Social coordination frameworks for social technical systems

Law, Governance and Technology Series 30 Huib Aldewereld Olivier Boissier Virginia Dignum Pablo Noriega Julian Padget Editors Social Coordination Frameworks for Social Technical Systems Law, Governance and Technology Series Volume 30 Series editors Pompeu Casanovas Institute of Law and Technology, UAB, Spain Giovanni Sartor University of Bologna (Faculty of Law-CIRSFID) and European University Institute of Florence, Italy The Law-Governance and Technology Series is intended to attract manuscripts arising from an interdisciplinary approach in law, artificial intelligence and information technologies The idea is to bridge the gap between research in IT law and IT-applications for lawyers developing a unifying techno-legal perspective The series will welcome proposals that have a fairly specific focus on problems or projects that will lead to innovative research charting the course for new interdisciplinary developments in law, legal theory, and law and society research as well as in computer technologies, artificial intelligence and cognitive sciences In broad strokes, manuscripts for this series may be mainly located in the fields of the Internet law (data protection, intellectual property, Internet rights, etc.), Computational models of the legal contents and legal reasoning, Legal Information Retrieval, Electronic Data Discovery, Collaborative Tools (e.g Online Dispute Resolution platforms), Metadata and XML Technologies (for Semantic Web Services), Technologies in Courtrooms and Judicial Offices (E-Court), Technologies for Governments and Administrations (E-Government), Legal Multimedia, and Legal Electronic Institutions (Multi-Agent Systems and Artificial Societies) More information about this series at http://www.springer.com/series/8808 Huib Aldewereld • Olivier Boissier Virginia Dignum • Pablo Noriega • Julian Padget Editors Social Coordination Frameworks for Social Technical Systems 123 Editors Huib Aldewereld Delft University of Technology Delft, The Netherlands Virginia Dignum Delft University of Technology Delft, The Netherlands Julian Padget Department of Computer Science University of Bath Bath, UK Olivier Boissier Laboratoire Hubert Curien UMR CNRS 5516 Institut Henri Fayol, Mines Saint-Etienne Saint-Étienne, France Pablo Noriega Intitut d’Investigació en Intel ligència Artificial (IIIA) Consejo Superior de Investigaciones Científicas (CSIC) Barcelona, Spain ISSN 2352-1902 ISSN 2352-1910 (electronic) Law, Governance and Technology Series ISBN 978-3-319-33568-1 ISBN 978-3-319-33570-4 (eBook) DOI 10.1007/978-3-319-33570-4 Library of Congress Control Number: 2016949076 © Springer International Publishing Switzerland 2016 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG Switzerland Contents Part I Preliminaries Introduction Huib Aldewereld, Olivier Boissier, Virginia Dignum, Pablo Noriega, and Julian Padget Conceptual Map for Social Coordination Huib Aldewereld, Sergio Álvarez-Napagao, Emilia García, Jorge J Gomez-Sanz, Jie Jiang, and Henrique Lopes Cardoso 11 Part II Social Coordination Frameworks ANTE: A Framework Integrating Negotiation, Norms and Trust Henrique Lopes Cardoso, Joana Urbano, Ana Paula Rocha, António J.M Castro, and Eugénio Oliveira 27 Electronic Institutions: The EI/EIDE Framework Pablo Noriega and Dave de Jonge 47 INGENIAS Jorge J Gomez-Sanz and Rubén Fuentes Fernández 77 InstAL: An Institutional Action Language 101 Julian Padget, Emad ElDeen Elakehal, Tingting Li, and Marina De Vos The JaCaMo Framework 125 Olivier Boissier, Jomi F Hübner, and Alessandro Ricci ROMAS-Magentix2 153 Emilia García, Soledad Valero, and Adriana Giret v vi Contents OperA/ALIVE/OperettA 173 Huib Aldewereld, Sergio Álvarez-Napagao, Virginia Dignum, Jie Jiang, Wamberto Vasconcelos, and Javier Vázquez-Salceda 10 Specifying and Executing Open Multi-agent Systems 197 Alexander Artikis, Marek Sergot, Jeremy Pitt, Dídac Busquets, and Régis Riveret 11 Frameworks Comparison 213 Olivier Boissier, Virginia Dignum, and Emilia García Part III Applications and Challenges 12 Application Domains 231 Julian Padget, Huib Aldewereld, Pablo Noriega, and Wamberto Vasconcelos 13 Challenges for M4SC 265 Julian Padget, Huib Aldewereld, and Wamberto Vasconcelos Contributors Huib Aldewereld Delft University of Technology, Delft, The Netherlands Sergio Álvarez-Napagao Universitat Politècnica de Catalunya, Barcelona, Spain Alexander Artikis University of Piraeus, Piraeus, Greece NCSR Demokritos, Athens, Greece Olivier Boissier Laboratoire Hubert Curien UMR CNRS 5516, Institut Henri Fayol, Mines Saint-Etienne, Saint-Étienne, France Dídac Busquets Electrical & Electronic Engineering Department, Imperial College London, London, UK António J.M Castro LIACC – Laboratrio de Inteligência Artificial e Ciência de Computadores, Porto, Portugal Dave de Jonge Artificial Intelligence Research Institute (IIIA), Spanish National Research Council (CSIC), Barcelona, Spain Marina De Vos Department of Computer Science, University of Bath, Bath, UK Virginia Dignum Delft University of Technology, Delft, The Netherlands Emad ElDeen Elakehal Department of Computer Science, University of Bath, Bath, UK Rubén Fuentes Fernández Universidad Complutense de Madrid, Madrid, Spain Emilia García Universitat Politècnica de València, Valencia, Spain Adriana Giret Universitat Politècnica de Valencia, Valencia, Spain Jorge J Gomez-Sanz Universidad Complutense de Madrid, Madrid, Spain Jomi F Hübner DAS-UFSC, Federal University of Santa Catarina, Florianópolis SC, Brazil Jie Jiang Delft University of Technology, Delft, The Netherlands vii viii Contributors Tingting Li Department of Computer Science, University of Bath, Bath, UK Henrique Lopes Cardoso Faculdade de Engenharia, Departamento de Engenharia Informática, Universidade Porto, Porto, Portugal LIACC – Laboratrio de Inteligência Artificial e Ciência de Computadores, Porto, Portugal Pablo Noriega Intitut d’Investigació en Intel ligència Artificial (IIIA), Consejo Superior de Investigaciones Científicas (CSIC), Barcelona, Spain Eugénio Oliveira Faculdade de Engenharia, Departamento de Engenharia Informática, Universidade Porto, Porto, Portugal LIACC – Laboratrio de Inteligência Artificial e Ciência de Computadores, Porto, Portugal Julian Padget Department of Computer Science, University of Bath, Bath, UK Jeremy Pitt Electrical & Electronic Engineering Department, Imperial College London, London, UK Alessandro Ricci DEIS, Alma Mater Studiorum – Università di Bologna, Cesena (FC), Italy Régis Riveret Electrical & Electronic Engineering Department, Imperial College London, London, UK Ana Paula Rocha Faculdade de Engenharia, Departamento de Engenharia Informática, Universidade Porto, Porto, Portugal LIACC – Laboratrio de Inteligência Artificial e Ciência de Computadores, Porto, Portugal Marek Sergot Department of Computing, Imperial College London, London, UK Joana Urbano LIACC – Laboratrio de Inteligência Artificial e Ciência de Computadores, Porto, Portugal Soledad Valero Universitat Politècnica de Valencia, Valencia, Spain Wamberto Vasconcelos University of Aberdeen, Aberdeen, UK Javier Vázquez-Salceda Universitat Politècnica de Catalunya, Barcelona, Spain Part I Preliminaries 12 Application Domains 261 Hiel, M., H Aldewereld, and F Dignum 2011 Modeling warehouse logistics using agent organizations In Collaborative agents – Research and development Vol 6066 of Lecture notes in computer science, ed C Guttmann, F Dignum, and M Georgeff, 14–30 Berlin/Heidelberg: Springer doi:10.1007/978-3-642-22427-0_2, http://dx.doi.org/10.1007/978-3-642-22427-0_2 Jensen, A.S., J.S Spurkeland, and J Villadsen 2013 Formalizing theatrical performances using multi-agent organizations In Twelfth Scandinavian conference on artificial intelligence (SCAI 2013), Aalborg, 20–22 Nov 2013 Vol 257 of Frontiers in artificial intelligence and applications, ed M Jaeger, T.D Nielsen, and P Viappiani, 135–144 IOS Press doi:10.3233/978-161499-330-8-135, http://dx.doi.org/10.3233/978-1-61499-330-8-135 Jiang, J., H Aldewereld, V Dignum, and Y Tan 2015 Compliance checking of organizational interactions ACM Transactions on Management Information Systems 5(4): 23:1–23:24 doi:10.1145/2629630, http://doi.acm.org/10.1145/2629630 Jiang, J., V Dignum, and Y Tan 2011 An agent-based inter-organizational collaboration framework: Opera+ In Coordination, organizations, institutions, and norms in agent system VII, COIN 2011 international workshops, COIN@AAMAS 2011, Taipei, May 2011; COIN@WIIAT 2011, Lyon, Aug 2011 Revised selected papers, Lecture notes in computer science, ed S Cranefield, M.B van Riemsdijk, J Vázquez-Salceda, and P Noriega, 58–74 Springer doi:10.1007/978-3-642-35545-5_4, http://dx.doi.org/10.1007/978-3-642-35545-5_4 de Jonge, D., and C Sierra 2015 SIMPLE: A language for the specification of protocols, similar to natural language In Proceedings of the XIXth international workshop on coordination, organizations, institutions and norms in multiagent systems (COIN@AAMAS2015), Istanbul King, T.C., T Li, M.D Vos, V Dignum, C.M Jonker, J Padget, and M.B van Riemsdijk 2015 A framework for institutions governing institutions In Proceedings of the 2015 international conference on autonomous agents and multiagent systems (AAMAS 2015), Istanbul, 4–8 May 2015, ed G Weiss, P Yolum, R.H Bordini, and E Elkind, 473–481 ACM http://dl.acm.org/ citation.cfm?id=2772940 Koster, A., J Madrenas, N Osman, M Schorlemmer, J Sabater-Mir, C Sierra, D de Jonge, A Fabregues, J Puyol-Gruart, and P García 2012 u-help: Supporting helpful communities with information technology In Proceedings of the first international conference on agreement technologies (AT 2012), Dubrovnik, vol 918, 378–392 Lee, J 2015 Norm awareness for virtual characters behaviour: A socio-cognitive approach Ph.D thesis, University of Bath http://opus.bath.ac.uk/47059/ Retrieved 2015-10-07 Lee, J., T Li, and J Padget 2013 Towards polite virtual agents using social reasoning techniques Computer Animation and Virtual Worlds 24(3–4): 335–343 doi:10.1002/cav.1517, http://dx doi.org/10.1002/cav.1517 Lee, J., J Padget, B Logan, D Dybalova, and N Alechina 2014 N-Jason: Run-time norm compliance in AgentSpeak(L) In Engineering multi-agent systems – Second international workshop (EMAS 2014), Paris, 5–6 May 2014 Revised selected papers, Vol 8758 of Lecture notes in computer science, ed F Dalpiaz, J Dix, and M.B van Riemsdijk, 367–387 Springer doi:10.1007/978-3-319-14484-9_19, http://dx.doi.org/10.1007/978-3-319-14484-9_19 Li, T 2014 Normative conflict detection and resolution in cooperating institutions Ph.D thesis, University of Bath http://opus.bath.ac.uk/45254/ Retrieved 2015-10-07 Li, T., T Balke, M De Vos, J.A Padget, and K Satoh 2013 A model-based approach to the automatic revision of secondary legislation In International conference on artificial intelligence and law, ed E Francesconi, and B Verheij, 202–206 New York: ACM Li, T., T Balke, M.D Vos, J Padget, and K Satoh 2013 Legal conflict detection in interacting legal systems In JURIX Vol 259 of Frontiers in artificial intelligence and applications, ed K.D Ashley, 107–116 Amsterdam: IOS Press Lopes Cardoso, H., J Urbano, A.P Rocha, A.J Castro, and E Oliveira 2013 ANTE: Agreement negotiation in normative and trust-enabled environments In Agreement technologies Vol of Law, governance and technology series, ed S Ossowski, 549–564 Dordrecht: Springer doi:10.1007/978-94-007-5583-3_32, http://dx.doi.org/10.1007/978-94-007-5583-3_32 262 J Padget et al Lopes Cardoso, H., J Urbano, A.P Rocha, A.J.M Castro, and E Oliveira 2016 ANTE a framework integrating negotiation, norms and trust In Social coordination frameworks for social technical systems Vol 30 of Law, governance and technology series, ed Huib Aldewereld, Olivier Boissier, Virginia Dignum, Pablo Noriega, and Julian Padget Cham: Springer Luke, S., C Cioffi-Revilla, L Panait, K Sullivan, and G Balan 2005 Mason: A multiagent simulation environment Simulation 81(7): 517–527 Mensonides, M., B Huisman, and V Dignum 2008 Towards agent-based scenario development for strategic decision support In Agent-oriented information systems IV, 53–72 Berlin/New York: Springer Montero, R., D de la Cruz, and P Noriega 2013 Prototipo de una plataforma de negociación on-line para el mercado de residuos tr-iiia-2013-02 Technical report, IIIA – CSIC, Barcelona Noriega, P., and D de Jonge 2016 Electronic institutions The EI/EIDE framework In Social coordination frameworks for social technical systems Vol 30 of Law, governance and technology series, ed Huib Aldewereld, Olivier Boissier, Virginia Dignum, Pablo Noriega, and Julian Padget Cham: Springer North, D.C 1994 Institutions, institutional change and economic performance Cambridge: Cambridge university press Padget, J., E.E Elakehal, T Li, and M.D Vos 2016 InstAL: An institutional action language In Social coordination frameworks for social technical systems Vol 30 of Law, governance and technology series, ed Huib Aldewereld, Olivier Boissier, Virginia Dignum, Pablo Noriega, and Julian Padget Cham: Springer Padget, J., E.E Elakehal, K Satoh, and F Ishikawa 2014 On requirements representation and reasoning using answer set programming In IEEE 1st international workshop on artificial intelligence for requirements engineering (AIRE 2014), Karlskrona, 26 Aug 2014, ed N Bencomo, J Cleland-Huang, J Guo, and R Harrison, 35–42 IEEE doi:10.1109/AIRE.2014.6894854, http://dx.doi.org/10.1109/AIRE.2014.6894854 Pavón, J., J Gómez-Sanz, A Fernández-Caballero, and J.J Valencia-Jiménez 2007 Development of intelligent multisensor surveillance systems with agents Robotics and Autonomous Systems 55(12): 892–903 Pavón, J., C Sansores, and J.J Gómez-Sanz 2006 Agent based modeling of social systems In EUMAS, Lisbon Pavón, J., C Sansores, and J.J Gómez-Sanz 2008 Modelling and simulation of social systems with INGENIAS International Journal of Agent-Oriented Software Engineering 2(2): 196– 221 Persson, C., G Picard, F Ramparany, and O Boissier 2012 A JaCaMo-based governance of machine-to-machine systems In Advances on practical applications of agents and multiagent systems, proceedings of the 10th international conference on practical applications of agents and multi-agent systems (PAAMS 12) Vol 155 of Advances in soft computing series, ed Y Demazeau, J.P Müller, J.M.C Rodríguez, and J.B Pérez, 161–168 Springer doi:10.1007/978-3-642-28786-2_18, http://dx.doi.org/10.1007/978-3-642-28786-2_18 Picard, G., C Persson, O Boissier, and F Ramparany 2015 Multi-agent self-organization and reorganization to adapt M2M infrastructures In Ninth IEEE international conference on selfadaptive and self-organizing systems (SASO’15), Cambridge IEEE Computer Society Pieters, W., J Padget, F Dechesne, V Dignum, and H Aldewereld 2015 Effectiveness of qualitative and quantitative security obligations Journal of Information Security and Applications 22: 3–16 doi:http://dx.doi.org/10.1016/j.jisa.2014.07.003, http://www.sciencedirect.com/science/ article/pii/S2214212614000805 Pitt, J., L Kamara, M.J Sergot, and A Artikis 2006 Voting in multi-agent systems Computer Journal 49(2): 156–170 doi:10.1093/comjnl/bxh164, http://dx.doi.org/10.1093/comjnl/ bxh164 Quillinan, T.B., F Brazier, H Aldewereld, F Dignum, V Dignum, L Penserini, and N Wijngaards 2009 Developing agent-based organizational models for crisis management In Proceedings of the 8th international joint conference on autonomous agents and multi-agent systems (AAMAS 2009), Budapest, 45–51 12 Application Domains 263 Robles, A., P Noriega, and F Cantú 2009 An agent oriented hotel information system In Proceedings of 8th international conference on autonomous agents and multiagent systems (AAMAS’09), Budapest, ed K.S Decker, J.S Sichman, C Sierra, and C Castelfranchi, 1415– 1416 International Foundation for Autonomous Agents and Multiagent Systems ACM Press Robles, A., P Noriega, M Luck, F Cantú, and F Rodriguez 2006 Multi agent approach for the representation and execution of medical protocols In 4th workshop on agents applied in health care (ECAI 2006), Riva del Garda, ed A Moreno, U Cortés, R Annicchiarico, and J Nealon, 11–16 Rodríguez, J.A., P Noriega, C Sierra, J Padget 1997 FM96.5 A java-based electronic auction house In Proceedings of 2nd conference on practical applications of intelligent agents and multiagent technology (PAAM’97), London, 207–224 ISBN 0-9525554-6-8 Roloff, M., M Stemmer, J.F Hübner, R Schmitt, T Pfeifer, and G Hüttermann A multi-agent system for the production control of printed circuit boards using JaCaMo and Prometheus AEOlus In Proceedings of 12th IEEE international conference on industrial informatics (INDIN 2014), Porto Alegre, 236–241 IEEE doi:10.1109/INDIN.2014.6945514 Sánchez-Anguix, V., S Esparcia, E Argente, A Garca-Fornes, and V Julian 2010 Collaborative information extraction for adaptive recommendations in a multiagent tourism recommender system In 8th international conference on practical applications of agents and multi-agent systems (PAAMS 2010), Salamanca, vol 70, 35–40 Springer Scott, W 1995 Institutions and organizations Thousand Oaks: Sage Publications Serrano, E., J.J Gómez-Sanz, J.A Botía, and J Pavón 2009 Intelligent data analysis applied to debug complex software systems Neurocomputing 72(13): 2785–2795 Sorici, A., G Picard, O Boissier, A Santi, and J Hübner 2012 Multi-agent oriented reorganisation within the JaCaMo infrastructure In The 3rd international workshop on infrastructures and tools for multiagent systems (ITMAS 2012), Valencia Thompson, M., J Padget, and S Battle 2015 Governing narrative events with institutional norms In 6th workshop on computational models of narrative (CMN 2015) Vol 45 of OpenAccess series in informatics (OASIcs), ed M.A Finlayson, B Miller, A Lieto, and R Ronfard, 142–151 Dagstuhl: Schloss Dagstuhl–Leibniz-Zentrum fuer Informatik doi:http://dx.doi.org/10.4230/OASIcs.CMN.2015.142, http://drops.dagstuhl.de/opus/volltexte/ 2015/5288 Toledo, C.M., R.H Bordini, O Chiotti, and M.R Galli 2011 Developing a knowledge management multi-agent system using JaCaMo In ProMAS, Taipei, 41–57 Trescak, T., M Esteva, and I Rodriguez 2010 A virtual world grammar for automatic generation of virtual worlds The Visual Computer 26(6): 521–531 doi:10.1007/s00371-010-0473-7, http://dx.doi.org/10.1007/s00371-010-0473-7 Westra, J., F Dignum, and V Dignum 2010 Keeping the trainee on track In 2010 IEEE symposium on computational intelligence and games (CIG), Copenhagen, 450–457 IEEE Westra, J., H van Hasselt, F Dignum, and V Dignum 2009 Adaptive serious games using agent organizations In Agents for games and simulations, 206–220 Berlin/New York: Springer Wooldridge, M 2009 An introduction to multiagent systems Chichester: John Wiley & Sons Yaich, R., O Boissier, G Picard, and P Jaillon 2013 Adaptiveness and social-compliance in trust management within virtual communities Web Intelligence and Agent Systems 11(4): 315–338 Chapter 13 Challenges for M4SC Julian Padget, Huib Aldewereld, and Wamberto Vasconcelos 13.1 Introduction So far, this book has covered the motivation for using models of social coordination (M4SC), the different models available for specifying M4SC, a first attempt at unifying the various models, and an overview of the domains of application of M4SC We observe a distinct advantage arising from the use of M4SC in domains that intrinsically have institutional and organisational aspects, and whose requirements include regulation, opacity of actors, and independent aims of participants There are, however, still many issues in the application of M4SC, ranging from the shortterm that might be resolved by connection to and adoption of technology from complementary research fields (like Semantic Web), to the long-term which includes the societal and industrial uptake of the frameworks for M4SC In this chapter, we wrap up the discussion on M4SC with a presentation of open issues and future perspectives that M4SC research must address and resolve to be successful in the near and more distant future J Padget ( ) Department of Computer Science, University of Bath, Bath, UK e-mail: j.a.padget@bath.ac.uk H Aldewereld Delft University of Technology, Delft, The Netherlands e-mail: h.m.aldewereld@tudelft.nl W Vasconcelos University of Aberdeen, Aberdeen, UK e-mail: w.w.vasconcelos@abdn.ac.uk © Springer International Publishing Switzerland 2016 H Aldewereld et al (eds.), Social Coordination Frameworks for Social Technical Systems, Law, Governance and Technology Series 30, DOI 10.1007/978-3-319-33570-4_13 265 266 J Padget et al 13.2 Open Issues In identifying open issues, we are attempting to put forward both intra- and interdisciplinary perspectives on what we perceive as shortcomings that can be tackled through technical means (in contrast to Sect 13.3, which considers broader nontechnical aspects) In doing so, our aim is to construct a comprehensive, but not exhaustive, list of sometimes complementary, sometimes overlapping points of view, that illustrate both the range but also the inter-dependence of the issues we cite and indicate how we imagine progress on one can help contribute to others 13.2.1 Semantic Integration The focus, as is fairly clear from the descriptions of all the frameworks (see Chaps 3, 4, 5, 6, 7, 8, 9, and 10), of research on M4SC over the last 10+ years, has been on how to formalize norms through mathematical theories informed by the ideas and principles originating from the humanities (Harré and Secord 1972; Weber 1978) and logic (Alchourrón and Bulygin 1971; von Wright 1951), with the explicit intention of enabling the construction of computational models to support programmatic reasoning about normative positions These models are now reaching a stage of maturity that allows them to be used on less artificial problems, but they remain symbolic models without any semantic association for the symbols they manipulate Thus, an open issue is how to create and maintain connections between the computational components and semantic representations of the concepts they seek to model as a necessary step on the way to establishing explainability, accessibility, and, eventually, correspondence with legal frameworks Technology from the field of Semantic Web might be leveraged to help in solving this issue 13.2.2 Semantic Interoperability Next to the issue of Semantic integration above, interoperability is another open issue that could benefit from Semantic Web technologies Any (social) interaction among actors in a system (be they human actors, software actors, or mixed populations of those) requires a common, or shared, notion of the environment and the concepts used in the organization and regulation of that environment (i.e., concepts used in the M4SC, like ‘role’, ‘obligation’, ‘power’, ‘duty’, etc.) Often, the difficulty of ensuring shared concepts is side-stepped by making the assumption that either all actors in the system are designed only for the purpose of that system, or that the actors have the necessary mechanisms to create the required links between their internal ontology and that of the society The former leads to issues of reusability: 13 Challenges for M4SC 267 agents designed for a system defined in framework A cannot operate in a system defined in framework B, even if the domain and objectives are the same The latter assumption leads to issues of semantic integration as mentioned in the previous subsection: research on agent reasoning has largely been focussed on computational models and components, side-stepping issues of the semantic association of the symbols Both issues, however, require a common ontology describing what the M4SC means with the concepts used in the definition of the system The metamodel for M4SC, as presented in Chap 2, can be used as a first step to creating such an ontology (in a unified way) 13.2.3 Accessibility to Non-specialists Socio-technical systems (Pasmore 1988) are fast becoming part of everyone’s dayto-day life We envisage a not-too-far future when people, with little or no training, will be involved in the design, implementation and operation of such systems or components thereof Approaches, methodologies, and tools which currently only cater for experts should thus be adapted and extended to cater for nonspecialists Research on M4SC is in a privileged position in this respect, with explicit representations of meta-models which can be refined and extended to cater for non-experts An important challenge, however, lies in offering means to customise, in a rational and disciplined fashion, general-purpose models, methods and tools, to suit (and adapt to) the (changing) needs and skills of potential users More sophisticated life-cycles should arise whereby users’ learning experience will inform how the model, method or tool should be changed to better support nonexperts in their various activities 13.2.4 Verification of Models/Systems Verification is typically associated with safety-critical systems, where there are significant consequences for incorrect or untimely actions and responses, and where verification means cast-iron guarantees that the code, typically written in a procedural language, will never violate (design) assertions The situation in M4SC is one step removed, in that decision-making and action selection are purposely delegated to (autonomous) software entities and are driven by environmental percepts that may correspond to brute (Searle 1995) events or to normative positions, depending on the sophistication of the agent In classical safety-critical implementations, the sequences of actions are hard-coded (design) decisions based on a pre-determined set of environmental states, and it is those that are verified In contrast, in (normative) practical reasoning systems, while the actions can be verified individually, the sequences cannot because they are determined dynamically and are arbitrarily long Although the set of possible sequences may be large, 268 J Padget et al their analysis offers the benefit of bringing verification closer to validation, since the ‘good’ and the ‘bad’ classes of sequences can be directly associated with requirements Furthermore, the reasoning process that leads to the selection of a particular action sequence can feed into a justification and explanation process: a notion that is taken up in the following section It is these two aspects that have the potential to enable the creation of systems that are safe – despite the delegation of autonomy – and trustable – because they have the data available to explain what they and why Thus, the challenge is how to model ‘precisely enough’ (Beck 2000), across a range of levels within M4SC, to allow static (design time) and continuous (run-time) validation and be able to adapt, in the latter, to (dynamically) changing scrutability requirements 13.2.5 Explainability A largely unexplored issue in complex distributed systems concerns how to explain their behaviours/outcomes to stakeholders These systems may involve hundreds or thousands of components, interacting in sophisticated ways and exchanging vasts amounts of information and knowledge as they pursue individual and societal goals We need to offer means for designers, implementors and operators to ask for explanations as to why the system (or any of its components) behaved in a particular way, as well as being able to query the alternatives and asking for reasons as to why such alternatives were not pursued Initial investigation on the topic (Caminada et al 2014; Tintarev and Kutlak 2014) has unveiled many important computational and cognitive issues when attempting to provide explanations in natural language, using formal argumentation techniques, using distributed plans Even more challenging is the issue of providing explanations about the system design (rather than its execution) as a whole and what alternatives there were Opacity in computational solutions may hinder its uptake as stakeholders have little trust in them Increasing their transparency, via explanations, may counter this Opacity, as we have noted in Chap 12, is an essential characteristic in M4SC, but at the same time, black boxes impede the establishment of trust, since only their actions can be observed Hence the incorporation of explanation provides a mechanism for a form of transparency that should be able to provide sufficient justification to counter negative perceptions, while retaining actors’ necessary privacy 13.2.6 Timeliness/Efficiency Of paramount importance to time-critical solutions such as, for instance, air traffic control systems, systems to support military and emergency relief operations, and so on, are guarantees about the timeliness of behaviours and/or outputs of individual components and the overall system Other domains or applications which are not 13 Challenges for M4SC 269 time-critical may still have constraints or expectations as to how long it takes for behaviours or outputs to be reached: most forms of encounters or interactions have expectations about when moves should be made or messages should be sent M4SC frameworks have not been designed with this requirement in mind The frameworks presented in this book are all intentionally very expressive, allowing them to model various different systems, domains and situations Such representational richness comes at a cost: associated mechanisms to reason with and about the represented knowledge are computationally costly, and efficiency has not been factored in This is evident in that most frameworks use sophisticated modal logics (e.g., deontic logic, temporal logic, etc) to represent knowledge formally, and such modal logics have undesirable computational properties such as semi-decidability and exponential complexity (Hemaspaandra and Schnoor 2008) It remains an open issue to explore techniques such as anytime computing (Zilberstein 1996) or practical reasoning (Meneguzzi et al 2015) within M4SC, to provide solutions meeting timeliness and efficiency requirements 13.2.7 Connection to Legal Frameworks This aspect links with verification, openness, semantic integration and explainability, as well as addressing the broader agenda of the creation of trust in complex systems, by seeking to represent elements of the situating legal framework of a M4SC platform in terms that software actors may reason about and that can be monitored for compliance From a technical point of view this appears in principle similar to the verification problem, once the necessary features of the legal framework have been captured in a suitable form for mechanised reasoning; the difficulty lies in the extraction of legal knowledge from natural language legal text and its accurate translation into machine-processable form Furthermore, since laws change (Li et al 2013a), this knowledge transfer process needs to be reliably repeatable and verifiable If this were achieved, it would make a considerable contribution to openness and explainability of such systems because the legal framework and its formal representation becomes a reference for system behaviour and at the same time makes such systems accountable through the situating legal context and hence, makes legislators accountable for creating appropriate laws for the governance of mixed systems (see also discussion of ethics below) An additional significant challenge is how to resolve the matter of applicable jurisdiction (or jurisdictions) governing a sequence of actions (Li et al 2013b) 13.2.8 Openness of Systems The combination of the essential characteristics of M4SC models regarding explicit regulation and opacity of actors (perhaps combined with the non-essential char- 270 J Padget et al acteristic of hybrid, technology-agnostic systems) leads to a vision of systems which any agent can join For instance, in virtual marketplaces, it is unknown at design time who the participants will be, yet the market, through its structure and regulations, performs just as well, regardless of the (type of) participants This idea of open systems (Arcos et al 2005), however, is more utopian than at first glance As demonstrated in Dignum et al (2007), openness of systems have much more stringent requirements than just the existence of (external) regulations, opacity of actors, and technology-agnostic agency The problem lies in that open systems make the assumption that any agent that joins: (i) shares a common language with other participants in the system to facilitate interactions, i.e., a shared ontology (see issue 13.2.2); (ii) has the ability to read and make sense of the regulatory specification, i.e., awareness of the ‘rules of the game’ or organisation-awareness; and (iii) has sufficient reasoning capabilities to interpret and understand the institutional state of the system and the normative consequence(s) of actions (its own, and those of the other actors involved), i.e., normative reasoning capabilities or norm-awareness Limitations in one or another of these required characteristics can be circumvented by an appropriate definition of regulation, structure of the system, or imposed mechanisms (for assistance) Examples of these are the use of governors (e.g., see Esteva et al 2004) to assist ignorant or less capable agents to perform their tasks in an appropriate fashion (from the system’s point of view) However, the success of such measures in the design is questionable, and whether a system can be truly considered to be ‘open’ is a nontechnical issue that we address in the next section 13.3 Future Perspectives In contrast to the issues presented above, some issues with M4SC involve nontechnical, or long-term, investments In this section, we look at those issues that could make M4SC successful on the long term, where issues like industrial uptake and methodological approaches play pivotal roles 13.3.1 Open Systems? In Sect 13.2.8 above we discussed the issue of open systems, describing the possible (technical) solutions required to make such systems closer to reality There is a larger philosophical issue at stake, however, concerning when a system can indeed be considered as open That is to say, how many mechanisms or how much information to assist participants can a system have, while still being considered open, or – flipping the perspective to the agents that want to participate – how much of a system should an agent understand to be able to operate within it Compared to human actors in real-world system – when joining a new organisation (e.g., filling 13 Challenges for M4SC 271 a position in another university), or visiting a (formally regulated) setting for the first time (e.g., an auction) – humans need to settle in just as well, getting to know the ‘rules’ that govern that system However, humans are also assisted by social conventions (or shared strategies (Ostrom 1990)) to bootstrap their understanding of how the system works (i.e., ‘do as most people do’) Even more, in most situations, people interact solely based on the shared strategies active in the system, rather than having a full conception of the formal rules of interaction of the system The notions of social conventions and shared strategies have not yet been applied to artificial systems in such a way so as to make these open 13.3.2 Value-Centred Design During system design many choices need to be made at higher levels of detail, which shape the nature of the resulting system The reasoning underlying each decision is ultimately based on abstract organisational values, like, e.g., integrity, trust, security, or fairness A failure to comply with such values may lead to resistance of the introduction of the system by the organisations or the society (van den Hoven 2007) Values manifest themselves at different levels of applications: e.g., in policy analysis or collective decision making, the use of values leads to a better understanding of the individual motives, thus allowing a more comprehensive (participatory) decision making process In human-machine interactions, e.g., in mixed populations of human and software agents (as in, for instance, virtual environments) or in situations where agents act on behalf of humans (as in, for instance, online markets), the use of values plays an important role in creating the required level of acceptance for the system: e.g., if a market is deemed ‘unfair’, people, or their agents, would prefer to avoid it A clear conception of what values play a role and how they affect the system is thus required M4SC, with its focus on multi-stakeholder problems, its connection to underlying legal frameworks (see Sect 13.2.7), and ability to model social norms, provides the right ingredients to model such value-sensitive issues and might assist in the design of value-sensitive systems 13.3.3 Engineering/Methodologies There are many existing environments, tools and functionalities to support software engineering activities To be adopted, these artifacts need to be accompanied by methodologies which provide guidelines on how best to use them Some of M4SC approaches in this volume come with associated methodologies (e.g., ROMAS (García et al 2015) and INGENIAS (Pavón and Gómez-Sanz 2003; Pavón et al 2005)), thus providing support and “entry points” for adopters Without methodologies, adoption is hampered as it may involve a very steep learning curve for which many companies and businesses may not have resources M4SC 272 J Padget et al methodologies also need to cater for a variety of distinct stakeholders (e.g., domain experts, technologists, human and software agents, and so on) which come together in socio-technical systems Additionally, the methodologies need to define novel software development life-cycles (e.g., requirements, as captured via the explicit representation of norms, that are easily changed and that these changes can be supported) with potential changes in the order in which tasks are done (e.g., the monitoring of the system execution which can lead to changes in requirements – represented as norms – and how this can be automated (Morales et al 2015)) Most importantly, M4SC methodologies should ideally be technology-agnostic, but with the means to integrate disparate technologies in hardware and software 13.3.4 Commercial/Industrial Uptake Development of M4SC has so far, understandably, remained an academic endeavour Uptake and use of M4SC frameworks, by industry and society, has been occurring only slowly through collaborative projects between universities and industry, for example: • The ALIVE project (FP7-215890: Coordination, Organisation and Model Driven Approaches for Dynamic, Flexible, Robust Software and Services Engineering, 2008–2010), which involved Thales, as well as two SMEs; publications Aldewereld et al (2010), Nieves et al (2011), and Quillinan et al (2009) • The MIRED-CON (Renewable distributed microgeneration/minigeneration and its control) project (MINECO (Ministry of Economy and Competitiveness) and FEDER (European Regional Development Fund), 2012–2014), which was lead by smart-grid company, the ZiV group, CIEMAT (Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas) along with three universities; publications Gómez-Sanz et al (2015) • The SensCity project (FUI Minalogic: Sensors and Services for Sustainable Cities, 2011–2012),1 which involved, among others, the RED Grenoble, Webdyn, and OrangeLabs; publications Persson et al (2012) and Picard et al (2015) as well as a range of more narrowly focussed demonstrators that clearly tackle aspects of commercial and regulatory domains, such as the examples found in the application chapter of this volume, e.g.: • mWater, realised in both Magentix and EIDE, whose specification was prepared in conjunction with the Confederación Hidrográfica del Júcar (Hydrographic Federation of the Júcar River); publications García et al (2015) and Garrido et al (2013) http://www.senscity-grenoble.com/ 13 Challenges for M4SC 273 • warehouse management and control system, modelled in OperA, and developed in conjunction with VanDerLande Industries; publications Aldewereld et al (2011), Aldewereld et al (2012), and Hiel et al (2011) • the Praise musical social networking platform (FP7-318770: Practice and peRformance Analysis Inspiring Social Education, 2012–2015),2 in conjunction with Sony CSL (Paris), and its successor project PeerLearn for the study of English and musical performance, both modelled in EIDE and the latter now being trialled in conjunction with the Escola Superior de Música de Catalunya; publications Brito et al (2015) In each of the above, there was substantial academic involvement in demonstrator development, highlighting the level of knowledge and experience that is currently typically required both to work with the concepts and with the (relatively fragile) tools, as already indicated by the open issue of accessibility to non-specialists (see Sect 13.2.3) The transition to M4SC will, as with earlier software evolutions, depend upon a critical mass of developers that understand the concepts (Sect 13.2.3), a range of M4SC-friendly methodologies (Sect 13.3.3), probably as adaptations of existing approaches to aid entry, and a variety of tool support Currently, EIDE, INGENIAS, MAGENTIX, and OperA/ALIVE offer visual modelling tools, with the former three also providing visual programming environments In the cases of INGENIAS and MAGENTIX, the tools are accompanied by an methodology that integrates with the tools INGENIAS, as well as ALIVE to a minor extent, also provide automatic (model-driven) code generation tools for ease of deployment OperA/ALIVE created modelling tools for Semantic Web service integration, and monitoring tools for run-time analysis While this demonstrates a promising variety of activity, these tools are far from mainstream and only persist thanks to their supporting research teams The short to medium term challenge is how to initiate and sustain the migration and integration of M4SC concepts into mainstream methodologies and tools, so that uptake occurs organically along with their continued maintenance and development 13.4 Summary In this book, we presented work on models for social coordination over the past 10 years Social coordination, with its intrinsic institutional and organisational aspects presents a distinct advantage in the modelling and engineering of sociotechnical systems In those past 10 years, many different frameworks (see Chaps 3, 4, 5, 6, 7, 8, 9, and 10) have been proposed and developed, each with their own focus and strengths (see Chap 11 for their comparison) Recently, with the frameworks reaching sufficient maturity, the models have been successfully applied to various http://www.iiia.csic.es/praise/ 274 J Padget et al differing use-case domains, as illustrated by the many applications presented in the previous chapter (Chap 12) Research on Models for Social Coordination is not finished, however As we argued in this chapter, there are many challenges, open issues and lines of future research that still need to be tackled to bring Models for Social Coordination to the next level With the penultimate goal of having social embedding and commercial/industrial uptake (see Sects 13.3.2 and 13.3.4), a focus needs to be placed on making (i) the frameworks easier to understand and use by non-experts (see Sect 13.2.3), i.e., easier to use by any/all modellers/users, and not just by academic people who worked on the modelling tools for several years, (ii) enhance the capabilities of the frameworks with respect to aspects that are essential to the industry, such as efficiency/timeliness (see Sect 13.2.6) and verification (see Sect 13.2.4), as well as (iii) ensuring that the frameworks and tools are accompanied by methodologies to guide users as they interact with the tool with a view to engineering their socio-technical systems and components (see Sect 13.3.3) Although there are still many challenges ahead for M4SC, this chapter and, even more so, this book, illustrates that the field is very much alive and still making steady progress on solving the open issues and challenges ahead For example, the attempt at unifying the models (see Chap 2) is a step forward in solving essential Semantic interoperability issues present in the current frameworks (see Sect 13.2.2) Additionally, the examples, even though these are currently but a few, of successful co-operations between academia and industrial/commercial partners presented in the previous subsection, indicate the growing industrial interest in the field of M4SC Lastly, the framework chapters themselves (see Chaps 3, 4, 5, 6, 7, 8, 9, and 10) provide a compact presentation of the recent developments in each of the fields, as illustrated by the various (recent) publications This volume will provide readers with academic and/or industrial interests with: (i) a comprehensive survey of approaches, tools and, in some cases, methodologies of Models for Social Coordination and how these have been put to use; (ii) a comparison among the many approaches, highlighting their advantages and shortcomings; (iii) a curated list of essential references for further reading, and our hope is that readers join us in this exciting area of research and development References Alchourrón, C.E., and E Bulygin 1971 Normative systems Vienna/New York: Springer Aldewereld, H., J Padget, W Vasconcelos, J Vázquez-Salceda, P Sergeant, and A Staikopoulos 2010 Adaptable, organization-aware, service-oriented computing IEEE Intelligent Systems 25(4): 26–35 Aldewereld, H., F Dignum, and M Hiel 2011 Re-organization in warehouse management systems In Proceedings of the IJCAI 2011 workshop on artificial intelligence and logistics (AILog-2011), 67–72 Barcelona Aldewereld, H., F Dignum, and M Hiel 2012 Decentralised warehouse control through agent organisations In Automation in warehouse development, ed R Hamberg and J Verriet, 33–44 London: Springer 13 Challenges for M4SC 275 Arcos, J.L., M Esteva, P Noriega, J.A Rodríguez-Aguilar, and C Sierra 2005 Engineering open environments with electronic institutions Engineering Applications of Artificial Intelligence 18(2): 191–204 Beck, K 2000 Extreme programming explained: Embrace change Reading: Addison-Wesley Professional Brito, I., P Gutierrez, K Hazelden, D De Jonge, L Lemus, N Osman, B Rosell, C Sierra, and C Roig 2015 Music learning with massive open online courses (MOOCs) In The future of learning, ed L Steels, vol 6, chap 10: Collaborative peer assessment with PeerLearn, 145– 174 IOS Press ISBN online 978-1-61499-593-7 Caminada, M.W., R Kutlak, N Oren, and W.W Vasconcelos 2014 Scrutable plan enactment via argumentation and natural language generation In Proceedings of the 13th international conference on autonomous agents and multi-agent systems (AAMAS’14), 1625–1626 Richland: International Foundation for Autonomous Agents and Multiagent Systems Dignum, F., V Dignum, J Thangarajah, L Padgham, and M Winikoff 2007 Open agent systems? In Agent-oriented software engineering VIII, 8th international workshop (AOSE 2007), Honolulu, ed M Luck and L Padgham, 14 May 2007 Revised selected papers, Vol 4951 of Lecture notes in computer science, 73–87 Springer Esteva, M., B Rosell, J.A Rodriguez-Aguilar, and J.L Arcos 2004 Ameli: An agent-based middleware for electronic institutions In Proceedings of the third international joint conference on autonomous agents and multiagent systems, New York, vol 1, 236–243 IEEE Computer Society García, E., A Giret, and V Botti 2015 Case studies In Regulated open multi-agent systems (ROMAS), Computational intelligence and complexity, ed E García, A Giret, and V Botti, Chap 9, 129–139 Springer García, E., A Giret, and V Botti 2015 ROMAS methodology In Regulated open multi-agent systems (ROMAS), ed E García, A Giret, and V Botti, Chap 6, 51–95 Springer Garrido, A., A Giret, V Botti, P Noriega 2013 mWater, a case study for modeling virtual markets In Agreement technologies, ed S Ossowski, 565–582 Dordrecht: Springer Gómez-Sanz, J.J., N Cuartero-Soler, and S Garcia-Rodriguez 2015 A testbed for agent oriented smart grid implementation In Engineering multi-agent systems Vol 9318 of Lecture notes of artificial intelligence, ed M Baldoni, L Baresi, and M Dastani, 92–108 Cham: Springer Harré, R., and P Secord 1972 The explanation of social behaviour Blackwells ISBN 631 14220 Hemaspaandra, E., and H Schnoor 2008 On the complexity of elementary modal logics CoRR abs/0802.1884 Hiel, M., H Aldewereld, and F Dignum 2011 Modeling warehouse logistics using agent organizations In Collaborative agents – Research and development Vol 6066 of Lecture notes in computer science, ed C Guttmann, F Dignum, and M Georgeff, 14–30 Heidelberg/ New York: Springer Li, T., T Balke, M De Vos, J.A Padget, and K Satoh 2013a A model-based approach to the automatic revision of secondary legislation In International conference on artificial intelligence and law, Rome, 202–206 Li, T., T Balke, M.D Vos, J Padget, and K Satoh 2013b Legal conflict detection in interacting legal systems In JURIX, Bologna, 107–116 Meneguzzi, F., O Rodrigues, N Oren, W.W Vasconcelos, and M Luck 2015 BDI reasoning with normative considerations Engineering Applications of Artificial Intelligence 43: 127–146 Morales, J., M López-Sánchez, J.A Rodríguez-Aguilar, M Wooldridge, and W Vasconcelos 2015 Synthesising liberal normative systems In Proceedings of the 14th international conference on autonomous agents and multi-agent systems (AAMAS’15), ed R Bordini, E Elkind, G Weiss, and P Yolum, 433–441 Richland: International Foundation for Autonomous Agents and Multiagent Systems 276 J Padget et al Nieves, J.C., J Padget, W Vasconcelos, A Staikopoulos, O Cliffe, F Dignum, J VázquezSalceda, S Clarke, and C Reed 2011 Coordination, organisation and model driven approaches for dynamic, flexible, robust software and services engineering In Service engineering, ed D Schahram and F Li, 85–115 Wien: Springer ISBN: 978-3-7091-0414-9 Ostrom, E 1990 Governing the commons The evolutions of institutions for collective action Cambridge: Cambridge University Press Pasmore, W.A 1988 Designing effective organizations: The sociotechnical systems perspective, vol New York: John Wiley & Sons Pavón, J., and J Gómez-Sanz 2003 Agent oriented software engineering with INGENIAS In Multi-agent systems and applications III, 394–403 Berlin: Springer Pavón, J., J.J Gómez-Sanz, and R Fuentes 2005 The INGENIAS methodology and tools AgentOriented Methodologies 9: 236–276 Persson, C., G Picard, F Ramparany, and O Boissier 2012 A multi-agent based governance of machine-to-machine systems In Multi-agent systems, 205–220 Springer Picard, G., C Persson, O Boissier, and F Ramparany 2015 Multi-agent self-organization and reorganization to adapt M2M infrastructures In Ninth IEEE international conference on selfadaptive and self-organizing systems (SASO’15), Cambridge, MA.: IEEE Computer Society Quillinan, T.B., F Brazier, H Aldewereld, F Dignum, V Dignum, L Penserini, and N Wijngaards 2009 Developing agent-based organizational models for crisis management In Proceedings of the eighth international joint conference on autonomous agents and multi-agent systems (AAMAS 2009), Budapest, 45–51 International Foundation for Autonomous Agents and Multiagent Systems Searle, J.R 1995 The construction of social reality Allen Lane: The Penguin Press Tintarev, N., and R Kutlak 2014 Sassy-making decisions transparent with argumentation and natural language generation In IUI 2014 workshop on interacting with smart objects, Haifa van den Hoven, J 2007 ICT and value sensitive design In The information society: Innovation, legitimacy, ethics and democracy: In honor of Professor Jacques Berleur sj, ed J Berleur and Ph Goujon, 67–72 New York: Springer von Wright, G 1951 Deontic logic Mind 60: 1–15 Weber, M 1978 Economy and society: An outline of interpretive sociology Berkeley: University of California Press Zilberstein, S 1996 Using anytime algorithms in intelligent systems AI Magazine 17(3): 73–83 ... Conceptual Map for Social Coordination, which presents the ongoing work on the unification of the different models for social coordination into one unifying meta-model for social coordination Chapter... better systems This book introduces models for social coordination (M4SC) and their related tools that fill that gap 1.2 Positioning Models for social coordination (M4SC) apply normative, social. .. Aldewereld et al (eds.), Social Coordination Frameworks for Social Technical Systems, Law, Governance and Technology Series 30, DOI 10.1007/978-3-319-33570-4_1 H Aldewereld et al coordination is a many-faceted