Download as pdf or txt
Download as pdf or txt
You are on page 1of 70

Innovations for Community Services

22nd International Conference I4CS


2022 Delft The Netherlands June 13 15
2022 Proceedings Frank Phillipson
Visit to download the full and correct content document:
https://ebookmeta.com/product/innovations-for-community-services-22nd-internationa
l-conference-i4cs-2022-delft-the-netherlands-june-13-15-2022-proceedings-frank-phill
ipson/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Brain Informatics 15th International Conference BI 2022


Padua Italy July 15 17 2022 Proceedings Mufti Mahmud

https://ebookmeta.com/product/brain-informatics-15th-
international-conference-bi-2022-padua-italy-
july-15-17-2022-proceedings-mufti-mahmud/

Integer Programming and Combinatorial Optimization 23rd


International Conference IPCO 2022 Eindhoven The
Netherlands June 27 29 2022 Proceedings Lecture Notes
in Computer Science 13265 Karen Aardal (Editor)
https://ebookmeta.com/product/integer-programming-and-
combinatorial-optimization-23rd-international-conference-
ipco-2022-eindhoven-the-netherlands-june-27-29-2022-proceedings-
lecture-notes-in-computer-science-13265-karen-aardal/

Social Informatics 13th International Conference


SocInfo 2022 Glasgow UK October 19 21 2022 Proceedings
Frank Hopfgartner

https://ebookmeta.com/product/social-informatics-13th-
international-conference-socinfo-2022-glasgow-uk-
october-19-21-2022-proceedings-frank-hopfgartner/

Intelligent Tutoring Systems 18th International


Conference ITS 2022 Bucharest Romania June 29 July 1
2022 Proceedings Scott Crossley

https://ebookmeta.com/product/intelligent-tutoring-systems-18th-
international-conference-its-2022-bucharest-romania-
june-29-july-1-2022-proceedings-scott-crossley/
Formal Methods for Industrial Critical Systems 27th
International Conference FMICS 2022 Warsaw Poland
September 14 15 2022 Proceedings Jan Friso Groote

https://ebookmeta.com/product/formal-methods-for-industrial-
critical-systems-27th-international-conference-fmics-2022-warsaw-
poland-september-14-15-2022-proceedings-jan-friso-groote/

HCI International 2022 Posters 24th International


Conference on Human Computer Interaction HCII 2022
Virtual Event June 26 July 1 2022 Proceedings Part II
Constantine Stephanidis
https://ebookmeta.com/product/hci-
international-2022-posters-24th-international-conference-on-
human-computer-interaction-hcii-2022-virtual-event-
june-26-july-1-2022-proceedings-part-ii-constantine-stephanidis/

HCI International 2022 Posters 24th International


Conference on Human Computer Interaction HCII 2022
Virtual Event June 26 July 1 2022 Proceedings Part IV
Constantine Stephanidis
https://ebookmeta.com/product/hci-
international-2022-posters-24th-international-conference-on-
human-computer-interaction-hcii-2022-virtual-event-
june-26-july-1-2022-proceedings-part-iv-constantine-stephanidis/

HCI International 2022 Posters 24th International


Conference on Human Computer Interaction HCII 2022
Virtual Event June 26 July 1 2022 Proceedings Part I
Constantine Stephanidis
https://ebookmeta.com/product/hci-
international-2022-posters-24th-international-conference-on-
human-computer-interaction-hcii-2022-virtual-event-
june-26-july-1-2022-proceedings-part-i-constantine-stephanidis/

HCI International 2022 Posters 24th International


Conference on Human Computer Interaction HCII 2022
Virtual Event June 26 July 1 2022 Proceedings Part III
Constantine Stephanidis
https://ebookmeta.com/product/hci-
international-2022-posters-24th-international-conference-on-
human-computer-interaction-hcii-2022-virtual-event-
Frank Phillipson
Gerald Eichler
Christian Erfurth
Günter Fahrnberger (Eds.)

Communications in Computer and Information Science 1585

Innovations for
Community Services
22nd International Conference, I4CS 2022
Delft, The Netherlands, June 13–15, 2022
Proceedings
Communications
in Computer and Information Science 1585

Editorial Board Members


Joaquim Filipe
Polytechnic Institute of Setúbal, Setúbal, Portugal
Ashish Ghosh
Indian Statistical Institute, Kolkata, India
Raquel Oliveira Prates
Federal University of Minas Gerais (UFMG), Belo Horizonte, Brazil
Lizhu Zhou
Tsinghua University, Beijing, China
More information about this series at https://link.springer.com/bookseries/7899
Frank Phillipson · Gerald Eichler ·
Christian Erfurth · Günter Fahrnberger (Eds.)

Innovations for
Community Services
22nd International Conference, I4CS 2022
Delft, The Netherlands, June 13–15, 2022
Proceedings
Editors
Frank Phillipson Gerald Eichler
Netherlands Organisation for Applied Deutsche Telekom Technology & Innovation
Scientific Research Darmstadt, Germany
The Hague, The Netherlands
Günter Fahrnberger
Christian Erfurth University of Hagen
University of Applied Sciences Jena Hagen, Germany
Jena, Germany

ISSN 1865-0929 ISSN 1865-0937 (electronic)


Communications in Computer and Information Science
ISBN 978-3-031-06667-2 ISBN 978-3-031-06668-9 (eBook)
https://doi.org/10.1007/978-3-031-06668-9

© The Editor(s) (if applicable) and The Author(s), under exclusive license
to Springer Nature Switzerland AG 2022
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, expressed or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword

This year, our three-day annual event, the International Conference on Innovations
for Community Services (I4CS 2022), went ahead as planned despite the ongoing
COVID-19 pandemic. But something else was new. Celebrating its 22nd edition, for
the first time it took place in The Netherlands. We met the fellows, who had to stay at
home last year, at Delft University in person, though the Organizing Committee still
prepared a hybrid event.
Twenty-one years ago, at the Technical University of Ilmenau, Germany, Herwig
Unger and Thomas Böhme called for the Workshop on Innovative Internet Community
Systems (IICS). It has continued its success story under its revised names of I2 CS
and, since 2014, I4CS. The workshop published its proceedings in Springer’s Lecture
Notes in Computer Science series (LNCS) until 2005, followed by Gesellschaft für
Informatik (GI) and Verein Deutscher Ingenieure (VDI). I4CS had commenced with
proceedings published by the Institute of Electrical and Electronics Engineers (IEEE)
before it switched back to Springer’s Communications in Computer and Information
Science (CCIS) series in 2016 and created a permanent partnership in 2018. The unique
combination of printed proceedings and the SpringerLink online edition has generated
much interest from external readers, with more than 3200 downloads since June 2021.
The selection of conference locations reflects the conference concept, and members
of the Program Committee (PC) offer suitable locations for us to live our passion.
For 2022, the I4CS Steering Committee had the honor to hand over the organization
responsibility to Frank Phillipson. Given his affiliation with TNO, The Hague, and
professorships in Delft and Maastricht, there was a choice of three beautiful Dutch
cities, with Delft coming out on top. Located close to The Hague, we were near to
the heart of quantum computing research which perfectly reflected this year’s motto
“Exploring Future Computing”.
We were proud to offer a good selection of scientific presentations, combined with
a keynote, two invited talks – among them the conference founder – and a great social
conference program to strengthen the cultural community spirit. The proceedings of
I4CS 2022 comprise 15 full and five short papers that were selected from 43 submissions,
received from authors in 13 countries. Interdisciplinary thinking is a key success factor
for any community. Hence, I4CS 2022 covered the renewed plurality of scientific,
academic, social, and industrial topics, bundled into the three key areas: Technology,
Applications, and Socialization.

Technology: Distributed Architectures and Frameworks

• Data architectures and enablers for community services,


• Block chain and secure multi-party computation,
• 5G/6G technologies and ad-hoc mobile networks,
• Data models, artificial intelligence, and big data analytics,
• Distributed and hybrid quantum computing.
vi Foreword

Applications: Communities on the Move

• Social networks, news, and mobile work,


• Open collaboration and eLearning didactics,
• Recommender solutions and context awareness,
• Augmented reality, robotics, and location-based gaming,
• Intelligent transportation, logistics, and connected cars.

Socialization: Ambient Work and Living

• Pandemic-related challenges and eHealth-assisted living,


• Smart energy and home control,
• Smart cities and municipal infrastructure,
• Internet of Things and cyber-physical systems,
• Security, identity, and data protection.

Many thanks to the 28 members of the Program Committee, representing 14 countries


worldwide, for their 152 worthwhile reviews, especially to the Program Chair Christian
Erfurth and furthermore to our Publication Chair Günter Fahrnberger, who continued
a very successful cooperation with the Springer publishing board to keep the high
reputation of I4CS, rated C-level at CORE.
Following the alternation rule, the 23rd I4CS will take place in Germany around
June 2023. Due to the requirement to hold last year’s I4CS completely online, as an
exception, the Steering Committee decided to vote again for the University of Bamberg
as host. Please check the permanent conference URL at http://www.i4cs-conference.org/
for more details.
Proposals on emerging topics as well as applications for prospective Program
Committee members along with potential conference hosts and locations are kindly
welcome at request@i4cs-conference.org.
Kind regards on behalf of the entire Steering Committee and the Editors’ Board.

June 2022 Gerald Eichler


I4CS Steering Chair
Preface

As Conference and Program Chairs, it is our pleasure to present this CCIS volume
with its unique contributions on distributed architectures and software frameworks, both
classical and quantum, and advanced applications in machine learning and cryptography
as well as services supporting digital communities and smart mobility. The related papers
were presented at the 22nd International Conference on Innovations for Community
Services (I4CS 2022), held during June 13–15, 2022, in Delft, the Netherlands.
The conference was hosted by TNO, the Netherlands organization for applied
scientific research, at the venue of Delft University of Technology. TNO connects people
and knowledge to create innovations that boost the competitive strength of industry
and the well-being of society in a sustainable way. Information and Communication
Technology (ICT) offers excellent opportunities to solve all kinds of societal challenges,
ensure innovation, and continue economic growth. It is a key area for many sectors and
other technologies. The rapid digital transformation of our society requires a framework
for action. TNO’s ambition is to help to achieve complex digital transformations by using
its knowledge of technology, policy, and business models. The connection of TNO and
Delft University of Technology here is a representation of the purpose of the conference,
which stands not only for scientific breakthroughs but also for the connection of science
to industry, giving room to present experiences and applications.
We were happy to be in Delft and have a physical event again. Delft is a popular
tourist destination in the Netherlands, famous for its historical connections with the
reigning House of Orange-Nassau, for its blue pottery, for being home to the painter
Johannes Vermeer, and for hosting Delft University of Technology. Historically, Delft
played a highly influential role in the Dutch Golden Age. In terms of science and
technology, thanks to the pioneering contributions of Antonie van Leeuwenhoek and
Martinus Beijerinck, Delft represents the birthplace of microbiology.
Currently, Delft is one of the main centers of quantum computing in the world. Again,
TNO and Delft University of Technology have joined efforts and created QuTech, whose
mission is to develop scalable prototypes of a quantum computer and an inherently safe
quantum Internet, based on the fundamental laws of quantum mechanics. Quantum
computing is only one example of the (fundamental) new computation techniques that
will change our lives. Areas such as photonic computing will grow rapidly in the coming
years. New cryptographic techniques will arise, partly assisted and forced by the rise
of the quantum computer and cryptographic applications, such as secure multi-party
computation. Artificial intelligence (AI) will continue to develop, enhanced by new
computing techniques and driving further demands for explainability. The consideration
of these trends and the effect on society was the main theme of this year’s conference,
having the title “Exploring Future Computing”.
The keynote and invited talks gave attention to these topics, and you will find a
summary of them in this volume. Thijs Veugen explained secure multi-party computation
and showed some of its applications. Herwig Unger, one of the founders of the
conference, gave his view on developments in Artificial Intelligence (AI), especially
viii Preface

on brain-inspired approaches to natural language processing and explainable AI. Darian


Meacham looked at the future of work and asked how all this new technology will
influence the way we are working. A visit to the quantum laboratories of QuTech was
an interesting excursion during the conference.
As Conference and Program Chairs, we express our gratitude to all members of
the Program Committee and all external reviewers for their dedicated service, for
maintaining the quality objectives of the conference, and for the timely provision of their
valuable reviews. We again managed to realize at least three reviews for each submitted
paper. We thank all the authors for their submitted contributions, all the speakers for
their vivid presentations, and all the participants for their contributions and interesting
discussions. We thank EasyChair and Springer for their excellent services.
We thank TNO and Delft University of Technology for hosting the conference, giving
us the opportunity to meet again after two years of restrictions due to COVID-19 in the
nice city of Delft.
It was a pleasure to work together with Gerald Eichler and Günter Fahrnberger.
Together, we have realized another version of the conference and hope the tradition lives
on.

June 2022 Frank Phillipson


Christian Erfurth
Organization

Conference Chair
Frank Phillipson TNO, The Netherlands

Steering Chair
Gerald Eichler Deutsche Telekom Technology and Innovation
Darmstadt, Germany

Program Chair
Christian Erfurth University of Applied Sciences Jena, Germany

Publication Chair
Günter Fahrnberger University of Hagen, Germany

Program Committee
Sebastian Apel Technical University of Applied Sciences
Ingolstadt, Germany
Gilbert Babin University of Montréal, Canada
Gerald Eichler Deutsche Telekom Darmstadt and Innovation
Darmstadt, Germany
Christian Erfurth University of Applied Sciences Jena, Germany
Günter Fahrnberger University of Hagen, Germany
Hacène Fouchal University of Reims Champagne-Ardenne, France
Sapna Ponaraseri Gopinathan Coimbatore Institute of Technology, India
Michal Hodoň University of Žilina, Slovakia
Mikael Johansson CSC - IT Center for Science, Finland
Kathrin Kirchner Technical University of Denmark, Denmark
Udo Krieger University of Bamberg, Germany
Peter Kropf University of Neuchâtel, Switzerland
Ulrike Lechner Bundeswehr University Munich, Germany
Andreas Lommatzsch Technical University of Berlin, Germany
Karl-Heinz Lüke Ostfalia University of Applied Sciences, Germany
Raja Natarajan Tata Institute of Fundamental Research, India
Deveeshree Nayak University of Washington Tacoma, USA
x Organization

Olli Nurmi VTT Technical Research Centre, Finland


George Papadopoulos University of Cyprus, Cyprus
Dana Petcu West University of Timisoara, Romania
Frank Phillipson TNO, The Netherlands
Siddharth Rautaray Kalinga Institute of Industrial Technology, India
Joerg Roth Nuremberg Institute of Technology, Germany
Amardeo Sarma NEC Laboratories Europe, Germany
Pranav Kumar Singh Indian Institute of Technology Guwahati and
Central Institute of Technology Kokrajhar,
India
Julian Szymanski Gdansk University of Technology, Poland
Rob van der Mei CWI and Vrije Universiteit Amsterdam, The
Netherlands
Leendert W. M. Wienhofen City of Trondheim, Norway

Additional Reviewers

Baier, Harald
Verlande, Lisa
Contents

Invited Papers

Secure Multi-party Computation and Its Applications . . . . . . . . . . . . . . . . . . . . . . . 3


Thijs Veugen

Brain-Inspired Approaches to Natural Language Processing


and Explainable Artificial Intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Erik Deussen, Herwig Unger, and Mario M. Kubek

Back to the Future of Work: Old Questions for New Technologies . . . . . . . . . . . . 11


Darian Meacham

Energy Harvesting and Environment Protection

Solar Energy Harvesting for the Mobile Robotic Platform . . . . . . . . . . . . . . . . . . . 17


Michal Hodoň, Peter Ševčík, Juraj Miček, Veronika Olešnániková,
Peter Šarafín, Ján Kapitulík, and Lukáš Čechovič

Demonstrating Feasibility of Blockchain-Driven Carbon Accounting –


A Design Study and Demonstrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Karl Seidenfad, Tobias Wagner, Razvan Hrestic, and Ulrike Lechner

Applied Security and Privacy

A Platform for Offline Voice Assistants: Development of Assistant


Applications Without Being Connected to Commercial Central Services . . . . . . . 49
Jörg Roth

Consensus Algorithms in Cryptocurrency and V2X-IoT: Preliminary Study . . . . 63


Fatima Chahal, Dominique Gaiti, and Hacène Fouchal

Realtime Risk Monitoring of SSH Brute Force Attacks . . . . . . . . . . . . . . . . . . . . . 75


Günter Fahrnberger

eHealth and Infrastructure

Emergency Evacuation Software Simulation Process for Physical Changes . . . . . 99


Dan Wu, Imran Shafiq Ahmad, and Rachit Tomar
xii Contents

Context Information Management in a Microservice Based Measurement


and Processing Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Steffen Späthe, Florian Greulich, and Sebastian Apel

A Web Architecture for E-Health Applications Supporting the Efficient


Multipath Transport of Medical Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Kibriya Inamdar, Oormila Ramanandan Kottayi Pilapprathodi,
Jopaul John, Markus Wolff, Marcel Großmann, and Udo R. Krieger

Smart Mobility and Routing

Understanding Human Mobility for Data-Driven Policy Making . . . . . . . . . . . . . . 155


Jesper Slik and Sandjai Bhulai

Research on Detecting Similarity in Trajectory Data and Possible Use Cases . . . 168
Moritz Peter Weber and Ilenia Salvadori

Misbehavior Verification on Cooperative Intelligent Transport System . . . . . . . . 185


Emilien Bourdy, Marwane Ayaida, and Hacène Fouchal

Smart Cities and Standardization

A Multi Service Capacitated Facility Location Problem with Stochastic


Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
L. J. Kim, F. Phillipson, and R. S. Wezeman

The Role of a Data Marketplace for Innovation and Value-Added Services


in Smart and Sustainable Cities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Rebekka Alvsvåg, Anthony Bokolo Jr., and Sobah Abbas Petersen

Challenges of Future Smart and Secure IoT Networking . . . . . . . . . . . . . . . . . . . . 231


Dirk Von Hugo, Gerald Eichler, and Behcet Sarikaya

Applied Machine Learning

Towards Simulating a Global Robust Model for Early Asthma Detection . . . . . . 257
Bhabesh Mali and Pranav Kumar Singh

Practical Method for Multidimensional Data Ranking:


Application for Virtual Machine Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Chérifa Boucetta, Laurent Hussenet, and Michel Herbin
Contents xiii

Quantum Computing

Distributed Quantum Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281


Niels M. P. Neumann and Robert S. Wezeman

A Quantum Annealing Approach for Solving Hard Variants of the Stable


Marriage Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Christoph Roch, David Winderl, Claudia Linnhoff-Popien,
and Sebastian Feld

Learning Based Hardware-Centric Quantum Circuit Generation . . . . . . . . . . . . . . 308


Merel A. Schalkers and Matthias Möller

A Quantum Approach for Tactical Capacity Management of Distributed


Electricity Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Frank Phillipson and Irina Chiscop

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335


Invited Papers
Secure Multi-party Computation and Its
Applications

Thijs Veugen1,2(B)
1
TNO, Anna van Buerenplein 1, The Hague, The Netherlands
thijs.veugen@tno.nl
2
CWI, Amsterdam Science Park, Amsterdam, The Netherlands

Abstract. There is an urgent need for secure data sharing solutions,


such that different organisations can jointly compute with their data,
without revealing sensitive data to each other. Secure multi-party com-
putation is an innovative cryptographic technology that has recently
become more mature, such that it can be used to obtain very secure
data sharing solutions for real life applications. Its broad usability is
illustrated by different use cases from various domains.

Keywords: Secure data sharing · Secure multi-party computation

1 Secure Data Sharing

In the current era of big data, artificial intelligence, and privacy, there is a
growing need for secure data sharing solutions [1]. This can be explained by
two conflicting developments observed within our society. The first one is the
increased availability of data from different sources, and the need to add value
to data by artificial intelligence, machine learning, and similar technologies [2].
By sharing various data sources we can improve services, mitigate risks, and
create new business opportunities.
The second development is the improved regulation for protecting the pri-
vacy of personal data, avoiding large data lakes by commercial companies [3],
and the increased awareness of people and organisations to carefully handle
privacy-sensitive data. Data can be sensitive because of privacy reasons, but
organisations are also reluctant to share commercially sensitive data. These two
developments illustrate the apparent paradox that we have created. On one hand
we want to share data and jointly compute with it to create value, but on the
other hand we do not, because this will leak sensitive data to other parties, who
might use it for malicious purposes.
This paradox shows the need for technical solutions that people and organi-
sations can use to securely compute with distributed data to jointly create value,
without revealing sensitive data to others. A suitable and very secure technology
to tackle this problem is the innovative cryptographic field of secure multi-party
computation.
c The Author(s), under exclusive license to Springer Nature Switzerland AG 2022
F. Phillipson et al. (Eds.): I4CS 2022, CCIS 1585, pp. 3–5, 2022.
https://doi.org/10.1007/978-3-031-06668-9_1
4 T. Veugen

2 Secure Multi-party Computation


First introduced in 1982 [5], secure multi-party computation is a relatively
new field within cryptography. It consists of a large collection of techniques to
tackle a specific problem. Consider multiple parties, each having their own data.
Together they want to perform a certain computation with their data, techni-
cally described as a function that given all inputs, each party having an input,
will compute an output. The most important requirement is that the parties are
not allowed to learn each others input during the computation of the function.
Although the developed solutions are very secure, this initially required large
computational efforts by the parties, and lots of (secure) communication between
them. In recent decades, new academic discoveries [6,7] led to improvements that
reduced the computational and communication effort, which has made the tech-
nology suitable for tackling real life problems. Secure multi-party computation
has proven to be a serious candidate for multiple data sharing problems in var-
ious domains.

3 Applications
We mention a couple of applications of secure multi-party computation from dif-
ferent domains. The first example is the combat of financial fraud. Anti money
laundering is a worldwide huge problem, and financial institutions are investing
a lot of effort in their attempt to tackle it. A reason that this has not led to seri-
ous fraud reductions, is that banks are not allowed to share their data. This is
especially a problem for anti money laundering, as these activities typically con-
sist of large transaction chains over various banks and countries in an attempt
to legalise the fraudulent money. We developed a solution to enable banks to
search for fraudulent accounts within virtually joint transaction networks, with-
out leaking their own transaction data [4].
Another problem arises from the fact that data are often fragmentary stored
over different organisations. A frequently arising situation is that there are two
databases containing similar items or persons, and they would like to know how
many, or even which ones, they have in common, without learning the remainder
of the other database. This problem is technically known as private, or secure
set intersection [8]. This could e.g. be used to find people that are wanted in
different countries for fraudulent actions, without revealing sensitive national
black lists.
Within the health domain, sensitive medical data is stored at various organ-
isations, like hospitals, general practitioners, pharmacies, medical research, etc.
Interesting insights could be gained by running machine learning on the dis-
tributed data, such as measuring the impact of health innovations, developing
personalised medicines, and finding the cause of diseases. As a concrete use-case,
hospital Erasmus MC and health insurance company Achmea have data on indi-
viduals in the city of Rotterdam, which would in theory enable them to train a
regression model in order to identify high-impact lifestyle factors for heart failure
[9].
Secure Multi-party Computation and Its Applications 5

4 Conclusion
We explained the urgency of finding solutions for secure data sharing, where
different organisations can jointly compute with their data, without revealing
sensitive data to each other. Secure multi-party computation is an innovative
cryptographic technology that can be used to obtain very secure data sharing
solutions. Its broad usability is illustrated by different applications from various
domains.

References
1. Attema, T., Worm, D.: Finally, a privacy-friendly way to harness data. Whitepa-
per TNO (2021). https://publications.tno.nl/publication/34637925/StFCdC/TNO-
2021-technological.pdf
2. OECD: Enhancing Access to and Sharing of Data: Reconciling Risks and Benefits
for Data Re-use across Societies. OECD Publishing, Paris (2019)
3. European Union: European Comission: Proposal for a REGULATION OF THE
EUROPEAN PARLIAMENT AND OF THE COUNCIL on European data gover-
nance (Data Governance Act), 25 November 2020, COM (2020). 767 final
4. Sangers, A., et al.: Secure multiparty PageRank algorithm for collaborative fraud
detection. In: Goldberg, I., Moore, T. (eds.) FC 2019. LNCS, vol. 11598, pp. 605–
623. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32101-7 35
5. Yao, A.C.-C.: Protocols for secure computations (extended abstract). In: 23rd
Annual Symposium on Foundations of Computer Science, Chicago, Illinois, USA,
3–5 November 1982, pp. 160–164. IEEE Computer Society (1982)
6. Damgård, I., Pastro, V., Smart, N., Zakarias, S.: Multiparty computation from some-
what homomorphic encryption. In: Safavi-Naini, R., Canetti, R. (eds.) CRYPTO
2012. LNCS, vol. 7417, pp. 643–662. Springer, Heidelberg (2012). https://doi.org/
10.1007/978-3-642-32009-5 38
7. Keller, M., Orsini, E., Scholl, P.: MASCOT: faster malicious arithmetic secure com-
putation with oblivious transfer. In: Weippl, E.R., Katzenbeisser, S., Kruegel, C.,
Myers, A.C., Halevi, S. (eds.) Proceedings of the 2016 ACM SIGSAC Conference
on Computer and Communications Security, Vienna, Austria, 24–28 October 2016,
pp. 830–842. ACM (2016)
8. Freedman, M.J., Nissim, K., Pinkas, B.: Efficient private matching and set intersec-
tion. In: Cachin, C., Camenisch, J.L. (eds.) EUROCRYPT 2004. LNCS, vol. 3027,
pp. 1–19. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-24676-3 1
9. van Egmond, M.B., Spini, G., van der Galien, O., et al.: Privacy-preserving dataset
combination and Lasso regression for healthcare predictions. BMC Med. Inform.
Decis. Mak. 21, 266 (2021)
Brain-Inspired Approaches to Natural
Language Processing and Explainable
Artificial Intelligence

Erik Deussen1 , Herwig Unger1 , and Mario M. Kubek2(B)


1
University of Hagen, Hagen, Germany
{erik.deussen,herwig.unger}@fernuni-hagen.de
2
Georgia State University, Atlanta, USA
mkubek@gsu.edu

Abstract. Like no other medium, the World Wide Web became the
major information source for many people within the last years, some
even call it the brain of mankind. For any arising questions, any facts
needed, or any multimedia content wanted, a web page providing the
respective information seems to exist. Likewise, it seems that sometimes
there is nothing that has not been thought, written, painted, or expressed
in any other form before: most users simply feel overwhelmed by the
flood of available information. Consequently, there is a need for new
technologies for autonomous self-management, more timely information
handling, processing, and the user’s interaction with such huge amounts
of data. Indeed, Einstein’s saying Look deep into nature, then you will
understand everything better is a big inspiration and challenge to find the
required, new solutions. At this point, a short overview is given of existing
organizational and functional principles, which have been derived from
nature and in particular the human brain and which could be adapted
to realize the desired, new methods for natural language processing. The
methods mostly follow the strict natural design principle of locality, i.e.
work without overseeing the whole system or full set of data, and exhibit
a high degree of parallelism. Also, specific application fields for them will
be discussed.

Keywords: Brain-inspired natural language processing · Explainable


artificial intelligence · Attention

1 Introduction
Like no other medium, the World Wide Web (WWW) became the major infor-
mation source for many people within the last years, some even call it the brain of
mankind. For any arising questions, any facts needed, or any multimedia content
wanted, a web page providing the respective information seems to exist. Likewise,
it seems that sometimes there is nothing that has not been thought, written,
painted, or expressed in any other form before: most users simply feel over-
whelmed by the flood of available information. Consequently, there is a need for
c The Author(s), under exclusive license to Springer Nature Switzerland AG 2022
F. Phillipson et al. (Eds.): I4CS 2022, CCIS 1585, pp. 6–10, 2022.
https://doi.org/10.1007/978-3-031-06668-9_2
Brain-Inspired Approaches to Natural Language Processing 7

new technologies and methods supporting autonomous self-management, more


timely information handling, processing, and the user’s interaction with such
huge amounts of data. Indeed, Einstein’s saying Look deep into nature, then you
will understand everything better is a big inspiration and challenge to find the
required, new solutions. Therefore, in the following sections, a short overview
is given of existing organizational and functional principles, which have been
derived from nature and in particular the human brain and which are currently
being adapted to realize the desired, new methods for natural language process-
ing (NLP). These methods mostly follow the natural design principle of locality,
i.e. work without overseeing the whole system or full set of data, and exhibit a
high degree of parallelism. Also, specific application fields for them will be dis-
cussed. However, today’s existing approaches to Artificial Intelligence (AI) are
not able to emulate human intelligence in the required flexibility, e.g. to trans-
fer problem solutions and strategies of one domain to another domain, i.e. to
perform a generalization. On the other hand, because they are often used as a
“black box”, they are not able to justify and explain certain decisions and classi-
fications made in a way that is comprehensible to human users. Moreover, they
are characterized by a usually long training phase for the creation of models,
which is always strictly separated from the phase of their application.

2 Information Processing in the Human Brain

However, this approach does not correspond to the way the human brain works,
where processes of learning and applying take place simultaneously and contin-
uously in the neocortex across multiple cortical levels. In the context of the the-
ory of memory-prediction framework presented in [1], these processes and their
interconnectedness are discussed. An essential core aspect here is that sensory
input sequences are stored and recognized by the cortical processing hierarchy.
To this end, recurrent sequences are encoded in an invariant form (are assigned
identifiers) and stored in the different levels of the hierarchy, which allows to
represent facts according to their degree of abstraction. Furthermore, a pre-
diction of expected sequence components (based on already stored inputs) and
their continuous comparison with new inputs takes place. Deviations trigger a
learning process in which alternative representations are captured, which in turn
trigger new predictions. Implicitly, this provides for fault tolerance and allows
for the correction of errors. The process of prediction is also an important basis
for content generation, e.g. in speech production. Furthermore, sensory stimuli
of different origins are linked in parallel on the levels of the hierarchy. In this
way, a complete – from the point of view of the individual brain – but dynami-
cally adaptable and holistic model of the (perceptible) world is created. Despite
continuous improvement and extensions of techniques based on artificial neural
networks for sequence modeling [2], for matching data of different domains [3]
by linking the models representing them and transfer learning [4], the afore-
mentioned mechanisms and processes are not represented by adequate technical
analogues. Despite their importance for natural language understanding (NLU),
8 E. Deussen et al.

the neurophysiological processes and effects occurring in the human brain such
as remembering (e.g. fan effect [5]) and forgetting [6] are largely ignored in these
technical solutions. Furthermore, neuroscientists have found out that during its
development the human brain may not only learn many new words but is also
connecting them by relations, depending on how often they have been used
together. In such a manner, even spatial clusters of frequently and jointly used
words appear; this process is called modular segregation [7] in neuroscience and
finds a corresponding model with co-occurrence graphs [8] in natural language
processing.

3 Language Acquisition and Use


The sequence learning processes addressed can be illustrated particularly well
in the context of natural language processing (analysis and generation). It is
shown in [9] that the classical generative model of linguistic knowledge [10],
which is based on the use of static and limited resources (lexicons, grammati-
cal rules), is unsuitable for this purpose. Instead, humans learn to use language
through repeated (frequency dependence should be emphasized) and situational
exposure to idiomatic turns of phrase and collocations, which allows words to
be ascribed their individual meaning by the brain along with their contextual
profile and frame of use, thus also allowing for the resolution of ambiguities. Lin-
guistic experiences are therefore stored by both semantic and episodic memory.
In the context of stimulus-response experiments [8], it was also shown that term
associations by humans correlate well with significant co-occurrences in statis-
tical corpus analysis. The mentioned learning processes must therefore incorpo-
rate these approaches of human language acquisition and use. As a side effect, a
unique representation of the world is formed depending on the presented stimuli.

4 Brain-Inspired and Continuous Information Processing


In order to make this representation as comprehensive as possible, it is necessary
to acquire as many sensory stimuli as possible and process them simultaneously
in order to strengthen the mental representation through a manifold input (not
exclusively linguistic). The core aspect here is that this modeling must always
be oriented towards use. Furthermore, this learning process must be executed
continuously, not only once or as needed, e.g. during the training phase of artifi-
cial neural networks. In this way, it is possible for the agent/computer program
executing the learning process to dynamically adjust to changing situations and
contexts. However, a filtering mechanism is necessary to support the aforemen-
tioned abstraction of the cortical processing hierarchy. The aim is to capture the
essential information from an input sequence on the different levels of the hier-
archy and to store it there in an encoded form. Current language models such
as Google BERT [4] use so-called attention mechanisms to determine the impor-
tance of different input fragments. In this respect, an alternative, graph-based
method was developed in [11], which determines the semantically influential
Brain-Inspired Approaches to Natural Language Processing 9

terms and phrases – called source topics – of a given text or corpus. Through
these, it is possible (among other things), to create topical chains of terms (even
across text boundaries) and whole texts that reveal the fundamentals of top-
ics of interest. In a natural way, this creates a kind of topic-oriented signpost
system, which is able to reveal thematic dependencies and to further use them
to find relevant foundational documents through (possibly repeated) interactive
and automated topic tracking alike. Another advantage of this approach is its
inherent explainability: topical foundations can be found by repeated backtrack-
ing, starting from source topics connected to an initial set of topics of interest.
This way, these connections serve as an explanation of the topical chain. Fur-
thermore, using those and similar procedures, an implicit extraction of relevant
features from input streams of different sensory origins is possible.

5 Applications
The previous discussions stress the importance of the technical realization and
modeling of the brain’s neurophysiological processes in modern information sys-
tems. Especially, the extraction of semantic features along with their weights
to be used in numerous tasks such as sentiment analysis as well as document
classification and clustering can be greatly improved when applying them as
continuously executed steps in a preprocessing pipeline. This way, the quality of
the results returned by downstream text mining algorithms will be significantly
improved as well. Library services are a particular important application field
that can specifically benefit from these approaches. As those services are essential
for learning and teaching, they are increasingly being integrated into learning
management systems (LMS) [12]. For instance, the services of liaison librarians,
who act as active intermediaries or contact persons and know how scientists of
a certain domain use to communicate, can be offered in these systems in order
to sustainably and actively support scientific works and research processes. To a
certain extent, the brain-inspired approaches and methods mentioned in the pre-
vious sections can on one hand support these services on a technical level and on
the other hand can take a lot of work off the librarians’ hands, especially when it
comes to analyzing, filtering, and screening large amounts of scientific contents.
Furthermore, these services can be personalized and adapted to users’ needs.
As an example, they can actively support LMS-users in their specific research
tasks by continuously providing pointers to relevant literature from the WWW,
enriching search processes with complementary topical directions or search terms
and suggesting useful scientific communities. For this purpose, large amounts of
textual resources must be automatically analyzed and classified, not only with
respect to content-related aspects, but to time-related ones, too. In doing so, a
“Librarian of the Web” [13] is created which will act autonomously and on behalf
of the LMS-users, while at the same time taking a huge leap forward towards
real information literacy or competence.
10 E. Deussen et al.

6 Conclusion
This extended abstract pointed out that current approaches for artificial intelli-
gence lack the inclusion of an adequate technical realization of neurophysiological
processes in the human brain. However, the adoption and integration of brain-
inspired methods in modern information systems will yield significantly better
processing results, e.g. when classifying textual documents and search queries.
Furthermore, they enable the creation of information literate library-services
that can proactively and individually support scientists in their research tasks.

References
1. Hawkins J., Blakeslee, S.: On Intelligence. Henry Holt and Company (2004)
2. Paaß, G., Hecker, D.: Erfassung der Bedeutung von geschriebenem text. In: Paaß,
G., Hecker, D. (eds.) Künstliche Intelligenz, pp. 167–248. Springer, Wiesbaden
(2020). https://doi.org/10.1007/978-3-658-30211-5 6
3. Antol, S., et al.: VQA: visual question answering. In: IEEE International Confer-
ence on Computer Vision (ICCV), pp. 2425–2433. IEEE Computer Society (2015)
4. Devlin, J., et al.: BERT: pre-training of deep bidirectional transformers for lan-
guage understanding. In: Proceedings of the 2019 Conference of the North Amer-
ican Chapter of the Association for Computational Linguistics: Human Language
Technologies, vol. 1, pp. 4171–4186. ACL (2019)
5. Radvansky, G.A.: The fan effect: a tale of two theories. J. Exp. Psychol.: Gener.
128(2), 198–206 (1999)
6. Deussen, E., Immelnkämper, M., Unger, H.: Oblivion in time-dependent informa-
tion management. In: Unger, H., Kubek, M. (eds.) The Autonomous Web. Studies
in Big Data, vol. 101, pp. 149–158. Springer, Cham (2022). https://doi.org/10.
1007/978-3-030-90936-9 11
7. Baum, G.L., et al.: Modular segregation of structural brain networks supports
the development of executive function in youth. Curr. Biol. 17(11), 1561–1572.e8
(2017)
8. Heyer, G., Quasthoff, U., Wittig, T.: Text Mining: Wissensrohstoff Text: Konzepte,
Algorithmen, Ergebnisse. W3L-Verlag (2006)
9. Taylor, J.R.: The Mental Corpus. How Language is Represented in the Mind.
Oxford University Press, Oxford (2012)
10. Chomsky, N.: Knowledge of Language: Its Nature, Origin, and Use. Greenwood
Publishing Group (1986)
11. Kubek, M., Unger, H.: Detecting source topics by analysing directed co-occurrence
graphs. In: 12th International Conference on Innovative Internet Community Ser-
vices (I2CS 2012), pp. 202–211. Gesellschaft für Informatik e.V., Bonn (2012)
12. Roos, M.: Mögliche Anforderungen an ein LMS aus Bibliothekssicht. eleed 14
(2021)
13. Kubek, M.: Concepts and Methods for a Librarian of the Web. Springer, Cham
(2020). https://doi.org/10.1007/978-3-030-23136-1
Back to the Future of Work: Old Questions
for New Technologies

Darian Meacham(B)

Maastricht University, Maastricht, Netherlands


d.meacham@maastrichtuniversity.nl

Abstract. In the past decade or so the Future of Work question has emerged as
a major policy concern at national and international level. This is in large part
due to opportunities and challenges created by the development of data and AI
driven automation technologies, and in the past two years by the Covid pandemic,
which has led many employees and employers to rethink the ways in which they
work, as individuals and organisations. In the USA, there is now talk of a great
resignation, as many employees reconsider the value and quality of their working
lives. If there is one lesson that we have already learned it is that the future of
work question resists easy formulations and answers, nor is it primarily a matter
of jobs being replaced by automation. As work touches nearly every aspect of
our lives the future of work is bound to be a complex question in need to careful
investigation. In this talk I won’t offer predictions, but try to unpack the problem,
asking not so much what is the future of work, but rather how should we ask good
question(s) about it in the first place.

Keyword: Future of work

The question of the “future-of-work” has in the past decade or so re-emerged as a major
field of academic and policy inquiry. Interest in this topic has spanned disciplines with
significant contributions being made in economics [1, 2], sociology [3], anthropology
[4] and philosophy [5]. On the policy side, reports from the European Commission [6]
have been published alongside studies from major consultancies [7], European agencies
such as Eurofound [8] as well as organised labour [9].
The recent storm of activity has in large part been motivated by reflection, anticipation
and consternation surrounding opportunities and challenges created by the development
of data and AI driven automation technologies. Over the past two years, the Covid-19
pandemic, has also led many employees and employers to rethink the ways in which
they work, as individuals and organisations. In the USA, there is now talk of a “great
resignation”, as many employees reconsider the value and quality of their working lives.
The question of the future-of-work has taken a partner: what is “good work”?
Taking a step back, a first question to ask is why this refocus on “work”, and why has
it sustained itself. An immediate if rather coarse answer is a concern about unknowns
when it comes to both the quality and quantity of future employment. Modern forms of
employment, what we normally call work or jobs, play a formative role in pretty much
all aspects of how our individual and collective lives are structured in. There is really

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022


F. Phillipson et al. (Eds.): I4CS 2022, CCIS 1585, pp. 11–13, 2022.
https://doi.org/10.1007/978-3-031-06668-9_3
12 D. Meacham

nothing that we do that is not somehow touched by the work that we do or do not do and
the rewards and recognition derived from that work. Employment roles have also played
a decisive part in the formation of social and political identities that have constitutive
roles in the ways that the legacy institutions of liberal democracy have functioned for
generations. In other words, there is a sense that the structures of work could be entering
a period of accelerated change. And when work changes, everything changes.
These are not new concerns. In the 1980s, the French philosopher André Gorz pointed
out that the “crisis of work” [10], in terms of both its quality and quantity, is coeval with
the emergence of the modern concept and practices of employment. The “future of work”
question and consternation have been around since work’s inception. In his seminal
paper, “Economic Possibilities for Our Grandchildren” [11] from the early 1930s, John
Maynard Keynes introduced the term “technological unemployment”, postulating that by
today productivity gains would leave citizens of technologically developed nations with
such a surplus of time that it could constitute an existential crisis, provided the question
of distribution was adequately addressed. Keynes got the problem backwards; it is the
problem of distribution within societies of relative affluence that plagues policy makers
and not a surplus of leisure time, but the conceptual framework that he introduced persists
in the current debate. Functionalist anthropologists [12] have also argued for decades
that the paradigmatic political structures of the modern age arose from the exigencies
of transformations in the material and technological structures of employment, namely
the emergence of commercial and eventually industrial economies.
It is important to revisit these old questions and debates in light of new technological
environments, and specifically the development of AI. Decades of debate in these dis-
ciplines have prepared us to have different kinds of conversations about not only what
structures are subject to change, but what we value more or less in those structures.
There are nonetheless several novel aspects of the current debate. First is the question
of whether this time is somehow different from previous waves of automation. Again,
an old question, but there is the possibility that the answer will be different this time
around. Jobs are not monolithic entities, but rather complex practices composed of many
different tasks involving diverse human capacities, and human-machine assemblages.
When evaluating the potential effect(s) of automation on jobs, it makes more sense to
look at what human roles, capacities and tasks are being transformed, and how jobs
and roles could change qualitatively as well as quantitatively, rather than anticipating
whether this or that job will exist tout court at some point in the future.
Something else may be different this time as well. There exist an ever-growing
number of proposals for hard and soft governance mechanisms which aim to humanise,
if not AI, but the socio-technical environments that AI is transforming or may transform
[13]. These approaches, grouped under the umbrella of “human-centred AI”, which
now inform EU guidelines [14], national research funding programmes and iterations of
corporate social responsibility, have something very similar to the questions and concerns
that have informed the future of work debate for decades. They aim to facilitate the
development and application of AI that enhances human capacities rather than replacing
them, that upskills rather than deskills and that overall contributes to human flourishing
rather than redundancy. The primary aim of this talk then is to better understand what
Back to the Future of Work: Old Questions for New Technologies 13

the work programme that we might call “human-centred AI” can learn from the future-
of-work question.

References
1. Susskind, R.E., Susskind, D.: The Future of the Professions: How Technology will Transform
The Work of Human Experts. Oxford University Press, USA (2021)
2. Avent, R.: The Wealth of Humans: Work, Power, and Status In The Twenty-First Century. St
Martins Press, New York (2016)
3. Dejours, C.: The Return of Work in Critical Theory : Self, Society. Columbia University
Press, Politics (2018)
4. Graeber, D.: Bullshit Jobs: A Theory. Penguin, London (2019)
5. Geuss, R.: A philosopher looks at work. Cambridge University Press, Cambridge (2021)
6. Arregui, P.: The changing nature of work and skills in the digital age. In: Gonzalez Vazquez,
I., et al. (eds) EUR 29823 EN, Publications Office of the European Union, Luxembourg, ISBN
978–92–76–09207–0, JRC117505 (2019). https://doi.org/10.2760/373892
7. https://www2.deloitte.com/global/en/pages/human-capital/articles/future-of-work.html
8. Eurofound Employee Monitoring and Surveillance: The Challenges of Digitalisation.
Publications Office of the European Union, Luxembourg (2020)
9. Freeman, R.: Who owns the robots rules the world. IZA World Labor. 5 (2015). https://doi.
org/10.15185/izawol.5
10. Gorz, A.: Critique of Economic Reason. Verso, London (2010)
11. Keynes, J.M.: Economic possibilities for our grandchildren. In: Essays in Persuasion. Palgrave
Macmillan.London (2010). https://doi.org/10.1007/978-1-349-59072-8_25
12. Gellner, E., Breuilly, J.: Nations and Nationalism. Blackwell Publishing, Malden (2013)
13. International KES Conference on Human Centered Intelligent Systems. In: Zimmermann,
A., Howlett, R.J., Jain, L.C., Schmidt, R. (eds.) Human centred intelligent systems. In:
Proceedings of KES-HCIS 2021 Conference https://doi.org/10.1007/978-981-15-5784-2
14. European Commission. High-level expert group on artificial intelligence. Ethics Guidelines
for Trustworthy AI. Publications office of the European Union (2019)
Energy Harvesting and Environment
Protection
Solar Energy Harvesting for the Mobile Robotic
Platform

Michal Hodoň(B) , Peter Ševčík, Juraj Miček, Veronika Olešnániková, Peter Šarafín,
Ján Kapitulík, and Lukáš Čechovič

Faculty of Management Science and Informatics, Department of Technical Cybernetics,


University of Žilina, Univerzitná 8215/1, 01026 Žilina, Slovakia
michal.hodon@fri.uniza.sk

Abstract. This paper describes the educational mobile robotic platform Solarcar.
The platform is used for teaching purposes, especially for description of theoret-
ical as well as practical aspects of using alternative energy sources in embedded
system design. The robot was designed and built as the supportive material for the
students of study field Computer engineering, for the subject Embedded systems.
Through its usage, the principles of so called Green computing in the background
of complex physics are taught in a playful, and for students interesting way.

Keywords: Solar energy · Robot · Educational platform · Green computing

1 Introduction
Electrical energy is often stored in batteries (electrochemical cell). These can be divided
into primary cells that are non-rechargeable and secondary cells that can be recharged.
Note that secondary electrochemical cells have a limited number of charge/discharge
cycles. Capacitors are another possibility of electrical storage. Not long ago, high-
capacity capacitors with the commercial name supercapacitors, or supercapacitors, or
ultracapacitors, have appeared. Their capacitance is several orders of magnitude higher
than that of conventional electrolytic capacitors. With their application together with
solar cell as an energy harvesting technology, a kind of energy-efficient embedded sys-
tem for various application fields can be revealed. The scope of this paper is not to list
and discuss pros and cons of supercapacitor and batteries or other energy harvesting
technologies in detail. An aim of the authors is to introduce and describe small robotic
platform powered via solar panel as a simple educational tool, which can used for better
explanation of various physical laws.
However, because the supercapacitor is a relatively new component, which comes
hand in hand with the developments in nanotechnology, let’s mention its basic properties
in general. It is obvious that the supercapacitor serves to store electricity. Thanks to the
very thin dielectric and large active areas of the electrodes, it achieves significantly
larger capacities than conventional electrolytic capacitors. This feature predestines the
supercapacitor for the energy storing. It performs the same function as rechargeable
batteries. Although, at present time, it is not yet able to fully replace the batteries [1–3],

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022


F. Phillipson et al. (Eds.): I4CS 2022, CCIS 1585, pp. 17–27, 2022.
https://doi.org/10.1007/978-3-031-06668-9_4
18 M. Hodoň et al.

in many applications, especially in low-power applications, supercapacitors successfully


compete with them. Gross comparison of Li-Ion batteries and supercapacitors is given
in the following Table 1.

Table 1. Gross comparison of Li-Ion batteries and supercapacitor

From the table is evident (light-grey filling), in which parameters supercapacitors


exceeds the second type of batteries.

2 Solar Mobile Robotic Platform


The block schematic of the solar mobile robotic platform, further named as Solar car,
can be seen in the Fig. 1 below.

Vcc VM
VM1
DC motor 6V
ADCV1 AD0 M
NSL, IN1, IN2 DRV1 100rpm
VM1

AD1
VM

ADCV2
DC motor 6V
M
Vcc DRV2 100rpm

MCU
R Atmega328P Vcc
ON-OPT
ON-GY
S (Q3)
S (Q5)
FOTORES.
OPT1 OPT2 GYRO
CNY70 CNY70 MPU-6050

Fig. 1. A block schematic of the solar car


Solar Energy Harvesting for the Mobile Robotic Platform 19

As can be seen from the figure above, the solar car, as the educational robotic plat-
form, is built on the popular 8-bit AVR microcontroller Atmega328p as the main control
unit. Further, there are integrated two 6V DC motors 100rpm for the robot driving,
which are operated through DRV8837 1.8A low-voltage DC motor driver. Two optical
sensors OPTx CNY70 together with the MPU-6050 module, which includes triple axis
accelerometer and gyro sensory, allow to navigate the robot through the defined envi-
ronment. Photoresistor is used for the purposes of power optimization through the light
source sensing. The solar car outline can be seen in the Fig. 2.

Fig. 2. Prototype of Solarcar

We will briefly describe the energy management solution of a solar car (Fig. 3). Not
everything mentioned is necessary for students when programming their applications.
Knowledge of the operation of circuits that provide power to the microcontroller and
motors is not necessary during programming.
Basic knowledge is that the supercapacitor C1 is continually charged from the solar
panel. This is the place, where an energy is stored. If the voltage at its electrodes reaches
1.4 V, the DC/DC boost converter will start its operation to increase the input voltage
to 3.4 V. All circuits including the microcontroller (MCU) are then supplied with this
voltage.
From the moment the DC/DC boost converter starts working, the student (or program)
can control solar car’s next activity so that it can complete the task with minimum power
consumption. The car is designed so the students do not have to wait for nice weather
(sun) when programming and tuning programs. For this reason, it is possible to power
the car’s control circuits via the USBASP programmer from the computer’s USB port.
20 M. Hodoň et al.

ENERGY MANAGEMENT

ADCV1
VM1
Umax= 3.4V 1.4 ->3.4V
Solar cell
E=1 Over START/
6V/110mA Superca DC/DC
voltage STOP Vcc
pacitor protecon 1.4/0.8 boost
1F/6V
0.8 – 3.4V

Fig. 3. Block schematic of the power-management circuit of Solarcar

3 Detailed Description of a Solar Car Power Circuitry

From the block diagram of the power-management circuitry is evident, that the protection
of supercapacitor together with the protection as well as starting of the DC/DC conver-
sion, is handled through the circuit design. It should be noted that more elegant solution
would be such that these protections against higher voltage could be solved by the
software. However, in this case the cars may not survive while students developing and
debugging the program. For this reason, these functions are assigned to electronics, even
this is not optimal solution in terms of power consumption optimization. Nevertheless,
basic goal of the proposed solution is the minimization of its power consumption.
As it was already mentioned, the main source, which powers the solar car is solar
panel together with the supercapacitor. The energy stored in a capacitor is given by:
1
E= C · U 2 [J , F, V ] (1)
2
The supercapacitor 1 V/6 V is used as a mean of storing energy in the solar vehicle.
It has a capacity of 1 F and the maximum permissible voltage at its electrodes is 5.5 to
6 V (depending on the supercapacitor type used). It follows an amount of the energy,
which can be stored in it:

E = 0.5 · 1F · (5.5 V)2 = 15.125 J (2)

Power to the supercapacitor is supplied by a solar cell with a maximum power of 0.66
W. The maximum voltage is 6 V. The supercapacitor is protected against overvoltage by
a Zener diode with a breakdown voltage of 5.6 V (Fig. 4).
Then, at the supercapacitor electrodes there is a voltage U which is used:

– for supplying H bridges of DRV8837 DC motors.


– for powering solar car control circuits via DC/DC boost converter (MCP16251) with
3.4 V output.
Solar Energy Harvesting for the Mobile Robotic Platform 21

Fig. 4. Supercapacitor protection

Since the input voltage of the DC/DC converter must not exceed the output voltage
(3.4 V), while the voltage on the supercapacitor can reach up to 5.8 V, the input voltage
limiting circuits are essential. These circuits provide overvoltage protection at the side
of DC/DC converter input (Fig. 5).

Fig. 5. Overvoltage protection of DC/DC boost module

The MCP111 monitoring circuit has at the input voltage lower than 3 V a voltage
of 0 V at the OUT/MCP111 output, which causes the transistor Q1 to be open and the
voltage U to be applied to the DC/DC converter input. The same voltage is applied to
resistor divider R5, R6 and therefore transistor Q8 will be closed, the base of transistor
Q7 will not flow. For this reason, base Q6 will be close to zero and transistor Q6 will
be closed. When the voltage U reaches 3 V, the circuit U2 closes, and at its output the
voltage level U (via resistor R2).
This will cause:

– closure of transistor Q1,


– opening of transistor Q6, which acts as a voltage limiter due to diode D5.

Voltage U on the emitter of Q6 transistor is equal to:

– if 3 V < U < Uz (where Uz = 3,6 V), Uv = U – Ube (where Ube = 0,6 V).
– if U > Uz, Uv = Uz – Ube = 3 V.
22 M. Hodoň et al.

The described circuit part is shown in the Fig. 6.

Fig. 6. Overvoltage protection circuit

To improve efficient operation, the DC/DC converter is preceded by a START/STOP


circuit that starts/stops the converter - MCP16251 via the EN pin when the defined
voltage levels are reached (Fig. 7).

Fig. 7. Block schematic of power circuitry

Start-stop circuits are shown in the following Fig. 8.


If the voltage U1 reaches 0.4 V (Ube Q2), transistor Q2 opens, causing an increase
in current to the base of transistor Q4, which opens “shortens resistance R3” causing
an increase in base current Q2. The voltage at the EN input reaches a value close to U.
It is apparent from the above relations that by suitable selection of resistors R1 to R3 a
suitable combination of Ustart and Ustop voltages can be set over a wide range.
If we choose the values of resistors R1, R2 and R3, R1 = 1.8 M, R2 = 1.8 M,
R3 = 2.7 M, then Ustart = 1.4 V and Ustop = 0.8 V.
Solar Energy Harvesting for the Mobile Robotic Platform 23

Fig. 8. Start-stop circuits

Note that the efficiency of the MCP16251 converter decreases with the decreasing
input voltage. At the output voltage Vout = 3.4 V and the input voltage 3 V the efficiency
at the consumption of 2 to 70 mA is above 95%, however at the voltage 1.5 V the efficiency
is only at the level of 87%. The dependence of the inverter efficiency from Iout and Vin
at Vout = 3.4 V is shown in the Fig. 9 below.

Fig. 9. Dependance of MCP16251 converter efficiency on the input voltage and current
consumption [4]

4 Power Consumption of Solar Car Peripherals


The main power consumption of a solar car is the consumption of a microcontroller.
Although the consumption of all other subsystems is significant, it can be controlled via
the software. All subsystems are powered through MCP 16251, with following supplies:

– MCU ATMEL328P @ RCOSC 1MHz, @ 3.4V, active mode - 0.6 mA


24 M. Hodoň et al.

– control of H-bridge DRV8837 without PWM / PWM50kHz - 0.3/0.7 mA


(SLEEP MODE – 5 nA)
– MPU6050 - 3.6 mA
– optical sensors CNY 70 - 20 mA
– LED - 2 mA

In the figure below, all peripherals are shown in block schematic (Fig. 10).

Fig. 10. Power supply of basic functional parts of a solar car

4.1 Optical Subsystem


As it follows from the above, the optical subsystem has the highest power consumption in
active mode. For this reason, it is necessary to turn it only when required, for the shortest
possible time. Circuitry of optical sensors is shown in the figure below (Fig. 11).
When the switch S1 is turned on, it is necessary to wait until the filter capacitors C1
and C2 are charged. The time constant of the filters, τ is:
τ = R1 · C1 = 0.484 ms (3)
If we assume that after a time of 4τ ≈ 2 ms, the value of output OPT1 can be measured
with sufficient accuracy. In the use scenario, the sensors are used for the navigation of
the robot across the black path at white underlay. So 2 ms is the minimal time to decide
on the reflectivity (color) of the path. The measurement then last 2.1 ms because the
setting of AD converter takes 50 us and AD conversion takes 50 us at clkIO = 1 MHz.
Power used to charge capacitors C1 and C2 is equal to:
1
E= CU 2 = 12.7 · 10−6 J (4)
2
The total energy to charge both capacitors is 0.0254 mJ. The current flowing through
the optocoupler diode can be calculated:
Ucc − UD 3.4 − 1.4
I= = = 9 mA (5)
R2 220
Solar Energy Harvesting for the Mobile Robotic Platform 25

Ucc
S1

R1 R2
47k
220 220
CNY70
ON-OPT C1
2M2 C2

2M2
OPT1
4k7

Fig. 11. Circuitry of optical sensors

The current flowing through the optocoupler transistor depends on the specific condi-
tions (reflection of the emitting infrared radiation). In the worst case, the current through
the transistor is equal to:
Ucc 3.4
IT = = = 0.7 mA (6)
R3 4700
To make it simpler, assume that a current I = 9.7 mA flows through the optocoupler
2.1 ms. Then the energy consumed in one measurement will be equal to:

E = U · I · t = 3.4 · 9.7 · 10−3 · 2.1 · 10−3 = 69 · 10−6 J (7)

The total energy per measurement using both optocouplers will be:

E = (0.0254 + 0.069) · 2 = 0.19 mJ (8)

Recall that the energy accumulated in the supercapacitor at a voltage of 5.5 V was E
= 15.125 J. A simple comparison shows that it is enough energy to perform about 80,000
measurements. Note that these considerations are too basic just to show the students how
the energy intensity of the measurement process is handled. If the measurement is done
every 200 ms then the mean current value is:
E 0.19
I= = = 0.28 mA (9)
U ·t 0.68
In continual measurement, the power consumption would be around 20 mA as it was
stated earlier.

4.2 Motor Drives


The DC motors of the solar car are controlled by two DRV3387 H-bridges. Their con-
sumption in sleep mode can be neglected. It is approximately 100 times lower than the
26 M. Hodoň et al.

ATmega328P @ 1 MHz MCU power consumption. In the active mode, their consump-
tion depends on the excitation. If the inputs do not have a PWM signal, then the typical
current is 0.3 mA. In the case of a 50 kHz PWM signal, the typical current is 0.7 mA.
Note that the consumption of the bridge control circuits is higher than that of the MCU.
For this reason, it is advisable to keep the bridges as long as possible in a sleep state.
Motor drives circuitry is shown in the Fig. 12 below.

Fig. 12. Motor drives circuitry

4.3 Main Controller

The consumption of the ATmega328P microcontroller depends mainly on the frequency


of the clock signal generator, the magnitude of the supply voltage and the active subsys-
tems of the microcontroller. From the datasheet and from measurements is obvious, that
current consumption at 1 MHz is about 0.5 mA. If 12 MHz external crystal oscillator
would be used, the consumption of the MCU in active mode and the 3.4 V supply would
rise to about 5 mA, i.e. about 10 times. In this case it is obvious, that computing power
is paid by consumption rise.

5 Conclusion

In this paper, the circuit solution of the special solar car robotic platform was described
in detail. Energetical requirements, necessary for its driving were discussed too. Though
the solution was successfully practically implemented, there are some improvements to
be applied. This solution is not ideal, because if the supercapacitor voltage rises above
3 V, the energy efficiency of the whole device decreases significantly. In the extreme
case up to 55% at 5.5 V. However, we assume that much of the time the car will operate
at voltages around 3 V. This is already in the hands of the SW developer. On the other
hand, in the case of direct sunlight, the solar panel can supply an amount of energy that
cannot be meaningfully use or conserve. Therefore, in this case, we will not be interested
in the efficiency of the device. From the point of view of the optimal use of the capacity
Solar Energy Harvesting for the Mobile Robotic Platform 27

of the solar panel, it would be necessary to choose a more suitable energy processing
strategy. In the simplest case, it would be necessary to ensure that the voltage on the
supercapacitor is approximately equal to Uc = 0.7Uo at each exposure, where Uo is the
voltage on the solar cell without load. This can be solved by a two-way energy pump
with secondary storage [5, 6].

Acknowledgement. “This publication was realized with support of Operational Program Inte-
grated Infrastructure 2014 - 2020 of the project: Innovative Solutions for Propulsion, Power and
Safety Components of Transport Vehicles, code ITMS 313011V334, co-financed by the European
Regional Development Fund”.

References
1. Musolino, V., Tironi, E., di Milano, P.: A comparison of supercapacitor and high-power lithium
batteries. In: Electrical Systems for Aircraft, Railway and Ship Propulsion, pp. 1–6 (2010).
https://doi.org/10.1109/ESARS.2010.5665263
2. Andreev, M.K.: An overview of supercapacitors as new power sources in hybrid energy storage
systems for electric vehicles. In: 2020 XI National Conference with International Participation
(ELECTRONICA), pp. 1–4 (2020). https://doi.org/10.1109/ELECTRONICA50406.2020.930
5104
3. Soni, S.R., Upadhyay, C.D., Chandwani, H.: Analysis of battery-super capacitor based stor-
age for electrical vehicle. In: 2015 International Conference on Energy Economics and
Environment (ICEEE), pp. 1–7 (2015). https://doi.org/10.1109/EnergyEconomics.2015.723
5110
4. MCP16251 datasheet. http://ww1.microchip.com/downloads/en/devicedoc/20005173b.pdf
5. Kochláň, M., Ševčík, P.: Supercapacitor power unit for an event-driven wireless sensor node.
In: 2012 Federated Conference on Computer Science and Information Systems (FedCSIS),
pp. 791–796 (2012)
6. Kochláň, M., Žák, S., Miček, J., Milanová, J.: Control unit for power subsystem of a wireless
sensor node, pp. 1239–1246 (2015). https://doi.org/10.15439/2015F355
Demonstrating Feasibility of Blockchain-Driven
Carbon Accounting – A Design Study
and Demonstrator

Karl Seidenfad(B) , Tobias Wagner, Razvan Hrestic, and Ulrike Lechner

Computer Science Department, Universität der Bundeswehr München, Munich, Germany


{karl.seidenfad,tobias.wagner,razvan.hrestic,
ulrike.lechner}@unibw.de

Abstract. Carbon accounting calls for innovative digital infrastructures. The


Paris Agreement and the 26th UN Climate Change Conference provide the frame
for designing information systems for carbon accounting. This paper explores
blockchain as a technology for carbon accounting and, in particular, for the Product
Carbon Footprint. This article analyses core architectural designs of carbon foot-
prints, consortia, and smart contract infrastructure. Experiences from the imple-
mentation of a carbon footprint blockchain demonstrator are reported. A coffee
supply chain exemplifies the approach. Hyperledger Fabric and Minifabric from
Hyperledger Labs are the technical frameworks used.

Keywords: Blockchain · Carbon accounting · Supply chain

1 Introduction

Digitalization and decarbonization are strategic topics that require considerable efforts
and innovations. The Paris Agreement sets climate goals for nation-states. A recent
milestone for developing the Paris Agreement [1] is marked by the 26th UN Climate
Change Conference (COP26), held in November 2021 in Glasgow. In the momentum
of the COP26, Article 6.2 of the Paris Agreement refers to decentralized, cooperative
approaches in carbon accounting – a departure from centralized architecture and com-
prehensive rules of present emission trading market mechanisms [2, 3]. Blockchain as
a technology comes with the value promise of decentralized architecture and secure
inter-organizational information flows. This article explores blockchain technology as
technical basis for carbon accounting.
Analysis and design are exemplified with the coffee supply chain. In the coffee
supply chain, farmer, production, logistics, retail provide consumers with roasted coffee
beans. Public authorities monitor the whole process. The use of blockchain to digitalize
the coffee supply chain may address three societal problems. First, a traceable product
history facilitates tracking and tracing to increase food safety [4]. Second, transparency
across the supply chain helps to address concerns of quality, fairness, and sustainability
as all this information can be provided to the end customer. We are interested in a third

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022


F. Phillipson et al. (Eds.): I4CS 2022, CCIS 1585, pp. 28–46, 2022.
https://doi.org/10.1007/978-3-031-06668-9_5
Demonstrating Feasibility of Blockchain-Driven Carbon Accounting 29

application: the transactions stored in the blockchain can be enriched to enable carbon
accounting use cases. This work presented in this article builds on previous work on
blockchain for enabling transparency, tracking, and tracing across the supply chain [5]
and the coffee supply chain is one of the use cases.
The main research interest is how to design blockchain technology for carbon
accounting and what functionalities are necessary. We are also interested in how to
integrate carbon accounting with other transactions in the supply chain for an efficient
solution. A demonstrator validates our design. We study design options and implement
a demonstrator for carbon accounting in the coffee supply chain. The article presents the
design of a demonstrator for blockchain as infrastructure for a coffee supply chain and
for capturing the associated carbon emissions. The article also reports on the learnings
for implementing blockchain-based infrastructures.
This article is organized as follows. Section 2 summarizes related work in blockchain
technology and carbon accounting. Section 3 describes the research design. Section 4
characterizes the scenario illustrated by the demonstrator. Section 5 documents the imple-
mentation results for the demonstrator. Section 6 concludes our experiences and outlines
future work.

2 State of the Art

This work contributes to understanding the design of blockchain-based infrastructures for


carbon accounting systems. The state-of-the-art section reviews Blockchain technology
first and then moves on to carbon accounting.

2.1 Blockchain Technology and Design of Blockchain

The Distributed Ledger Technology (DLT) shapes a family of concepts with different
designs, properties, and characteristics. Popular concepts such as Blockchain, block
directed acyclic graphs (blockDAG), and transaction-based directed acyclic graphs
(TDAG) differ in how data is concatenated. All these concepts have decentralized man-
agement of immutable data and transactions in common. Reports by the NIST [6] and
its German counterpart BSI [7] provide definitions and guidelines for blockchain tech-
nology. They outline the four key DLT characteristics: decentralization, transparency,
immutability, and cryptography. Classification approaches for blockchain technologies
distinguish network exposure (e.g., private or public) or the right to be part of the
consensus (e.g., permissioned or permissionless).
For the design process of blockchain-based information systems, Xu et al. [8] guide
the multitude of configuration possibilities and variants of blockchain-based systems and
summarize design decisions in a flowchart. Marchesi, Marchesi, and Tonelli [9] follow an
agile approach based on user stories and iterative development for designing blockchain
applications. They provide a UML activity diagram as orientation for the design process.
Hoiss, Seidenfad, and Lechner [10] develop an architecture for blockchain-based sys-
tems: blockchain and the integration with business information systems and integration
with services for operation are critical elements of this approach. The approach fosters
a clearer annotation of tasks and responsibilities inside a blockchain-based consortium.
30 K. Seidenfad et al.

Weking et al. [11] provide a taxonomy of blockchain-based business models. The


adoption of private permissioned blockchains in practice is recognized by projects such as
IBM’s TradeLens [12], BMW’s PartChain [13], and the Catena-X Automotive Network
[14]. A systematic review of applications in supply chain management contributed by
Dietrich et al. [15] lists more than 40 blockchain projects. Beyond famous use cases for
supply chain management, Schmiedel [16] connects an additive manufacturing machine
with a permissioned blockchain to enable business models like pay-per-use or predictive
maintenance.
The discussion about implementing carbon accounting and carbon markets with
blockchain technology is vital, and the bandwidth of approaches ranges from public
permissionless to private permissioned blockchain solutions [17–20]. Braden [21] eval-
uates Bitcoin, Ethereum, Hyperledger Fabric, and EOS regarding the applicability for
climate policy applications, with the result that Hyperledger Fabric and Ethereum are
the most promising technology stacks to date. Schletz, Franke, and Salomo [22, 23]
introduce architecture to embed Article 6.2 requirements into a blockchain-based mar-
ket model. In this regard, the vibrant open source community around Hyperledger Fabric
[24] coined the term ‘operating System for Climate Action’ [25].

2.2 Carbon Accounting


Carbon accounting is how organizations track and report their greenhouse gas (GHG)
emissions. In combination with pricing, these are the well-known instruments to realize
cost-effective GHG mitigation [3, 26]. For interoperability, international standards such
as ISO14040/14044, PAS2050, and the Greenhouse Gas Protocol [25] have been intro-
duced. Carbon accounting aims to calculate carbon footprints, as the product carbon
footprint (PCF) and the corporate carbon footprint (CCF). While the CCF collects car-
bon emissions associated with all the activities of a person or other entity (e.g., building,
corporation, country, etc.) [27], the PCF collects all carbon emissions in producing and
distributing a product along the supply chain and serves as meaningful information for
consumers.

Fig. 1. The (simplified) principle of exchanging certified emissions reductions (CER) by the
cap-and-trade scheme according to the German Emissions Trading Authority (DEHSt) [28]

According to the Kyoto protocol, NDC’s are stored in dedicated national registries.
To facilitate the exchange between national registries, the United Nations Framework
Demonstrating Feasibility of Blockchain-Driven Carbon Accounting 31

Convention on Climate Change (UNFCCC) operates the International Transaction Log


(ITL) [29]. The European Union Transaction Log (EUTL) is linked with the ITL and
acts as a database for the European Emissions Trading System (EU ETS) [2]. The EU
ETS is the world’s most extensive emission trading system [30, 31] and utilizes the
cap-and-trade scheme [32] (cf. Fig. 1). In the EU ETS, an individual emissions cap is
assigned for each issuer. This emissions cap acts as an annual threshold. The issuers need
to decide whether to reduce emissions or buy emissions allowances for compensation.
To reduce carbon emissions, the individual emissions cap is lowered by a percentage
amount each year according to regulation.
According to Zhao and Chan [33], the practical problems of current cap-and-trade
implementations include issues in operation and security. The current cap-and-trade
systems, including transaction settlement, credit registration, compliance checking, and
emissions verification, require a large amount of administrative effort. Besides the indus-
tries which already contribute to carbon markets (e.g., energy and aviation), these con-
cerns are also mentioned in the context of onboarding new sectors to carbon markets,
such as, agricultural industry [34, 35]. Also, many of the existing implementations lack
effective punishment mechanisms or their non-strict execution [36]. The centralized
architecture of current cap-and-trade systems makes them vulnerable to hacking and cor-
ruption [37]. Security concerns regard fraud claims for emissions and double accounting
[38]. However, there is also a considerable agreement about the importance of voluntary
carbon accounting (e.g., by small measures) and the challenges of incentivization and
auditing [39, 40].
In 2020, the Kyoto Protocol reached its end of life, and the Paris Agreement became
effective. Schneider et al. [41] compare the Kyoto Protocol and Paris Agreement. For
the operationalization of the Paris Agreement, Müller and Michaelowa [42] discuss two
Internationally Transferred Mitigation Outcomes (ITMO’s) approaches for achieving
NDC’s.

3 Research Design

According to Hevner et al. [42], guided by the design science paradigm, we work in an
iterative design process. Our research builds on a codebase, scenarios, a demonstrator,
and an understanding of designing blockchain to increase security and resilience for the
food supply chain. We have identified design principles to guide our conceptualization,
design, and implementation in our work. One design principle aims for fully decentral-
ized structures, i.e., a blockchain that runs on distributed peers operated by different
organizations (in contrast to the many approaches hosted in a single cloud). This article
is about a use case carbon accounting and extends the current use case, namely tracking,
tracing, and resilience in the food supply chain.
The demonstrator we present in this article marks the first design iteration of
blockchain for carbon accounting as part of project LIONS. Note that the scenario devel-
opment for our blockchain-based carbon accounting demonstrator is strongly inspired
by the theoretical work of Richardson and Xu [17].
This work extends our Blockchain design knowledge base in the fields of operation
[10], design principles [43], design patterns [5], and IoT integration [44, 45].
32 K. Seidenfad et al.

4 The Carbon Accounting Scenario


This section is about the scenario of carbon accounting and core business and archi-
tectural design decisions. We consider three different topics with implications for the
blockchain design: (1) the carbon footprint, (2) the consortium architecture, and (3) the
tokens and smart contracts to run carbon accounting (and potentially compensation).
The coffee supply chain scenario concludes the section.
In our analysis, we assume the existence of a blockchain-based infrastructure for
food safety, similar to our approach in NutriSafe [4]. This infrastructure connects actors
in the supply chain, and each actor adds information to the blockchain. The consumer
then may access this information.

4.1 Carbon Footprint

There are two prominent types of carbon footprint: Product Carbon Footprint and Cor-
porate Carbon Footprint, and both require different kinds of architectures of digital
infrastructure.

Product Carbon Footprints (PCF). The first option to implement carbon accounting
is with a Product Carbon Footprint. This carbon footprint is the sum of all the green-
house gas emissions in production and distribution. To build on the blockchain-based
infrastructure for food safety (similar to our approach in NutriSafe [4]).

Fig. 2. Architecture model excerpt with business Fig. 3. Class diagram representation of the
process, event, and data layer of NutriSafe meta definition concept of NutriSafe
introduced by Lamken et al. [5] introduced by Lamken et al. [5]

This infrastructure connects all actors in the supply chain, and each actor adds infor-
mation to the blockchain. To build on this infrastructure, the data model of the product
with the PCF as a new product attribute using the meta-model introduced by Lamken
et al. [5] (see Fig. 3). We propose for this first option to carbon accounting a design
that extends the existing data model such that all blockchain entries have this additional
information. All relevant actors inside the supply chain provide data about carbon emis-
sions for each product (see Fig. 2). Also, a smart contract is necessary to cumulate carbon
Demonstrating Feasibility of Blockchain-Driven Carbon Accounting 33

emissions between the predecessor and successor of a product in the supply chain. The
flow of information for food safety purposes and the PCF follows the same architecture.
The information on production and distribution might serve as a meaningful parameter
in buying decisions for consumers.

Corporate Carbon Footprints (CCF). The second option for implementing carbon
accounting is the corporate carbon footprint (CCF). CCFs have an essential role in deter-
mining the greenhouse gas emissions according to the Paris agreement and the need for
compensation. This part of the scenario is inspired by the EU ETS and its architecture
depicted in Fig. 4. Independent national registries shape the EU ETS and each registry is
linked with the European Union Transaction Log (EUTL) of the European Commission.
The actors’ roles are described as follows: First, each government must take an adminis-
trative part. Hence, each government manages the issuance of allowances, surrender of
permits, the verification of reported emissions, and grants verification rights to trustwor-
thy actors inside the registry. Second, an installation represents a site, which is emitting.
The holder of an installation transfers permissions to other actors, receive allowances
from the government, and issues credits e.g., by mitigation projects. For a digital infras-
tructure, this means that various actors are connected hierarchically. However, not many
connections exist in this scenario. This structure is depicted in Fig. 4.

Fig. 4. Architecture for the centralized approach of the EU ETS [17]

Note that PCF and CCF are not independent of each other. The PCF of a product will
rely to some extent on data that determine the CCF of the organizations involved. The
PCF is related to the single steps in production and a CCF can serve as a proxy to PCFs
when exact numbers are not known. Hence, the computation of PCF and CCF needs to
be integrated.

The Implications for a Blockchain Infrastructure: The product carbon footprint


can build on existing business transactions and, eventually a blockchain like the one
we have established in project NutriSafe. The product carbon footprint can build on
34 K. Seidenfad et al.

existing transactions and potentially has a high number of transactions; the corporate
carbon footprint potentially needs a new infrastructure with fewer transactions. How-
ever, both need to be linked – to provide data in the blockchain and plausibility and
correctness checks or auditing and certification.

4.2 Consortium Architecture


Public permissionless blockchains such as, e.g., Bitcoin being open for everyone and
granting equal rights to all participants and transactions carry a minimal amount of
usable payload. Implementing a blockchain-based EU ETS demands mechanisms: role-
based access rights to sensitive data, separation of concerns in a channel-like topology
of sidechains, (preferable) feeless transactions, and advanced transaction performance.
According to these requirements, the findings mentioned by Braden [21], and considering
our own experience from NutriSafe [4], we follow the ideas of private permissioned
blockchains and rely on Hyperledger Fabric as our technology platform.
Establishing a productive private permissioned blockchain infrastructure requires the
setup of a corresponding consortium. We analyze the possible relationships and moti-
vations to join a consortium. Studying publications of the German Emissions Trading
Authority (DEHSt) [2, 41, 46], architecture discussions for business models [47, 48],
blockchain-based carbon markets [17, 22, 23, 33, 49], and documentation regarding
the Paris Rulebook in practice [50], we identify three archetypes of carbon accounting
consortia (see Fig. 5).

Fig. 5. Architectures of carbon accounting consortia by information systems, mitigation potential,


and managing emissions shares

Horizontal Consortium. Actors in this consortium are in a similar field of business


with the same kind of value-creating activities. The actors may compete in their business
environment. They share the motivation to reduce the effort of accounting and report
the CCF. Synergies in carbon accounting are leveraged by sharing data and collecting
data on transactions to compute carbon footprints. Carbon accounting is a driver for
horizontal integration.
Another random document with
no related content on Scribd:
to place the most implicit confidence. On such occasions a magical
drum is usually employed. This instrument is formed of a piece of
wood of a semi-oval form, hollow on the flat side, and there covered
with a skin, on which various uncouth figures are depicted; among
which, since the introduction of Christianity into that country, an
attempt is usually made to represent the acts of our Saviour and the
apostles. On this covering several brass rings of different sizes are
laid, while the attendants dispose themselves in many antic postures,
in order to facilitate the charm; the drum is then beat with the horn
of a rein-deer, which occasioning the skin to vibrate, puts the rings in
motion round the figures, and, according to the position which they
occupy, the officiating seer pronounces his prediction[46].
“The remedy,” says a late writer[47], “specifically appropriated for
these maladies of the mind, is the cultivation of natural knowledge;
and it is equally curious and gratifying to observe, that though the
lights of science are attained by only a small proportion of the
community, the benefits of it diffuse themselves universally; for the
belief of ghosts and witches, and judicial astrology, hardly exists, in
these days, even amongst the lowest vulgar. This effect of knowledge,
in banishing the vain fears of superstition, is finely alluded to in the
last words of the following admirable lines quoted from Virgil, e. g.—
Felix qui potuit rerum cognoscere causas,
Atque metus omnes et inexorable fatum,
Subjicit pedibus, Strepitumque Acherontes avari.

But in order to shew with what fervour the belief in witches and
apparitions was maintained about a century and a half ago, we lay
before our readers, as it is scarce, “Doctor Henry More, his letter,
with the postscript to Mr. J. Glanvil[48], minding him of the great
expedience and usefulness of his new intended edition of the Dæmon
of Tedworth, and briefly representing to him the marvellous
weakness and gullerie of Mr. Webster’s[49] display of Witchcraft.”
“Sir,
“When I was at London, I called on your bookseller, to know in
what forwardness this new intended impression of the story of the
Dæmon of Tedworth (see p. 223) was, which will undeceive the
world touching that fame spread abroad, as if Mr. Mompesson and
yourself had acknowledged the business to have been a meer trick or
imposture. But the story, with your ingenious considerations about
witchcraft, being so often printed already, he said it behoved him to
take care how he ventured on a new impression, unless he had some
new matter of that kind to add, which might make this edition the
more certainly saleable; and therefore he expected the issue of that
noised story of the spectre at Exeter, seen so oft for the discovery of a
murther committed some thirty years ago. But the event of this
business, as to juridical process, not answering expectation, he was
discouraged from making use of it, many things being reported to
him from thence in favour of the party most concerned. But I am told
of one Mrs. Britton, her appearing to her maid after her death, very
well attested, though not of such a tragical event as that of Exeter,
which he thought considerable. But of discoveries of murther I never
met with any story more plain and unexceptionable than that in Mr.
John Webster his display of supposed Witchcraft: the book indeed
itself, I confess, is but a weak and impertinent piece; but that story
weighty and convincing, and such as himself, (though otherwise an
affected caviller against almost all stories of witchcraft and
apparitions,) is constrained to assent to, as you shall see from his
own confession. I shall, for your better ease, or because you may not
haply have the book, transcribe it out of the writer himself, though it
be something, chap. 16, page 298, about the year of our Lord 1632,
(as near as I can remember, having lost my notes and the copy of the
letters to Serjeant Hutton, but I am sure that I do most perfectly
remember the substance of the story.)
“Near unto Chester-le-Street, there lived one Walker, a yeoman of
good estate, and a widower, who had a young woman to his
kinswoman, that kept his house, who was by the neighbours
suspected to be with child, and was towards the dark of the evening
one night sent away with one Mark Sharp, who was a collier, or one
that digged coals underground, and one that had been born in
Blakeburn hundred, in Lancashire; and so she was not heard of a
long time, and no noise or tittle was made about it. In the winter
time after, one James Graham, or Grime, (for so in that country they
call them) being a miller, and living about 2 miles from the place
where Walker lived, was one night alone in the mill very late
grinding corn, and about 12 or 1 a clock at night, he came down stairs
from having been putting corn in the hopper: the mill doors being
shut, there stood a woman upon the midst of the floor with her hair
about her head hanging down and all bloody, with five large wounds
on her head. He being much affrighted and amazed, began to bless
himself, and at last asked her who she was and what she wanted? To
which she said, I am the spirit of such a woman who lived with
Walker, and being got with child by him, he promised to send me to
a private place, where I should be well lookt too till I was brought in
bed and well again, and then I should come again and keep his
house. And accordingly, said the apparition, I was one night late
sent away with one Mark Sharp, who upon a moor, naming a place
that the miller knew, slew me with a pick, such as men dig coals
withal, and gave me these five wounds, and after threw my body
into a coal pit hard by, and hid the pick under a bank; and his shoes
and stockings being bloody, he endeavoured to wash ’em; but seeing
the blood would not forth, he hid them there. And the apparition
further told the miller, that he must be the man to reveal it, or else
that she must still appear and haunt him. The miller returned home
very sad and heavy, but spoke not one word of what he had seen, but
eschewed as much as he could to stay in the mill within night without
company, thinking thereby to escape the seeing again of that frightful
apparition. But notwithstanding, one night when it began to be dark,
the apparition met him again, and seemed very fierce and cruel, and
threatened him, that if he did not reveal the murder she would
continually pursue and haunt him; yet for all this, he still concealed
it until St. Thomas’s Eve, before Christmas, when being soon after
sun-set in his garden, she appeared again, and then so threatened
him, and affrighted him, that he faithfully promised to reveal it next
morning. In the morning he went to a magistrate and made the
whole matter known with all the circumstances; and diligent search
being made, the body was found in a coal pit, with five wounds in the
head, and the pick, and shoes and stockings yet bloody, in every
circumstance as the apparition had related to the miller; whereupon
Walker and Mark Sharp were both apprehended, but would confess
nothing. At the assizes following, I think it was at Durham, they were
arraigned, found guilty, condemned, and executed; but I could never
hear they confest the fact. There were some that reported the
apparition did appear to the judge or the foreman of the jury, who
was alive in Chester-le-Street about ten years ago, as I have been
credibly informed, but of that I know no certainty: there are many
persons yet alive that can remember this strange murder and the
discovery of it; for it was, and sometimes yet is, as much discoursed
of in the North Country as any that almost has ever been heard of,
and the relation printed, though now not to be gotten. I relate this
with great confidence, (though I may fail in some of the
circumstances) because I saw and read the letter that was sent to
sergeant Hutton, who then lived at Goldsbrugh, in Yorkshire, from
the judge before whom Walker and Mark Sharp were tried, and by
whom they were condemned, and had a copy of it until about the
year 1658, when I had it, and many other books and papers taken
from me; and this I confess to be one of the most convincing stories,
being of undoubted verity, that ever I read, heard, or knew of, and
carrieth with it the most evident force to make the most incredulous
to be satisfied that there are really sometimes such things as
apparitions.” Thus far he.
“This story is so considerable that I make mention of it in my
Scholea, on the Immortality of the Soul, in my Volumen
Philosophicum, tom. 2, which I acquainting a friend of mine with, a
prudent, intelligent person, Dr. J. D. he of his own accord offered
me, it being a thing of much consequence, to send to a friend of his in
the north for greater assurance of the truth of the narrative, which
motion I willingly embracing, he did accordingly. The answer to this
letter from his friend Mr. Sheperdson, is this: I have done what I can
to inform myself of the passage of Sharpe and Walker; there are
very few men that I could meet that were then men, or at the tryal,
saving these two in the inclosed paper, both men at that time, and
both at the trial; and for Mr. Lumley, he lived next door to Walker,
and what he hath given under his hand, can depose if there were
occasion. The other gentleman writ his attestation with his own
hand; but I being not there got not his name to it. I could have sent
you twenty hands that could have said thus much and more by
hearsay, but I thought those most proper that could speak from
their own eyes and ears. Thus far (continues Dr. More,) Mr.
Sheperdson, the Doctor’s discreet and faithful intelligencer. Now for
Mr. Lumly, or Mr. Lumley. Being an ancient gentleman, and at the
trial of Walker and Sharp upon the murder of Anne Walker, saith,
That he doth very well remember that the said Anne was servant to
Walker, and that she was supposed to be with child, but would not
disclose by whom; but being removed to her aunt’s in the same town
called Dame Caire, told her aunt (Dame Caire) that he that got her
with child, would take care both of her and it, and bid her not trouble
herself. After some time she had been at her aunt’s, it was observed
that Sharp came to Lumley one night, being a sworn brother of the
said Walker’s; and they two that night called her forth from her
aunt’s house, which night she was murdered; about fourteen days
after the murder, there appeared to one Graime, a fuller, at his mill,
six miles from Lumley, the likeness of a woman with her hair about
her head, and the appearance of five wounds in her head, as the said
Graime gave it in evidence; that that appearance bid him go to a
justice of peace, and relate to him, how that Walker and Sharp had
murthered her in such a place as she was murthered; but he, fearing
to disclose a thing of that nature against a person of credit as Walker
was, would not have done it; upon which the said Graime did go to a
justice of peace and related the whole matter[50]. Whereupon the
justice of peace granted warrants against Walker and Sharp, and
committed them to a prison; but they found bail to appear at the next
assizes, at which they came to their trial, and upon evidence of the
circumstances, with that of Graime of the appearance, they were
both found guilty and executed.
“The other testimony is that of Mr. James Smart and William
Lumley, of the city of Durham, who saith, that the trial of Sharp and
Walker was in the month of August 1631, before judge Davenport.
One Mr. Fanhair gave it in evidence upon oath, that he saw the
likeness of a child stand upon Walker’s shoulders during the time of
the trial, at which time the judge was very much troubled, and gave
sentence that night the trial was, which was a thing never used in
Durham before nor after; out of which two testimonies several things
may be counted or supplied in Mr. Webster’s story, though it be
evident enough that in the main they agree; for that is but a small
disagreement as to the years, when Mr. Webster says about the year
of our Lord 1632, and Mr. Fanhair, 1631. But unless at Durham they
have assizes but once in the year, I understand not so well how Sharp
and Walker should be apprehended some little time after St.
Thomas’s day, as Mr. Webster has, and be tried the next assizes at
Durham, and yet that be in August, according to Mr. Smart’s
testimony. Out of Mr. Lumley’s testimony the christian name of the
young woman is supplied, as also the name of the town near Chester-
le-Street, namely, Lumley: the circumstance also of Walker’s sending
away his kinswoman with Mark Sharp are supplied out of Mr.
Lumley’s narrative, and the time rectified, by telling it was about
fourteen days till the spectre after the murder, when as Mr. Webster
makes it a long time.”
We shall not follow the learned Doctor through the whole of his
letter, which principally now consists in rectifying some little
discrepancies in the account of the murder of Anne Walker, and the
execution of the murderers, upon circumstantial evidence, supported
by the miller’s story of the apparition, between the account given by
Mr. Webster, and that here related by Lumley and Sharp. Mr.
Webster’s account, it would appear, was taken from a letter written
by Judge Davenport to Sergeant Hutton, giving a detailed narrative
of the whole proceeding as far as came within his judicial
observation, and the exercise of his functions; which it also appears
Dr. More likewise saw; a copy of which, he states, he had in fact by
him for some considerable time, but which he unfortunately lost: his
account, therefore, is from sheer recollection of the contents of this
letter, but as there is very little difference in the material points,
unless with respect to the date of the year, between the account given
by Webster, and that related from the Doctor’s memory, we shall
offer no further observation than that the whole savours so much of
other similar stories, the result of superstition and ignorance, that it
claims an equal proportion of credit: for if, at the time we allude to,
they would hang, burn, or drown a woman for a witch, either upon
her own evidence, or that of some of her malignant and less
peaceably disposed neighbours, it cannot be matter of surprise, that
two individuals, for a crime really committed, should be hanged as
murderers upon the testimony of the apparition of a murdered
person, given through the organ of a miller, who resided only six
miles from the spot.
That Dr. Henry More was not only an enthusiast and a visionary,
(both of which united in the same person, constitute a canting
madman) but also a humorous kind of fellow when he chose to be
jocular, and it would appear he was by no means incapable of
relaxing the gravity of his countenance as occasion served him, may
be still further inferred from the following extracts of the sequel of
his letter to the Reverend Joseph Glanvil:—
“This story of Anne Walker, (says Dr. M.) I think you will do well
to put amongst your additions in the new impression of your new
edition of your Dæmon of Tedworth, it being so excellently well
attested, AND SO UNEXCEPTIONABLE IN EVERY RESPECT; and hasten as fast
as you can that impression, to undeceive the half-witted world, who
so much exult and triumph in the extinguishing the belief of that
narration, as if the crying down the truth of that of the Dæmon of
Tedworth, were indeed the very slaying of the devil, and that they
may now, with more gaiety and security than ever, sing in a loud
note, that mad drunken catch—
Hay ho! the Devil is dead, &c.

Which wild song, though it may seem a piece of levity to mention,


yet, believe me, the application thereof bears a sober and weighty
intimation along with it, viz. that these sort of people are very
horribly afraid that there should be any spirit, lest there should be a
devil, and an account after this life; and therefore they are impatient
of any thing that implies it, that they may with a more full swing, and
with all security from an after reckoning, indulge their own lusts and
humours; and I know by long experience that nothing rouses them so
much out of that dull lethargy of atheism and sadducism, as
narrations of this kind, for they being of a thick and gross spirit, the
most subtle and solid deductions of reason does little execution upon
them; but this sort of sensible experiments cuts them and stings
them very sore, and so startles them, that a less considerable story by
far than this of the drummer of Tedworth, or of Ann Walker, a
Doctor of Physic cryed out presently, if this be true I have been in a
wrong box all this time, and must begin my account anew.
“And I remember an old gentleman, in the country, of my
acquaintance, an excellent justice of peace, and a piece of a
mathematician, but what kind of a philosopher he was you may
understand from a rhyme of his own making, which he commended
to me at my taking horse in his yard; which rhyme is this:—
Ens is nothing till sense finds out;
Sense ends in nothing, so naught goes about.

Which rhyme of his was so rapturous to himself, that at the reciting


of the second verse the old man turned himself about upon his toe as
nimbly as one may observe a dry leaf whisked round in the corner of
an orchard walk, by some little whirlwind. With this philosopher I
have had many discourses concerning the immortality of the soul
and its destruction: when I have run him quite down by reason, he
would but laugh at me, and say, this is logic, H., calling me by my
christian name; to which I replied, this is reason, Father L., (for I
used and some others to call him) but it seems you are for the new
lights and the immediate inspirations, which I confess he was as little
for as for the other; but I said so only in the way of drollery to him in
those times, but truth is, nothing but palpable experience would
move him, and being a bold man, and fearing nothing, he told me he
had used all the magical ceremonies of conjuration he could to raise
the devil or a spirit, and had a most earnest desire to meet with one,
but never could do it. But this he told me, when he did not so much
as think of it, while his servant was pulling off his boots in the hall,
some invisible hand gave him such a clap upon the back that it made
all ring again; so, thought he, I am invited to converse with a spirit;
and therefore so soon as his boots were off and his shoes on, out he
goes into the yard and next field to find out the spirit that had given
him this familiar slap on the back, but found him neither in the yard
nor the next field to it.
“But though he did not feel this stroke, albeit he thought it
afterwards (finding nothing came of it) a mere delusion; yet not long
before his death it had more force with him than all the philosophical
arguments I could use to him, though I could wind him and non-plus
him as I pleased; but yet all my arguments, how solid soever, made
no impression upon him, wherefore after several reflections of this
nature, whereby I would prove to him the soul’s distinction from the
body, and its immortality, when nothing of such subtile
considerations did any more execution in his mind, than some
lightening is said to do, though it melts the sword on the fuzzy
consistency of the scabbard: Well, said I, Father L., though none of
these things move you, I have something still behind, and what
yourself has acknowledged to me to be true, that may do the
business: do you remember the clap on your back, when your
servant was pulling off your boots in the hall? Assure yourself, said
I, Father L., that goblin will be the first that will bid you welcome in
the other world. Upon that his countenance changed most sensibly,
and he was more confounded with rubbing up of his memory than
with all the rational and philosophical argumentations that I could
produce.”
How the various commentators on holy writ have reconciled to
their minds the existence of spirits, witches, hobgoblins, devils, &c.
we are unable to decide, for the want of a folio before us; but, if there
are none of this evil-boding fraternity “wandering in air” at the
present day, they must be all swamped in the Red sea, ready to be
conjured up from the “vasty deep,” by the king of spirits alone; for as
sure as the Bible is the word of truth, we find therein such
descriptions of spirits, apparitions, witches, and devils, as would
make an ordinary man’s hair stand on end. And it is from this source
alone that Dr. More argues for their existence, and which he has fully
corroborated by his old hobby, “The Dæmon of Tedworth,” and the
unfortunate Anne Walker.
“Indeed (says the learned divine) if there were any modesty left in
mankind, the histories of the Bible might abundantly assure men of
the existence of angels and spirits.”
In another place he observes, “I look upon it as a special piece of
providence that there are ever and anon such fresh examples of
apparitions and witchcraft, as may rub up and awaken their
benumbed and lethargic minds into a suspicion at least, if not
assurance, that there are other intelligent beings besides those that
are clothed in heavy earth or clay; in this I say, methinks the divine
providence does plainly interest the powers of the dark kingdom,
permitting wicked men and women, and vagrant spirits of that
kingdom, to make leagues or covenant one with another, the
confession of witches against their own lives being so palpable an
evidence, besides the miraculous feats they play, that there are bad
spirits, which will necessarily open a door to the belief that there are
good ones, and lastly that there is a God.” There is beyond a doubt
much plausibility, supported by strong and appropriate argument, in
this declaration of the Doctor’s. But as it is not our province to
confute or explain texts or passages of Scripture, much less to warp
them round to particular purposes, we shall reply by observing that,
although we do not entirely concur in the belief of the non-existence
of witches, apparitions, &c. at an earlier period of the world; we do,
from our very souls, sincerely believe that there are no guests of this
description, at the present day, either in the water or roaming about
at large and invisible, on terra firma; or floating abroad in ether,
holding, or capable of holding, converse or communion, either by
word, deed, or sign, with the beings of this earth, civilized or
uncivilized, beyond those destined by the God of heaven to constitute
the different orders, classes, and genera of its accustomed and
intended inhabitants. However, as we live in a tolerant mixed age, we
have no fault to find with those who may attach faith to the opposite
side of our creed.
We shall now, previous to laying before our readers some of those
dismal stories of witches, wizards, apparitions, &c. of the days of
yore, give the postscript to Dr. More’s letter to the author of
“Saducismus Triumphatus;” a postscript, in fact, that might with
more propriety be styled a treatise on the subject it relates to; but the
rarity of the document, as well as its curiosity and the great learning
and ingenuity it betrays, will, we feel assured, be received as an
apology for bringing it under their view in this part of our paper, on
the subject matter it bears so strongly upon. We give it the more
cheerfully as it exemplifies certain passages of Scripture that have
never been handled, at least so well, by after-writers who have
attempted the illustration.
Witchcraft proved by the following texts of
Scripture.
Exodus, c. xxii, v. 18. Thou shalt not suffer a WITCH to live.
2 Chronicles, c. xxxiii, v. 6. And he caused his children to pass
through the fire in the Valley of the Son of Hinnom; also he
observed times, and used ENCHANTMENTS, and used WITCHCRAFT, and
dealt with a FAMILIAR SPIRIT, and with WIZARDS: he wrought much evil
in the sight of the Lord, to provoke him to anger.
Galatians, c. v, v. 20. Idolatry, WITCHCRAFT, hatred, variance,
emulations, wrath, strife, seditions, heresies.
Micah, c. v, v. 12. I will cut off WITCHCRAFTS out of thine hand; and
thou shalt have no more soothsayers.
Acts, c. xiii, v. 6, 8. ¶ And when they had gone through the isle
unto Paphos, they found a certain Sorcerer, a false prophet, a Jew,
whose name was Bar-jesus.
But Elymas the Sorcerer, (for so is his name by interpretation,)
withstood them, seeking to turn away the deputy from the faith.
Acts, c. viii, v. 9. ¶ But there was a certain man called Simon, which
before time in the same city used Sorcery, and bewitched the people
of Samaria, giving out that himself was some great one.
Deuteronomy, c. xviii, v. 10, 11. There shall not be found among
you any one, that maketh his son or his daughter to pass through the
fire, or that useth divination, or an observer of times, or an
enchanter, or a witch.
Or a charmer, or a consulter with familiar spirits, or a
necromancer.
12. For all that do these things are an abomination: and because of
these abominations, the Lord thy God doth drive them from before
thee.
Dr. More’s Postscript.
The following scarce, curious, and learned document, long since
out of print, forms a postscript written by Dr. More, who, it appears,
strenuously advocated the existence of preternatural agencies,
against the opinion of many eminent men, who wrote, at that time,
on the same subject; and however much the belief in witches, &c.
may have been depreciated of later years, we will venture to say that
few of the present day, layman or divine, could take up his pen, and
offer so learned a refutation against, as Dr. More has here done in
support of his opinions founded on Scripture.
“This letter lying by me some time before I thought it opportune to
convey it, and in the meanwhile meeting more than once with those
that seemed to have some opinion of Mr. Webster’s criticisms and
interpretations of Scripture, as if he had quitted himself so well
there, that no proof thence can hereafter be expected of the being of
a witch, which is the scope that he earnestly aims at; and I reflecting
upon that passage in my letter, which does not stick to condemn
Webster’s whole book for a weak and impertinent piece, presently
thought fit, (that you might not think that censure over-rash or
unjust) it being an endless task to shew all the weakness and
impertinencies of his discourse, briefly by way of Postscript, to hint
the weakness and impertinency of this part which is counted the
master-piece of the work, that thereby you may perceive that my
judgment has not been at all rash touching the whole.
“And in order to this, we are first to take notice what is the real
scope of his book; which if you peruse, you shall certainly find to be
this: That the parties ordinarily deemed witches and wizzards, are
only knaves and queans, to use his phrase, and arrant cheats, or deep
melancholists; but have no more to do with any evil spirit or devil, or
the devil with them, than he has with other sinners or wicked men,
or they with the devil. And secondly, we are impartially to define
what is the true notion of a witch or wizzard, which is necessary for
the detecting of Webster’s impertinencies.
“As for the words witch and wizzard, from the notation of them,
they signify no more than a wise man or a wise woman. In the word
wizzard, it is plain at the very first sight. And I think the most plain
and least operose deduction of the name witch, is from wit, whose
derived adjective might be wittigh or wittich, and by contraction
afterwards witch; as the noun wit is from the verb to weet, which is,
to know. So that a witch, thus far, is no more than a knowing woman;
which answers exactly to the Latin word saga, according to that of
Festus, Sagæ dictæ anus quæ multa sciunt. Thus in general: but use
questionless had appropriated the word to such a kind of skill and
knowledge, as was out of the common road, or extraordinary. Nor
did this peculiarity imply in it any unlawfulness. But there was after
a further restriction and most proper of all, and in which alone now-
a-days the words witch and wizzard are used. And that is, for one
that has the knowledge or skill of doing or telling things in an
extraordinary way, and that in virtue of either an express or implicit
sociation or confederacy with some evil spirit. This is a true and
adequate definition of a witch or wizzard, which to whomsoever it
belongs, is such, et vice versâ. But to prove or defend that there
neither are, nor ever were any such, is, as I said, the main scope of
Webster’s book: in order to which, he endeavours in his sixth and
eighth chapters to evacuate all the testimonies of Scripture; which
how weakly and impertinently he has done, I shall now shew with all
possible brevity and perspicuity.
“The words that he descants upon are Deut. c. xviii. v. 10, 11:
‘There shall not be found among you any one that useth divination,
or an observer of times, or an enchanter, or a witch, or a charmer, or
a consulter with familiar spirits, or a wizzard, or a necromancer.’ The
first word in the Hebrew is ‫קוסם קסמים‬, kosem kesamim, a diviner.
Here because ‫ קסם‬kasam, sometimes has an indifferent sense, and
signifies to divine by natural knowledge or human prudence or
sagacity; therefore nothing of such a witch as is imagined to make a
visible league with the devil, or to have her body sucked by him, or
have carnal copulation with him, or is really turned into a cat, hare,
wolf or dog, can be deduced from this word. A goodly inference
indeed, and hugely to the purpose, as is apparent from the foregoing
definition. But though that cannot be deduced, yet in that, this
divination that is here forbidden, is plainly declared abominable and
execrable, as it is v. 12, it is manifest that such a divination is
understood that really is so; which cannot well be conceived to be,
unless it imply either an express or implicite inveaglement with some
evil invisible powers who assist any kind of those divinations that
may be comprehended under this general term. So that this is plainly
one name of witchcraft, according to the genuine definition thereof.
And the very words of Saul to the witch of Endor, are, ‫קסומי נא לי‬
‫ ;באוב‬that is to say, ‘Divine to me, I pray thee, by thy familiar spirit.’
Which is more than by natural knowledge or human sagacity.
“The next word is ‫ מעונן‬megnonen, which, though our English
translation renders [gnon] (tempus,) ‘an observer of times;’ which
should rather be a declarer of the seasonableness of the time, or
unseasonableness of the time, or unseasonableness as to success; a
thing which is inquired of also from witches, yet the usual sense,
rendered by the learned in the language, is præstigiatur, an imposer
on the sight, Sapientes prisci, says Buxtorf, a ‫[ עין‬gnajin, oculus]
deduxerunt et ‫[ מעונן‬megnonen] esse eum dixerunt, qui tenet et
præstringit oculos, ut falsum pro vero videant. Lo, another word
that signifies a witch or a wizzard, which has its name properly from
imposing on the sight, and making the by-stander believe he sees
forms or transformations of things he sees not! As when Anne
Bodenham transformed herself before Anne Styles in the shape of a
great cat; Anne Styles’s sight was so imposed upon, that the thing to
her seemed to be done, though her eyes were only deluded. But such
a delusion certainly cannot be performed without confederacy with
evil spirits. For to think the word signifies præstigiator, in that sense
we translate in English, juggler, or a hocus-pocus, is so fond a
conceit, that no man of any depth of wit can endure it. As if a merry
juggler that plays tricks of legerdemain at a fair or market, were such
an abomination to either the God of Israel, or to his law-giver Moses;
or as if a hocus-pocus were so wise a wight as to be consulted as an
oracle: for it is said, v. 14, ‘For the nations which thou shalt possess,
they consult,’ ‫ מעוננים‬megnonenim. What, do they consult jugglers
and hocus-pocusses? No, certainly, they consult witches or wizzards,
and diviners, as Anne Styles did Anne Bodenham.’ Wherefore here is
evidently a second name of a witch.
“The third word in the text is ‫ מנחש‬menachesh, which our English
translation renders, an enchanter. And, with Mr. Webster’s leave,
(who insulteth so over their supposed ignorance) I think they have
translated it very learnedly and judiciously; for charming and
enchanting, as Webster himself acknowledges, and the words
intimate, being all one, the word, ‫ מנחש‬menachesh, here, may very
well signify enchanters, or charmers; but such properly as kill
serpents by their charming, from ‫ נחש‬nachash, which signifies a
serpent, from whence comes ‫ נחש‬nichesh, to kill serpents, or make
away with them. For a verb in pihel, sometimes (especially when it is
formed from a noun) has a contrary signification. Thus from ‫שרש‬
radix is ‫ שרש‬radices evulsit, from ‫ דשן‬cinis ‫ דשן‬removit cineres, from
‫ חטא‬peccavit ‫ חטא‬expiavit à peccato; and so lastly from ‫ נחש‬serpens,
is made ‫ נחש‬liberavit â serpentibus, nempe occidendo vel fugando
per incantationem. And therefore there seems to have been a great
deal of skill and depth of judgment in our English translators that
rendered ‫ מנחש‬menachesh, an enchanter, especially when that of
augur or soothsayer, which the Septuagint call Ὀιωνιζόμενον (there
being so many harmless kinds of it) might seem less suitable with
this black list: for there is no such abomination in adventuring to tell,
when the wild geese fly high in great companies, and cackle much,
that hard weather is at hand, but to rid serpents by a charm is above
the power of nature; and therefore an indication of one that has the
assistance of some invisible spirit to help him in this exploit, as it
happens in several others; and therefore this is another name of one
that is really a witch.
“The fourth word is ‫ מכשף‬mecasseph, which our English
translators render, a witch; for which I have no quarrel with them,
unless they should so understand it that it must exclude others from
being so in that sense I have defined, which is impossible they
should. But this, as the foregoing, is but another term of the same
thing; that is, of a witch in general, but so called here from the
prestigious imposing on the sight of beholders. Buxtorf tells us, that
Aben Ezra defines those to be ‫[ מכשפים‬mecassephim] qui mutant et
transformant res naturales ad aspectum oculi. Not as jugglers and
hocus-pocusses, as Webster would ridiculously insinuate, but so as I
understood the thing in the second name; for these are but several
names of a witch, who may have several more properties than one
name intimates. Whence it is no wonder that translators render not
them always alike. But so many names are reckoned up here in this
clause of the law of Moses, that, as in our common law, the sense
may be more sure, and leave no room to evasion. And that here this
name is not from any tricks of legerdemain as in common jugglers
that delude the sight of the people at a market or fair, but that it is
the name of such as raise magical spectres to deceive men’s sight,
and so are most certainly witches, is plain from Exod. chap. xxii, v.
18, ‘Thou shalt not suffer,’ ‫ מכשפה‬mecassephah, that is, ‘a witch, to
live.’ Which would be a law of extreme severity, or rather cruelty,
against a poor hocus-pocus for his tricks of legerdemain.
“The fifth name is ‫ חובר חבר‬chobher chebher, which our English
translators render charmer, which is the same with enchanter.
Webster upon this name is very tedious and flat, a many words and
small weight in them. I shall dispatch the meaning briefly thus: this
‫חובר חבר‬, chobher chebher, that is to say, socians societatem, is
another name of a witch, so called specially either from the
consociating together serpents by a charm, which has made men
usually turn it (from the example of the Septuagint, ἐπάδων
ἐπαοιδὴν,) a charmer, or an enchanter, or else from the society or
compact of the witch with some evil spirits; which Webster
acknowledges to have been the opinion of two very learned men,
Martin Luther and Perkins, and I will add a third, Aben Ezra, (as
Martinius hath noted,) who gives this reason of the word ‫חובר‬
chobher, an enchanter, which signifies socians or jungens, viz. Quòd
malignos spiritus sibi associat. And certainly one may charm long
enough, even till his heart aches, ere he make one serpent assemble
near him, unless helped by this confederacy of spirits that drive them
to the charmer. He keeps a pudder with the sixth verse of the fifty-
eighth Psalm to no purpose; whereas from the Hebrew, ‫אשר לא־ישמע‬
‫לקול מלחשים חובר חברים מחכם‬, if you repeat ἀπὸ κοινoῦ ‫ לקול‬before
‫חובר‬, you may with ease and exactness render it thus: ‘That hears not
the voice of muttering charmers, no not the voice of a confederate
wizzard, or charmer that is skilful.’ But seeing charms, unless with
them that are very shallow and sillily credulous, can have no such
effects of themselves, there is all the reason in the world (according
as the very word intimates, and as Aben Ezra has declared,) to
ascribe the effect to the assistance, confederacy, and co-operation of
evil spirits, and so ‫חובר חברים‬, chobher chabharim, or ‫חובר חבר‬
chobher chebher, will plainly signify a witch or wizzard according to
the true definition of them. But for J. Webster’s rendering this verse,
p. 119, thus, Quæ non audiet vocem mussitantium incantationes
docti incantantis, (which he saith is doubtless the most genuine
rendering of the place) let any skilful man apply it to the Hebrew
text, and he will presently find it grammatical nonsense. If that had
been the sense, it should have been ‫חברי חובר מחכם‬.
“The sixth word is ‫שואל אוב‬, shoel obh, which our English
translation renders, ‘a consulter with familiar spirits;’ but the
Septuagint Ἐγγαστρίμυθος. Which therefore must needs signifie him
that has this familiar spirit: and therefore ‫ שואל אוב‬shoel obh, I
conceive, (considering the rest of the words are so to be understood)
is to be understood of the witch or wizzard himself that asks counsel
of his familiar, and does by virtue of him give answers unto others.
The reason of the name of ‫ אוב‬obh, it is likely was taken first from
that spirit that was in the body of the party, and swelled it to a
protuberancy like the side of a bottle. But after, without any relation
to that circumstance, OBH signifies as much as pytho; as pytho also,
though at first it took its name from the pythii vates, signifies no
more than spiritum divinationis, in general, a spirit that tells hidden
things, or things to come. And OBH and pytho also agree in this, that
they both signify either the divinatory spirit itself, or the party that
has that spirit. But here in ‫שואל אוב‬, shoel obh, it being rendered by
the Septuagint Ἐγγαςείμυθος, OBH is necessarily understood of the
spirit itself, as pytho is, Acts xvi. 16, if you read πνεῦμα πύδωνα, with
Isaac Casaubon; but if πύθωνος, it may be understood either way. Of
this πνεύμα πύθων, it is recorded in that place, that ‘Paul being
grieved, turned and said to that spirit, I command thee, in the name
of Jesus Christ, to come out of her, and he came out at the same
hour;’ which signifies as plainly as any thing can be signified, that
this pytho or spirit of divination, that this OBH was in her: for
nothing can come out of the sack that was not in the sack, as the
Spanish proverb has it; nor could this pytho come out of her unless it
was a spirit distinct from her; wherefore I am amazed at the profane
impudence of J. Webster, that makes this pytho in the maid there
mentioned, nothing but a wicked humour of cheating and cozening
divination: and adds, that this spirit was no more cast out of that
maid than the seven devils out of Mary Magdalene, which he would
have understood only of her several vices; which foolish familistical
conceit he puts upon Beza as well as Adie. Wherein as he is most
unjust to Beza, so he is most grossly impious and blasphemous
against the spirit of Christ in St. Paul and St. Luke, who makes them
both such fools as to believe that there was a spirit or divining devil
in the maid, when according to him there is no such thing. Can any
thing be more frantic or ridiculous than this passage of St. Paul, if
there was no spirit or devil in the damsel? But what will this profane
shuffler stick to do in a dear regard to his beloved hags, of whom he
is sworn advocate, and resolved patron right or wrong?
“But to proceed, that ‫אוב‬, obh, signifies the spirit itself that divines,
not only he that has it, is manifest from Levit. xx. 27, Vir autem sive
mulier cùm fuerit [‫ ]בהם אוב‬in eis pytho. And 1 Sam. xxviii. 8, Divina
quæso mihi [‫ ]באוב‬per pythonem. In the Septuagint it is ἐν τῶν
Ἐγγαστρίμυθῳ, that is, by that spirit that sometimes goes into the
body of the party, and thence gives answers; but here it only signifies
a familiar spirit. And lastly, ‫בעלת אוב‬, bagnalath obh, 1 Sam. xxviii. 7,
Quæ habit pythonem; there OBH must needs signify the spirit itself,
of which she of Endor was the owner or possessor; that is to say, it
was her familiar spirit. But see what brazen and stupid impudence
will do here, ‫בעלת אוב‬, bagnalath obh, with Webster must not signify
one that has a familiar spirit, but the mistress of the bottle. Who but
the master of the bottle, or rather of whom the bottle had become
master, and by guzzling had made his wits excessively muddy and
frothy, could ever stumble upon such a foolish interpretation? But
because ‫ אוב‬obh, in one place of the Scripture signifies a bottle, it
must signify so here, and it must be the instrument forsooth, out of
which this cheating quean of Endor does ‘whisper, peep, or chirp like
a chicken coming out of the shell,’ p. 129, 165. And does she not, I
beseech you, put her nib also into it sometimes, as into a reed, as it is
said of that bird, and cries like a butter-bump? certainly he might as
well have interpreted ‫ בעלת אוב‬bagnalath obh, of the great tun of
Heidelberg, that Tom. Coriat takes such special notice of, as of the
bottle.
“And truly so far as I see, it must be some such huge tun at length
rather than the bottle, that is, such a spacious tub as he in his
deviceful imagination fancies Manasses to have built; a μανείον
forsooth, or oracular edifice for ‘cheating rogues and queans to play
their cozening tricks in;’ from that place 2 Chron. xxxiii. 6, ‫ועשה אוב‬,
Et fecit pythonem. Now, says he, how could Manasses make a
familiar spirit? or make one that had a familiar spirit? Therefore he
made a bottle a tun, or a large tub, a μαντεῖον, or oracular edifice ‘for
cheating rogues or queans to play their cozening tricks in.’ Very
wisely argued, and out of the very depth of his ignorance of the
Hebrew tongue, whereas if he had looked into Buxtorf’s Dictionary
he might have understood that ‫ עשה‬signifies not only fecit but also
paravit, comparavit, acquisivit, magni fecit, none of which words
imply the making of OBH in his sense, but the only appointing them
to be got, and countenancing them. For in Webster’s sense he did not
make ‫ ידעני‬jidegnoni neither, that is wizzards, and yet Manasses is
said to make them both alike. ‫יעשה אוב וידעני‬, Et fecit pythonem et
magos. So plain is it that ‫אוב‬, obh, signifies pytho, and that
adequately in the same sense that pytho does, either a familiar spirit,
or him that has that spirit of divination. But in ‫בעלת אוב‬, bagnalath
obh, it necessarily signifies the familiar spirit itself, which assisted
the witch of Endor; whereby it is manifest she is rightly called a
witch. As for his stories of counterfeit ventriloquists, (and who knows
but some of his counterfeit ventriloquists may prove true ones,) that
is but the threadbare sophistry of Sadducees and Atheists to elude
the faith of all true stories by those that are of counterfeits or feigned.
“The seventh word is ‫ידעוני‬, jidegnoni, which our English
translators render a wizzard. And Webster is so kind as to allow them
to have translated this word aright. Wizzards, then, Webster will
allow, that is to say, he-witches, but not she-witches. How tender the
man is of that sex! But the word invites him to it ‫ידעוני‬, jidegnoni,
coming from scire, and answering exactly to wizzard or wise man.
And does not witch from wit and weet signify as well a wise woman,
as I noted above? And as to the sense of those words from whence
they are derived, there is no hurt herein; and therefore if that were
all, ‫ידעוני‬, jidegnoni, had not been in this black list. Wherefore it is
here understood in that more restrict and worse sense: so as we
understand usually now-a-days witch and wizzard, such wise men
and women whose skill is from the confederacy of evil spirits, and
therefore are real wizzards and witches. In what a bad sense ‫ידעוני‬,
jidegnoni, is understood, we may learn, from Levit. xx. 27, ‘A man
also or woman that hath a familiar spirit, or that is a wizzard,
jidegnoni, shall be put to death, they shall stone them with stones,’
&c.
“The last word is ‫דורש המתים‬, doresh hammethim, which our
translators rightly render necromancers; that is, those that either
upon their own account, or desired by others, do raise the ghosts of
the deceased to consult with; which is a more particular term than
‫בעל אוב‬, bagnal obh: but he that is bagnal obh, may be also doresh
hammethim, a necromancer, as appears in the witch of Endor. Here
Webster by ‫המתים‬, hammethim, the dead, would understand dead
statues; but let him, if he can, any where shew in all the Scripture
where the word ‫המתים‬, hammethim, is used of what was not once
alive. He thinks he hits the nail on the head in that place of Isaiah,
viii. 19, ‘And when they say unto you, seek unto [‫האבות‬, that is, to
‫בעליה אוב‬, such as the witch of Endor was,] them that have familiar
spirits, and to wizzards that peep and that mutter; [the Hebrew has it
‫ המהגים‬and ‫ ;המצפצפים‬that is, speak with a querulous murmurant or
mussitant voice, when they either conjure up the spirit, or give
responses. If this be to ‘peep like a chicken,’ Isaiah himself peeped
like a chicken, xxxviii. 14,] should not a people seek unto their God?
for the living, (‫אל המתים‬,) to the dead?’ Where hammethim is so far
from signifying dead statues, that it must needs be understood of the
ghosts of dead men, as here in Deuteronomy. None but one that had
either stupidly or wilfully forgot the story of Samuel’s being raised by
that ‫בעלת אוב‬, bagnalath obh, the witch of Endor, could ever have the
face to affirm that ‫המתים‬, hammethim, here in Isaiah, is to be
understood of dead statues, when wizzards or necromancers were so
immediately mentioned before, especially not Webster, who
acknowledges that ‫שואל אוב‬, shoel obh, signifies a necromancer in
this Deuteronomical list of names. And therefore, forsooth, would
have it a tautology that doresh hammethim should signify so too. But
I say it is no tautology, this last being more express and restrict. And
besides, this enumeration is not intended as an accurate logical
division of witches or witchcraft, into so many distinct kinds, but a
reciting of several names of that ill trade, though they will interfere
one with another, and have no significations so precisely distinct. But
as I said before, this fuller recounting of them is made that the
prohibition in this form might be the surer fence against the sin. And
now therefore what will J. Webster get by this, if doresh hammethim

You might also like