Basically that it is public
domain or, better, no man's land domain, free to be grabbed and manipulated for
public reasons by main representatives in nation-states —LEAs (Law Enforcement
Agencies), Intelligence Services, State Agencies...
This paper deals with the relationships and differences between (Open) Social
Intelligence (OSI) and Open Source Intelligence (OSINT). That is to say, with the
ways to combine freedom and surveillance. This is at present one of the hottest topics
of European legislation that it is worthwhile to tackle from a regulatory point of view.
The so-called General Data Reform Package (GDRP) is at stake. The new rules
intend to put citizens back in control of their data, notably through: (i) a right to be
forgotten (when you no longer want your data to be processed and there are no
legitimate grounds for retaining it, the data will be deleted); (ii) easier access to your
own data (a right to data portability to make easier to transfer personal data between
service providers; (iii) putting citizens in control (requirement of explicit consent to
process personal data), (iv) Privacy by Design (PbD) and Privacy by Default (PbD)
—as they are becoming essential principles in EU Data Protection Rules [19].
The recent Opinion 28 released by the European Group on Ethics (EGE) of 20 May
2014 describes Ethics of Security and Surveillance Technologies [17]. The Opinion
advances a set of sixteen concrete recommendations for the attention of the EU,
member states, and a range of public and private stakeholders. It "challenges the
notion that 'security' and 'freedom' can be traded against one another", and "calls for
a more nuanced approach, in which the proportionality and effectiveness of security
and surveillance technologies are subject to rigorous assessment, and in which rights
are prioritized rather than traded". Certain core principles, such as human dignity,
cannot be bartered with [18].
What does it exactly mean? In this position paper, I will show that replacing the
mechanism of security as a general exception to rules by another approach in which
other principles apply adds some complexity to the balance between freedom and
security. I will follow up some ideas raised into the CAPER EU Project [10] 1,
stemming from some of the conclusions : (i) even in surveillance toolkits and serious
security issues some feasible ways to bridge PbD principles and citizens' rights are
possible; (ii) PbD can be broadly understood as a form of institutional design; (iii)
ethics are starting to play a major role in such regulatory tasks.
2 A meta-level rule of law?
Let's start in reverse order and put some theoretical statements first. The problem of
dealing with law is that law is not a well-defined field, where you can easily find its
constituent elements and model them into rules and consequences. This is a
discussion that has been lasting for centuries in the modern world. The French jurist
Jacques Cujas (Cujacius, 1522-1590) defined the legal method as "the method of lack
of method" in the 16c. Jeremy Bentham expressed the same feelings about lawyer's
cant. Rules, norms, principles and values are expressed in natural language, and there
is a real problem to face them analytically, because the same statute, article or concept
might be interpreted in different ways any time they are instantiated in a ruling or in a
decision-making process, and in fact they are.
1 Collaborative Information, Acquisition, Processing and Reporting for the Prevention of
Organized Crime (CAPER) http://www.fp7-caper.eu/
To put it crudely: law cannot be completely modelled. This is not saying that
cannot be modelled. Only that cannot be modelled up to the end. This means at least
two apparently opposing things: (i) there are objective limitations to formalize legal
statements, (ii) the analyst is bounded to complete this missing part by settling some
general framework for his own sake.
This can be like a leap into the dark, but it is most interesting to note that any
practical decision or implementation of norms entails a theoretical ground. That is to
say, from the epistemic point of view, the analyst is working at the same time through
a constructing language and from a structuring meta-system for such a language.
Models and meta-models come alike, and one of the interesting tasks is to make
explicit the inner structure of the framework (the meta-model) legal models are made
of or sorted out.
There are at least four kind of complexities related to: (i) semantic languages (and
the explosion of data and metadata on the web), (ii) the socio-cognitive structure of
individual and social mind (iii) the logical structure of nMAS (based on concepts like
agency), (iv) institutional building within social ecosystems. To my purposes, I will
focus on the latter one.
Sue Crawford and Elinor Ostrom defined institutions as formal and informal rules
that are understood and used by a community [15]. Therefore, as advanced by legal
realists in the thirties, institutions are not automatically what is written in formal rules.
It is widely known that Vincent and Elinor Ostrom were looking for a dynamic and
flexible framework expressible in a meta-model able to account for the fluidity of
property rights (access, contribution, extraction, removal, exclusion, management/
participation, exclusion, alienation) and to overcome the static view of the "tragedy of
the commons".
Their solution is generally referred to as the just right one, as balancing the aim of
reducing the high costs associated with political solutions with that of ameliorating
the absence of incentives to create solutions at all in the market-based approach [1].
Aligica and Boettke [2] have just highlighted that there is a deep social philosophy
behind the Ostroms' approach. They developed in fact two frameworks: the
Institutional Analysis and Development (IAD) framework as well as the Diagnostic
Ontological Social-Ecological Systems (DOSES) framework. Thus, this philosophy
holds in two different sides: (i) polycentricity, a notion that challenges two of the
most common assumptions of political and economic sciences in the twentieth
century (the monocentric vision of social order, and the “market” vs. “state”
dichotomy); (ii) a view of social order seen as a knowledge and learning process
within social ecosystems (factors engendering institutional order as a response to the
challenges posed by them). But as the authors point out, there is no necessary
connection linking the two perspectives involved in IAD and DOSES.
I will quote them at length, because Aligica and Boettke are fleshing out one of the
more pervasive and striking problems that we have to face at present:
" To sum up, social order and its institutional dynamics are seen as shaped by (and
operating under) the shadow of the ongoing tension between the “threat of chaos” and
the “threat of tyranny.” Force and political constraint can be used as “instruments of
tyranny as well as instruments to support productive and mutually advantageous
relationships.” Could the problem of elemental asymmetry in the relationship between
the “rulers” and the “ruled” be dealt in any way? Does, by its very nature, social order
require that someone rules over society and cannot be held accountable to other
members of society? Is it possible to conceptualize and organize the relationship
between rulers and ruled so that rulers themselves are subject to a rule of law? In
other words, could we design a “meta-level rule of law” where rulers themselves are
subject to enforceable rules? Could we encapsulate it creating a climate dominated by
deliberation and critical reason tempering the rulers and the application of force by
checking and balancing them not only with the force of rules but also with reason and
deliberation. The very effort of specifying such a solution is a means to appreciate the
deep tensions that are involved in establishing a system of governance where both the
threat of chaos and the threat of tyranny are circumvented. We are already in the
territory of the science and art of self-governance.”
This possibility of "a meta-level rule of law" is crucial, because so far the
mainstream of legal theory has not leaned on the understanding of decentralized
boundaries, but on a hierarchical structure of norms, obligations and rights that are
more based on the blueprint of normative or deontic languages than on the result of
social and ecological analyses. How this meta-level rule could be built then?
It is worth noticing that classical 20 c. legal solutions to this problem do not apply
here. Kelsenian Grundnorm, Hartian recognition rule or the assumption of selective
sources of law to set a principle of legality —Alf Ross' solution for the grounding of
national statutes— cannot constitute potential answers any more, as from the
ecosystems perspective law does not encompass the whole problem, but only one of
its dimensions. A similar problem is faced in agreement technologies, when Online
Dispute Resolution (ODR), nMAS and crowdsourcing enhance aggregated collective
information and social intelligence processes [8].
Besides, legal theory approaches lean on the (not always assumed) situated
perspective taken by lawyers, jurists and rule-makers in conflictive environments. The
Polish logician Jerzy Wróblewski put it quite clearly when he stated that lawyers
make the law of the land, and therefore the language of law is intertwined with the
language of lawyers. Legal design is to a great extent political design too. Let's face
the problem from another side.
3 Open Source Intelligence (OSINT) and Social Intelligence (OSI)
OSINT, is usually defined as intelligence collected from open sources. There is no
homogeneous approach to this concept, depending on the field, purposes and actions
in which it is used. Intelligence Services refer to it as "unclassified information that
has been deliberately discovered (...) to a select audience", i.e. military uses [30].
Reports for the US Congress are quite clear about this: "A consensus now exists that
OSINT must be systematically collected and should constitute an essential component
of analytical products" [5]. This is especially felt when working on the edges of law,
balancing freedom and security, surveillance and fundamental rights. General Michael
Hayden, former director of the NSA and the CIA, furnishes an extreme example. He
conspicuously stated in a recent debate with David Cole at John Hopkins University:
We kill people based on metadata [14].
But OSINT is considered also for other unrelated aims a cluster of tools to browse
the web, aggregate information, and getting reliable profiles from websites, blogs,
social networks, and other public settings. From this broader point of view, it may be
defined synthetically as "the retrieval, extraction and analysis of information from
publicly available sources" [4]. This approach is taken by many to get, structure and
manage information in a broad array of social domains —media [6], education [24],
business [20], disaster management [3].
The former is a definition of the concept referring to functions being performed by
a computer system —retrieval, extraction and analysis of information. It is ostensive
in nature, offering a descriptive meaning. Could OSINT somehow be related to
(Open) Social Intelligence (O)SI)?
Only in a superficial way. Both concepts have an operational side and denote the
circulation and transformation of non-structured information into structured
information on the web. In their history of the concept, Glassman and Klang offer a
communicative and cultural approach —"the Web as an extension of mind", OSINT
as the interface between fluid intelligence and crystallized intelligence [3].
If this is so, the field, methodology and theory of Social Intelligence comprehends
what is referred by OSINT, as the social mind is faced as a set of social affordances
that can be represented, described and reproduced computationally as inner
mechanisms, as social artifacts performing a collective work [11]. Social Intelligence
focuses on the human/machine coordination of artificial socio-cognitive technical
systems, assuming that they interact in a shared web-mediated space with aims,
purposes, intentions, etc. and are amenable to models and meta-models from a
theoretical point of view [27].
If the concept of OSINT tools or platforms is used to describe the operational
functionalities of a computational system, this use should be embedded into a
conceptually broader set of notions to be effective. In artificial socio-cognitive
systems "rationality is based on the model that agents have of the other agents in the
system" [28]. This epistemic assumption is not necessary in OSINT systems, more
pragmatically oriented and centered on visual analytics and on the interface between
intra and inter-organizational teams.
Let's face this problem stemming from a different side —which meta-model would
be needed to build the institutional design of civil rights protections (the balance
between freedom and security)?
4 Privacy by Design (PbD)
For some time now, the main concepts of Privacy by Design (PbD) —and Data
Protection by Design and by Default— have been widely spread over computer
science and legal research communities [12]. This conceptual body intends to develop
the Principles of Fair Information Practices (FIPs) 2 that follow (i) from the Alan
2 1. Openess and Transparency, 2. Individual Participation. 3. Collection Limitation, 4. Data
Quality, 5. Use Limitation, 6. Reasonable Security. 7. Accountability.
Westin tradition in private law 3, (ii) and the technological idea of a meta-layer to
manage and secure the identity of users on the web set by the Microsoft architect
Kim Cameron [7].
Ann Cavoukian has recently asserted that "it is not true that privacy and security
are mutually opposing", and that big and smart data "proactively builds privacy and
security in" [12]. It might be true, but it is not evident in the fight against organized
crime. Both in the military or humanitarian fields, to be effective OSINT tools have
been designed just for what they should be controlled: spotting as much as possible
and getting personal information about individuals and organizations.
Bert-Jaap Koops, Jaap-Henk Hoepman [23] and Ronald Leenes [25] [26]
experienced this void on the sidelines of law and technology when they faced the
problem of modelling the protections of General Data Reform Package into OSINT
platforms. 4 As might be expected, they found that privacy regulations cannot be
hardcoded —"‘privacy by design’ should not be interpreted as trying to achieve rule
compliance by techno-regulation. Instead, fostering the right mindset of those
responsible for developing and running data processing systems may prove to be
more productive. Therefore, in terms of the regulatory tool-box, privacy by design
should be approached less from a ‘code’ perspective, but rather from the perspective
of ‘communication’ strategies [emphasis added]".
Regarding the analysis, I am going along to the authors' guidelines, but perhaps
another conclusion could be drawn as well from these limitations. There are other
possibilities to embed PbD into surveillance platforms, albeit indirectly, i.e. adding
theoretical views not thinking of techno-regulation but of what law means when
constructed through technological means. It is properly legal theory what could be
worked out from the perspective of self-governance and socio-cognitive artificial
systems.
E,g. CAPER is a system comprising a platform whose main functionalities are
accorded to OSINT features: (i) implementing a framework to perform the task of
connecting multiple data sources with multiple visualization techniques via a
standardized data interface, including support for data-mining components; (ii)
enabling a quick import of data types from disparate data sources in order to improve
the ability of different LEAs to work collaboratively; (iii) supporting pattern
discovery, documentation and reuse, thus increasing progressively detection
capabilities. The architecture design has four components: (i) Data harvesting
(knowledge acquisition: data gathering), (ii) analysis (content processing), (iii)
semantic storage and retrieval, (iv) and advanced visualization and visual analytics of
data. For sake of clarity, Fig. 1 shows the interaction and workflow between
databases.
3 These historical origins must still be retraced and reconstructed carefully. I am grateful to
Graham Greenleaf for this observation.
4 VIRTUOSO (Versatile InfoRmation Toolkit for end-Users oriented Open-Sources
explOitations), http://www.virtuoso.eu/ .
Figure 1. CAPER Databases Overview.
We explored several related strategies that constitute an indirect approach to PbD
principles; and five different strategies were designed [10]:
1. CAPER workflow is addressed in fact to four different LEA´s analysts: (i)
The Generic Analyst (GA) (ii) the Advanced Analyst (LEA-AA) (iii) the
System Administrator, (iv) LEA's External User (LEU).
2. Well-defined scenarios are extracted from the experience of managing
specific types of crimes. CAPER tools operate only within the investigation
conducted by LEAs, helping them to better define the lines of research, but
avoiding any automated conceptualization of them.
3. Well-defined module interdependencies are drawn in advance. CAPER
crawling system is sustained by three modules: (i) crawler by keyboard, (ii)
by URL, (iii) by URL focusing on keyboards. But the crawler is able to
convert images and videos metadata into allowed mimetypes required by the
Visual Analytics module (VA). Multi-lingual semantics is added to the whole
process as well.
4. Two different kind of ontologies have been built. The first one is a Multi-
lingual Crime Ontology (MCO) for 13 languages, including Hebrew and
Arabic, with a proof of concept on drugs (346 nodes). MCO works
according to the country legislation (possession of drugs is a crime in UK,
but not in Spain e.g.).
5. The second one is properly a legal ontology, focused on European LEAs
Interoperability (ELIO). ELIO has been built with the purpose of improving
the acquirement and sharing of information between European LEAs [22].
This pack of actions constitutes a warrant for citizens' rights. Design means
institutional design as well. The notion of institutional-Semantic Web Regulatory
Model (i-SWRM) leans on this assumption: a self-regulatory model embeds the
dimension of PbD into a technological environment than can be represented as a
social ecosystem [8]. This is similar (with some differences) to the framework set in
[25] [26] and specifically in [23].
Hoepman's PbD strategies model focuses on the inner structure of the modeling,
signaling several points into a general framework for privacy (or data protection)
closed managerial system. This is consistent with the idea of concentrating the effort
onto the interpretation of the law, putting aside the specific problems, risk scenarios,
and asymmetric multilayered governance of OSINT platforms, end-users and LEA.
Conversely, i-SWRMs and specifically the regulatory model designed to regulate the
CAPER workflow system (CRM), are more focused on LEA's inner and outer
relationships. The social ecosystem centered on the specific data that users are
processing and "living by" can be outlined as a simple scheme described in advance
(see fig. 2).
Figure 2. CAPER regulatory scheme [22].
Affordances can be plotted on the CAPER workflow, along with data protection
rules. However, 9 out of the 23 rules, could not be situated because they don't apply to
the information process, but to LEA's behavior (analysts, controller, administrator).
This holds, e.g. for a positive obligation such as "Every LEA should perform a
specific Privacy Impact Assessment (PIA) according to the general framework
offered by the CAPER Regulatory Model (CRM)", a general prohibition as "No
automated classification of suspects, victims and witnesses can be inferred from
CAPER results", and for specific obligations such as "The alleged reasons to deny
access should be open to external supervision. The external supervisory authority
should have free access to documents justifying the refusal. A short time-span of three
months to give an answer to a previous request of access should be implemented".
5 The role of normative and institutional Semantic Web Regulatory Models (n-
SWRM, i-SWRM)
Let's get back to Aligika's and Boettke's [2] main questioning: Would it be possible to
conceptualize and organize the relationship between rulers and ruled so that rulers
themselves are subject to a rule of law? In other words, could we design a “meta-level
rule of law” where rulers themselves are subject to enforceable rules?
Semantic Web Regulatory Models (SWRM) co-organize the conceptual
architecture of (enforceable) hard law and (non-enforceable) soft law, best practices
and ethics. Compliance with norms is important, but it is not enough. Reaching a
sustainable behavioral model is equally important. The validity of the system is
shown through the emerging independent axis of institutional strengthening. This
approach presents the additional advantage of being measurable.
E.g., in the CAPER example dialogue with LEA and security experts is crucial to
understand where the problems are and why, and to let LEA's investigators participate
into the regulatory process. At the same time, control is exerted because binding
norms apply as well. E.g. the need for an internal and external DP controller
(competent DP authorities) and the obligation to set a strict log file to keep records to
sustain the accountability of the whole system.
So, the Caper Regulatory Model (CRM) lying behind the actions taken intends to
bridge negotiations of social agents with the normative requirements and conditions
of the rule of law (reinterpreted from this broader standpoint). This is why it can be
implemented among LEA's organizations and embedded into the CAPER system to
regulate the use of the platform. CRM is an example of i-SWRM.
There are two types of SWRM [8]. Normative-SWRMs (n-SWRMs) use RDF,
RuleML and computer versions of rights, duties and obligations. They may lean on
the use and reuse of nested ontologies or "ontological patterns". Digital Rights
Management (DRM), Rights Expression Languages (REL) are quite useful to
represent legal licenses, intellectual property rights, or patents as data and metadata to
be searched, tracked or managed in a global way on the Internet. Creative Commons
is a well known example for using REL to express licenses. Open Digital Rights
Language (ODRL) initiative constitutes another example. Therefore, end-users and
systems are linked through the same tool that is being used to manage and apply the
modeled legal knowledge.
Institutional-SWRM (i-SWRM) are focused on the relationship of end-users with
their own organizations. Inner coordination among electronic agents, outer interface
with human (collective) agents, and their dynamic interaction within different types of
scenarios and real settings are key. They can be applied to regulatory systems with
multiple normative sources and human-machine interactions between organizations,
companies and administrations. Thus, their conceptual scheme is linked with legal
pluralism and with existing models of asymmetric multi-layered and networked
governance. Those are conceptual constructs compatible with Ostrom's social
philosophy —polycentricism and social ecosystems— because their center of gravity
lie on their dynamic social bonds. End-users and systems are connected through the
social and legal bonds that externally link them.
We could now reframe the original questions we started with. How are i-SWRS
and n-SWRS related? How could they work together to frame this meta-level rule of
law? Which self-regulatory institutions could be based on such a model?
If we start finding reasonable answers perhaps we will be able to circumvent the
dichotomy between Ostrom's particular Scylla and Charibdis —the threats of chaos
and tyranny for the 21c.
Acknowledgments. This research has been funded by the F7 EU Project Collaborative
information, Acquisition, Processing, Exploitation and Reporting for the prevention of
organised crime (CAPER) —Grant Agreement 261712—; and partially by the National
Project Crowdsourcing, DER2012-39492-C02-01.
References
1. Aldrich, J.H.: Elinor Ostrom and the "just right" solution, Public Choice 143, pp. 269-273
(2010) DOI 10.1007/s11127-010-9630-9
2. Aligica, P.D., Boettke, P.: The Two Social Philosophies of Ostroms’ Institutionalism, The
Policy Studies Journal, 39,1, pp. 29-49 (2011)
3. Backfried, G., Schmidt, C., Pfeiffer, M., Quirchmayr, G., Markus Glanzer, M., Rainer, K.:
Open Source Intelligence in Disaster Management. 2012 European Intelligence and
Security Informatics Conference, EISIC, IEEE Computer Society, pp. 254-258 (2012)
4. Best, C.: Open Source Intelligence, in F. Fogelmann-Soulié et al.(Eds.), Mining Massive
Data Sets for Security: Advances in Data Mining, Search, Social Networks and Text
Mining, and Their Applications to Security 19 , Amsterdam, IOS Press, 2008, pp. 331-344.
5. Best, R., Cumming, A.: Open Source Intelligence (OSINT): Issues for Congress, CRS
Report for Congress, Order Code RL34270, Updated January 28 (2008)
6. Bradbury, D.: In plain view: Open Source Intelligence, Computer Fraud & Security 4, pp.
5-9 (2011)
7. Cameron, K.: The Laws of Identity …as of 5/11/2005. Microsoft Corporation,
http://www.identityblog.com/stories/2005/05/13/TheLawsOfIdentity.pdf
8. Casanovas, P.: Agreement and Relational Justice: A Perspective from Philosophy and
Sociology of Law, Sascha Ossowski (Ed.) Agreement Technologies, LGTS n. 8, Springer
Verlag, Dordrecht, Heidelberg, pp. 19-42 (2013)
9. Casanovas, P.: Semantic Web Regulatory Models: Why Ethics Matter, Philosophy &
Technology, DOI 10.1007/s13347-014-0170-y. 27, 2 (2014)
10. Casanovas, P., Arraiza, J., Melero, F., González-Conejero, J., Molcho, G., Cuadros, M.
Fighting Organized Crime Through Open Source Intelligence: Regulatory Strategies of the
CAPER Project. Submitted at JURIX 14, Cracow (2014)
11. Castelfranchi, C.: Minds as Social Institutions, Phenomenology and Cognitive Science DOI
10.1007/s11097-013-9324-0 (2013)
12. Cavoukian, A.: Privacy by Design. The 7 Foundational Principles. Implementation and
Mapping of Fair information Practices. Information and Privacy Commissioner, Ontario,
Canada, (2010) http://www.privacybydesign.ca/index.php/about-pbd/7-foundational-
principles/
13. Cavoukian, A.: Privacy by Design, IEEE Technology and Society Magazine, DOI
10.1109/MTS.2012.2225459, pp. 18-19 (2012)
14. Cole, D.: 'We kill people based on metadata', New York Review of Books, 10th May (
2014) http://www.nybooks.com/blogs/nyrblog/2014/may/10/we-kill-people-based-
metadata/?insrc=wbll
15. Crawford, S., Ostrom, E.: A Grammar of Institutions, in E. Ostrom (Ed.) Understanding
Institutional Diversity, pp. 137-174. Princeton University Press, New Jersey (2005)
16. Gangemi, A.: Ontology semantic web patterns for semantic web content. Y. Gil et al.
(Eds.), ISWC 2005, LNCS 3729, pp. 262–276 (2005)
17. EGE: Ethics of Security and Surveillance Technologies, Opinion no. 28 of the European
Group on Ethics in Science and new Technologies, Brussels, 20 May 2014,
http://ec.europa.eu/bepa/european-group-
ethics/docs/publications/ege_opinion_28_ethics_security_surveillance_technologies.pdf
18. EGE: Press Release on the EGE Opinion 28, of 20 May 2014.
http://ec.europa.eu/bepa/european-group-
ethics/docs/publications/press_release_ege_opinion_28_.pdf
19. EU Commission: Progress on EU data protection reform now irreversible following
European Parliament vote European Commission - MEMO/14/186 12/03/2014,
http://europa.eu/rapid/press-release_MEMO-14-186_en.htm
20. Fleisher, C. OSINT: Its Implications for Business/CompetitiveIntelligence Analysis and
Analysts, OSINT: Its Implications for Business/Competitive Intelligence Analysis and
Analysts. Inteligencia y Seguridad , 4, pp. 115-141 (2008)
21. Glassman, M., Kang, M.J.: Intelligence in the internet age: The emergence and evolution of
Open Source Intelligence (OSINT ), Computers in Human Behavior 28, p p. 673-682
(2012)
22. González-Conejero, J., Varela-Figueroa, R., Muñoz-Gómez, J., Teodoro, E.: Organized
Crime Structure modelling for European Law Enforcement Agencies Interoperability
through Ontologies, P. Casanovas, U. Pagallo, M. Palmirani, G.Sartor Ed.) AICOL 2013,
LNAI, Heidelberg, Dordrecht, Springer, 2014 (forthcoming)
23. Hoepman, J-H: Privacy Design Strategies (extended abstract), N. Cuppens-Boulahia et al.
(Eds.), ICT Systems Security and Privacy Protection - 29th IFIP TC 11 International
Conference, SEC 2014, Marrakech, Morocco, June 2-4, 2014. Proceedings. Springer 2014
IFIP Advances in Information and Communication Technology, pp. 446-459 (2014)
24. Kim, Y.,Glassman, M., Bartholomew, M., Hur, E.H.: Creating an educational context for
Open Source Intelligence: The development of Internet self-efficacy through a blogcentric
course, Computers & Education 69, pp. 332-342 (2013)
25. Koops, B.J., Hoepman, J.H., Leenes, R.: Open-source intelligence and privacy by design,
Computer Law & Security Review 29, pp. 676-688 (2013)
26. Koops, B-J, Leenes, R.: Privacy regulation cannot be hardcoded. A critical comment on the
‘privacy by design’ provision in data-protection law, International Review of Law,
Computers & Technology (2013) DOI: 10.1080/13600869.2013.801589.
27. Noriega, P., Padget, J., Verhagen, H., d'Inverno, M.: The Challenge of Artificial Socio-
Cognitive Systems, in: AMMAS 14’ Proceedings at:
http://aamas2014.lip6.fr/proceedings/workshops/AAMAS2014-W22/p12.pdf.
28. Noriega, P., d'Inverno, M.: Crowd-Based Socio-Cognitive Systems, in Crowd Intelligence:
Foundations, Methods and Practices. European Network for Social Intelligence, Barcelona,
January 2014, ed. M. Poblet, P. Noriega, E. Plaza (eds.) http://ceur-ws.org/Vol-
1148/CROWD2014
29. Ostrom, E.: Institutional Analysis and Development. Micro Workshop in Political Theory
and Political Analysis, Proceedings f the Policy Studies Organization, New series 9, pp.
851-878 (2010) http://www.ipsonet.org/proceedings/category/volumes/2010/no-9/
30. Steele, R.D: Open Source Intelligence, Loch Johnson (Ed.), Handbook of Intelligence
Studies, pp. 129-147, Routledge, New York (2007)