Content uploaded by Roel Lutkenhaus
Author content
All content in this area was uploaded by Roel Lutkenhaus on Jul 16, 2020
Content may be subject to copyright.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 1
Stimulating Conversations about Human
Germline Technology
A digital approach to societal debates
9 April, 2020
By Roel O. Lutkenhaus1,2, Jeroen Jansz2, and Martine P.A. Bouman1,2
1: Center for Media & Health
Peperstraat 35
2801 RD Gouda, the Netherlands
tel.: +31 (0)182-549445
2: Erasmus School of History, Culture and
Communication (ESHCC), Erasmus University
Rotterdam
Burgemeester Oudlaan 50
3062 PA Rotterdam, the Netherlands;
tel.: +31 (0)10-4089109
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 2
Chapter 5:
Stimulating Conversations about Human
Germline Technology
A digital approach to societal debates
In November 2018, Chinese biophysicist He Jiankui shocked the world by announcing the
birth of the world’s first-known genetically edited human babies: Lulu and Nana (Marchione,
2018; Regalado, 2018). He claimed to have used the CRISPR-Cas9 technique to edit the germline
genes of two embryos in order to lower their susceptibility to the HIV virus. Jiankui’s work met
significant critique, as he had not resolved the relevant safety and efficacy issues involved in
editing the genes of two human embryos nor was there wide public support for his experiment
(Krimsky, 2019; Normile, 2018). Thereby, he violated most of the conditions for responsible
application of the CIRSPR-Cas9 technique stipulated by the scientific community at the First
International Summit on Human Gene Editing in 2015 in Washington. D.C. (Olson, 2015).
The case of Lulu and Nana illustrates the growing tension between what is technically
possible and what applications are deemed ethical and acceptable as a society—an issue that
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 3
governments and legislators are struggling with worldwide. In their summit statement, the
organizing committee of the Summit on Human Gene Editing therefore urged societal debate in
order to exchange expertise and perspectives between different societal groups, including
“biomedical scientists, social scientists, ethicists, health care providers, patients and their
families, people with disabilities, policymakers, regulators, research funders, faith leaders, public
interest advocates, industry representatives, and members of the general public” (Olson, 2015,
pp. 7–8). The committee suggested that societal debates contribute to societal consensus but did
not elaborate on what this consensus would amount to or how it could be achieved.
Francoise Baylis, one of the members of the organizing committee, later suggested that
societal consensus can be “figured out” by building upon decision-making through consensus
strategies that have been developed by particular communities over the years (2017). For
example, she refers to John Beatty's “no-objection” or “let-it-stand” decision-making (2018),
implying a process that reaches unanimity or a situation in which the opposition feels it is no
longer worthwhile to express their views. Baylis also referred to a process described by the
Women's Encampment for a Future of Peace and Justice (1983) in which each perspective is given
proper hearing and nobody feels misunderstood. Baylis argued such processes will invariably
involve struggle, but this struggle is an inevitable part of the path leading to societal consensus.
The Dutch DNA Dialogue
Currently, in the Netherlands, the topic of human germline modification (HGM) is
discussed by a narrow range of experts and stakeholders, and has not yet involved wider groups
of citizens (Erfocentrum, 2018). The Dutch Ministry of Health, Welfare and Sport (VWS) has
therefore financially supported a consortium of research institutions, health organizations,
knowledge centers, and public advocacy groups to organize a societal debate on HGM in the
Netherlands: the DNA Dialogue (Dutch Ministry of Health Welfare and Sports et al., 2020). The
project’s goal is to organize societal debate about HGM that reaches and involves a broad range
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 4
of societal groups, propelling a wider exchange of topical expertise as well as diverse, real-world
perspectives. The DNA Dialogue aims to facilitate and stimulate processes in which the
perspectives of all societal stakeholders are heard, thereby improving the conditions for
consensus decisions that can be implemented by lawmakers. Therefore, the DNA Dialogue
Consortium (DDC) will ultimately map the most salient opinion and values that surface during
the DNA Dialogue and offer these to Dutch politicians and scientists.
The DDC seeks to draw lessons from earlier societal debates in the Netherlands on topics
such as nuclear energy, genetically modified foods, and nanotechnology. Previous debates have
included public discussion events as well as the production of TV shows, educational materials,
and opinion articles (Krabbenborg, 2012, 2016). An evaluation of a societal debate on
nanotechnology showed that it did not fully succeed in involving ordinary citizens—co-
organizing stakeholders and experts remained stuck to institutional frames of reference
traditionally concerned with informing and educating rather than fully realizing an open debate
featuring the exchange of day-to-day perspectives and experiences (Krabbenborg & Mulder,
2015).
The present-day media landscape also offers opportunities to extend societal debates to
virtual places such as websites, fora, and social media, utilizing interactive media formats to
stimulate conversations (Lutkenhaus et al., 2019b, 2019a). Expanding the debate to such
avenues can contribute to the inclusiveness of societal debates, as it allows for actively reaching
out to communities of audiences beyond the scope of traditional mass media and offline public
discussion events. Furthermore, social media platforms and formats allow audiences to directly
engage with the debate, thereby enabling substantial societal debate to take place online.
We have been commissioned by the DDC to provide formative insights and strategic
suggestions that contribute to the design and implementation of the Dutch DNA Dialogue’s
online media strategy (Lutkenhaus & Bouman, 2019). This chapter (a) describes how we
identified online communities around HGM, genetic technologies, and related topics; (b) provides
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 5
insight into these communities’ views, values, and interests; and (c) identifies potential
collaboration partners in these communities. Furthermore, we argue the ways in which exposure,
influence, and expertise of influential websites, social media pages, and micro-celebrities can be
activated to contribute to a wider exchange of expertise and diverse, real-world perspectives.
Theoretical Background
In recent years, media behaviors have diversified, leading audiences to websites,
platforms, and online communities around various niche interests (Castells, 2008; Couldry,
2008; Toepfl & Piwoni, 2015). There, likeminded audiences engage with each other on topics and
issues of interest (e.g., food, music, fashion)—they consume and interact with media content or
create their own content (Jenkins et al., 2013; Mukerjee et al., 2018). Such media engagement is
often associated with a collective type of sense-making (Kligler-Vilenchik & Thorson, 2015),
playing an important role in the circulation of media content (Jenkins et al., 2013), contributing
to setting the public agenda (Barberá et al., 2015; Boynton & Richardson, 2016), and
renegotiating each community’s culture (Alleyne, 2015). These processes occur against the
backdrop of networks of websites, fora, and social media profiles that provide the infrastructure
for content consumption, engagement, and re-circulation (Jenkins et al., 2013)—or simply, the
infrastructure for online discourse. In these networks, some websites, pages, or social media users
have grown more influential than others and play a role similar to that of opinion leaders in the
classical Two-Step Flow theory by Katz and Lazarsfeld (2006; Lutkenhaus et al., 2019a).
Similarly, some social media users have built a large following by sharing and creating user-
generated content and acquired micro-celebrity status in their respective communities (Senft,
2009; Usher, 2018). Their influence makes them attractive for private and public organizations
seeking to advertise their products, services, (Hearn & Schoenhoff, 2016) or health messages
(Lutkenhaus et al., 2019a), and has kickstarted career trajectories as social influencers (Abidin,
2017). Micro-celebrities are capable of addressing issues in unique and attractive ways and
directly engage with their audiences in the comment sections, allowing for sophisticated
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 6
parasocial interaction and building strong relationships with their fans (Burgess & Green, 2018;
Frobenius, 2014).
Mapping Communities and Finding Social Influencers
The DDC’s aim is to reach and involve online communities in the Dutch DNA Dialogue by
collaborating with influential websites and micro-celebrities. The DDC can initiate such
collaborations to diffuse expertise and stimulate meaningful conversations around the topic of
HGM in the websites’ and micro-celebrities’ respective communities. In order to identify the
online communities that could serve as avenues for public discussion, we mapped online
communities around HGM, genetic technology, and adjacent issues on the open web, Twitter,
and YouTube. Next, we identified the most influential accounts in each community to find
potential collaboration partners. The following sections (a) distinguish the different types of
networks that we retrieved and analyzed and (b) briefly explore the roles they play in today’s
media landscape.
Types of Networks. First, issue networks comprise web pages connected by hyperlinks
around specific themes and issues (Marres, 2017). These networks can be retrieved by following
links on a given set of web pages. Communities in these networks often signify a cluster of pages
that approach an issue in a similar manner. Influential pages in these networks are frequently
linked to other pages, often signifying authority or popularity of a page on the networks’ themes
and issues.
Second, ego networks include connected social media users (Arnaboldi et al., 2013; Myers
& Leskovec, 2014) and comprise the infrastructure allowing users to send and receive messages
and media content (Myers & Leskovec, 2014). Ego networks can be retrieved from platforms such
as Twitter; communities in these networks—which are smaller clusters of closely connected
members—often signify like-mindedness among their members. Influential users in these
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 7
networks are followed frequently and are well-connected, often signifying the social influence of
these users in the network or in a specific community.
Third, recommender networks comprise connected pieces of media content (e.g., related
YouTube videos), offering pathways between related or similar pieces of media content.
Recommender networks can be retrieved from platforms such as YouTube (Rieder, 2015).
Communities signify a certain degree of relatedness that is algorithmically derived from the
media behaviors of each platform’s users, tailored toward keeping audiences engaged with the
platform as long as possible (Helmond, 2015; Rieder et al., 2018). Influential pieces of media
content in these networks are linked with many other pieces of media content and signify
authority.
In reality, issue-, ego-, and recommender networks are interconnected—a user can
seamlessly browse from a web page and Twitter profile to a YouTube video. However, these
networks differ in their nature and can only be retrieved and analyzed separately. In our case
study, we retrieved issue networks of websites using Google Search, ego networks from Twitter,
and recommender networks from YouTube—all of which are services with a large user base in the
Netherlands (CBS, 2016; de Best, 2019). Although these networks are treated separately, it is
likely they overlap. For example, micro-celebrities may publish articles on their websites, link to
videos on YouTube, and engage with fans and foes on Twitter. To interpret the networks and find
potential collaboration partners, it was therefore important to look beyond the platforms from
where the data were originally retrieved and consider the partners’ other digital extensions (e.g.,
Facebook, Instagram, and Snapchat).
Collaboration and Media Formats
Online communities uphold narrative frameworks that comprise systems of symbols,
habits, and worldviews (Alleyne, 2015) and may approach HGM and genetic technology in unique
ways. Although influential websites, Twitter users, or YouTube channels can leverage these
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 8
frameworks to create media content and foster engagement on topics with which they are familiar
(Frobenius, 2014; Senft, 2009; Usher, 2018; Van Eldik et al., 2019), they might not be as familiar
with raising highly technical and ethical topics, such as that of HGM. To effectively collaborate
with micro-celebrities on the DNA Dialogue, partnership arrangements drive the creation of
media content, striking a balance between the interests of the collaboration partners and the
objectives of the societal debate.
Method
Within the DDC, we have been commissioned to study online networks in order to identify (a)
online communities as digital avenues for the DNA Dialogue and (b) websites and micro-
celebrities for possible collaborate. Our methods help map and analyze online communities
around HGM in order to provide shortlists of potential collaboration partners, strongly relying on
a data retrieval and processing pipeline consisting of four parts (see Figure 1.).
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 9
Figure 1.
Data Processing Pipeline.
Note. The data retrieval and processing pipeline uses a core set of key phrases to retrieve and analyze issue-, ego-,
and recommender networks from the open web, Twitter, and YouTube.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 10
During the first step, we defined the scope of our project by creating list of keywords and
phrases that approach the topic of HGM from different angles (e.g., ‘CRISPR-Cas9’, ‘designer
baby’, and ‘transhumanism’). During the second, third, and fourth steps, these key phrases were
used to identify relevant media content via Google Search, Twitter, and YouTube. For each service,
we (a) retrieved connections between websites (Google Search), people (Twitter), and videos
(YouTube); (b) visualized and analyzed the networks in order to identify online communities and
potential collaboration partners; and (c) studied the content circulated in these networks. We
then shared our scripts—including a more thorough description of the underlying steps—on
RPubs’ (Lutkenhaus, 2019).
Scope and Key Phrases
First, we created a mental model comprising lists of keywords and phrases to find relevant media
content via Google Search, Twitter, and YouTube. In order to identify a broad range of
communities, the model distinguished three levels of topical involvement:
1. Core level: Communities with a direct interest in HGM. This includes key phrases
such as ‘CRISPR-Cas9’, ‘human germline modification’, and ‘Lulu and Nana’.
2. Category level: Communities with an interest in genetic technology. This includes
keywords and phrases such as ‘genetic modification’, ‘GMO’, ‘genome sequencing’,
‘perfect baby’, and ‘repair DNA’.
3. Adjacent level: Communities around topics that signify an indirect or potential
interest in HGM or genetic technology, including future applications, related
philosophical questions, or genetic technology in popular culture. This includes key
phrases such as ‘fertility treatment’, ‘transhumanism’, ‘biotech startups’, and
‘gattaca’ (the title of a movie).
We used this model to structure the inputs of our consortium partners, each of whom we
asked to provide possible key phrases, topics, and themes fitting these different levels of interest.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 11
We received suggestions from (a) researchers and ethicists working for the Rathenau Institute, a
Dutch organization focused on research and dialogues related to the social aspects of science; (b)
the Erasmus Medical Center and the Amsterdam Medical Center, both of which are Dutch
university hospitals; and (c) the Center for Media & Health, a Netherlands-based nonprofit
focused on popular media and social change.
Based on their input, we noticed frequently recurring synonyms (e.g., ‘gentech’ vs.
‘genetic technology’); related concepts (e.g., ‘embryo’, ‘baby’, ‘child’, ‘human’, and ‘offspring’);
and different tenses (i.e., ‘genome modification’ vs. ‘modify genomes’). We translated the inputs
to an exhaustive list of key phrases potentially capable of identifying all relevant media content
around the key themes and topics of our project. We achieved this by splitting up key phrases into
groups of synonyms and combining these groups to compose thematic building blocks that
ultimately generated the list of 14,938 unique key phrases (see Figure 2).
Figure 2
Generating the List of Key Phrases
Note: We split the phrases provided by our consortium partners into words; clustered words into groups of synonyms
and/or tenses (e.g., ‘To modify’ or ‘To repair’), used words or groups of synonyms to define building blocks (e.g.,
‘Subject’ or ‘Action’), and combined the building blocks to generate an exhaustive list of keywords.
Next, we categorized specific combinations of building blocks (and the associated key
phrases) into thematic clusters: description (e.g., ‘crispr-cas9’ or ‘human germline modification’);
metaphors and imaginative future (e.g., ‘designer baby crispr’ or ‘perfect human gentech’);
application (e.g., ‘repair genetic defects’); startups (e.g., ‘human germline modification
company’); fertility and reproduction (e.g., ‘IVF alternatives’ or ‘embryo selection’); ethics and
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 12
philosophy (e.g., ‘transhumanism’ or ‘procreative beneficence’); current affairs (e.g., ‘He Jiankui’
or ‘Lulu and Nana’); and popular culture (e.g., ‘Gattaca’ or ‘Brave New World’). Key phrases in
these groups corresponded to different levels of interest. For example, the startups theme
comprises key phrases based on each of the three levels: ‘CRISPR-Cas9 startup’ (core), ‘genome
sequencing startup’ (category), and ‘biotech startup’ (adjacent).
Data Retrieval and Preprocessing
We used Google—the most widely used search engine in the Netherlands (de Best,
2019)—to search for the keywords and phrases, leading us to relevant media content. Google
allows users to combine multiple key phrases into single search queries (i.e., “‘key phrase 1’ OR
‘key phrase 2’ OR ‘key phrase 3’”) as long as the query does not exceed 32 words in total. We
combined our key phrases into 1,886 unique search queries before submitting the queries to
Google, where we disabled personalization and safe search, and searched from an IP address with
no search history. Due to rate limitations, we chose not to use all 1,886 search queries on Twitter
and YouTube. Instead, after filtering our queries for relevance, we selected the key phrases that
returned the most relevant results for each theme and on each level. We determined this by
filtering out keywords and phrases in queries (a) that did not return any results or (b) for which
the first three results provided an insufficient match with the content we were expecting
(‘designer baby’ returning web stores with designer clothes for babies).
After searching for relevant media content, we retrieved the surrounding issue-, ego- and
recommender networks using (a) Ahrefs.com, a service that offers high-quality link reports; (b)
rtweet, an open-source software to programmatically retrieve information about tweets and
Twitter users (Kearney, 2017); and (c) YouTube Tools, an open-source tool to retrieve networks
of related videos (Rieder, 2015). Prior to analysis, connections between the web pages in the issue
network were aggregated to only include links between top-level domains (TLDs; e.g.,
‘www.address.com’) with the number of underlying subpages (URLs; e.g.,
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 13
‘www.address.com/article.html’) as their properties. A more precise description of these steps
can be found on RPubs (Lutkenhaus, 2019).
Network Analysis
We used igraph and Gephi (Bastian et al., 2009; Csardi & Nepusz, 2006) to further process,
analyze, and visualize these networks. In particular, the Louvain algorithm (Blondel et al., 2008)
was used for community detection; Gephi’s ForceAtlas2 algorithm was used to map out the
network; and the centrality measures eigenvector centrality, betweenness, centrality and
PageRank were used to determine the connectedness and social influence of the individual
nodes—the connected entities such as websites, Twitter accounts, or YouTube videos—in the
networks (Barabási, 2016).
To understand the nature of the different communities, we analyzed the URLs, profile
texts, and video titles in each community by computing word occurrence and TF-IDF (Ramos,
2003). Next, we created word clouds, including the 25 most common words, where word size
expresses frequency and the shade of blue expresses the uniqueness of this word compared to
other communities.
Results
Table 1 shows the characteristics of the retrieved and filtered data, while Table 2 reports
the characteristics of the networks and their underlying community structures.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 14
Table 1
Characteristics of the Retrieved and Preprocessed Data
Web
issue network
Twitter
ego network
YouTube
recommender network
Data set before, and after filtering:
Tweets before, and after filtering:
Total
Relevant
Total
Relevant
Total
Queries
1,886
250
(13.3%)
Tweets
17,505
1,302
Videos
867
URLs
4,145
2,564
(61.8%)
Context
1,545
Unique channels
651
TLDs
2,112
1,317
(62.4%)
(Back)links of the URLs:
Users included in the network:
URLs TLDs
N
Backlinks
1,804
1,709
Authors of tweets
736
(26.8%)
Links
15,003
1,140
Authors of context
382
(13.9%)
Common audiences and
interests
1,627
(59.3%)
The results were filtered by query relevance. 1,234 queries did not return any results, and 402 queries did not return any
relevant results.
(Back)links of all the relevant URLs were retrieved using a paid subscription to Ahrefs.com, a platform offering detailed
(back)link reports.
We retrieved tweets published between January 1, 2019 and June 1, 2019 using the shortlist of keywords. Relevant tweets
are published by Dutch users and/or written in the Dutch language.
As we are interested in users that the authors of our tweets are commonly following (interests) or followed by (audiences),
we have only included users in the networks that are connected to at least 10% of our authors.
The conversational context of the tweets (i.e., associated (re)tweets, replies, and quotes).
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 15
Table 2
Characteristics of the Networks and Their Community Structures
Web
issue network
Twitter
ego network
YouTube
recommender network
Total
Relevant
Active
Passive
Total
Total
Relevant
Nodes (websites,
Twitter users, and
YouTube videos)
4,100
3,102
(75.7 %)
1,118
(43.4%)
1,627
(63.1%)
2,577
867
Edges (connections
between nodes)
5,485
5,277
(96.2%)
232,716
5,847
Communities
818
7
5
136
8
Modularity score
0.802
0.32
0.579
Part of the largest cluster of connected nodes in the network.
Communities spanning >5% of the network.
The Twitter network comprised one large component with 5 communities and did not need to be filtered.
Communities spanning > 1.5% of the network.
Authors of the tweets, and authors of tweets in the conversational context (i.e., associated (re)tweets, replies, and
quotes). Passive users include the users that the authors are commonly following (interests) or followed by (audiences).
Issue Network (Web)
Table 3 shows the main characteristics of the communities in the issue network, and the
visualizations in Figure 12 and Figure 4 show a subset of the issue networks.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 16
Table 3
Main Characteristics of the Communities in the Issue Network (Web)
Nodes
Engaged
Top Words
high frequency and/or salience
TLDs
Returned by Google
Search
Web page titles
Knowledge
548
13%
73
14%
thesis, genetic, DNA, know,
biotechnology, crispr
Press
507
12%
74
14%
News, DNA, science,
background, genetic
Social Media and
Universities
404
10%
102
20%
Calendar, gene dna, genetic, brave (new world),
diabetes
Popular Media
278
7%
42
8%
film, boys (from brazil), dna, baby, black mirror,
homo deus
Experts
272
7%
46
9%
news, dna, crispr, lives, engineerable
Health
221
5%
35
7%
kst (parliamentary documents), UMCG (medical
centers), check-up, genetics
Theatre and
Reference Work
218
5%
44
9%
university, lecture, brave (new world), genetic,
theatre
Study Material
162
4%
26
5%
report, term, book report,
science, question
Other
1,490
36%
72
14%
Nodes from communities spanning less than 5% of the network. This mostly concerns small clusters that are not
connected to a giant component.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 17
Note. The issue network, only including Top Level Domains (TLD) with at least two connections (n = 1,689). The
detected communities are distinguished by color and labeled based on the nature of the content in the community.
The size of the nodes expresses the number of times we have found unique URLs on each TLD.
Figure 3
Issues Network
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 18
Figure 4
Issues Network
Note. The same as Figure 12, but instead of TLDs being colored by community, the TLDs are colored according to
whether they were found by googling for our key phrases (green) or not (red).
Communities
Knowledge. The Knowledge community includes a mix of websites from universities,
knowledge centers, interest groups, and the government, as well as independent platforms
focused on knowledge and science. Potential collaboration partners include (but are not limited
to) nemokennislink.nl (a popular science communication platform that is part of the DDC);
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 19
cogem.net (an advisory committee focusing on the potential risks of genetic modification);
biotechnologie.nl (a platform about biotechnology, created by NEMO Kennislink, which is the
organization behind nemokennislink.nl); and waag.org (a networking organization focusing on
science, technology, and the arts).
Press. The Press community comprises a mix of websites of Dutch newspapers, news
magazines, online news magazines, professional press, and other news media. Potential
collaboration partners include newspapers and broadcasting organizations; independent
newsmagazines such as welingelichtekringen.nl and joop.vara.nl; the collaborating public
libraries of the Netherlands (nobb.nl); and npofocus.nl (a portal featuring documentaries by the
Dutch public broadcasting organizations).
Social Media and Universities. The Social Media and Universities community includes a
mix of websites and social media pages of Dutch universities, faculties, and research labs, as well
as three important social media platforms: Facebook, Instagram, and Twitter. Based on our search
results, universities and social media appear to have been clustered together into one community
because of the high level of integration with social media of Dutch university websites. In addition
to the universities, potential collaboration partners include thijmgenootschap.nl (an association
focused on science and philosophy); montesquieu-instituut.nl (an institute focused on the role
of science in society); and brightlands.com (a networking organization focused on bridging
research and business).
Popular Media. The Popular Media community comprises a mix of websites about science
fiction, technology, cinema, and literature—including web stores where movies and books can be
purchased or streamed for free. Potential collaboration partners include (but are not limited to)
moviemeter.nl (a Dutch website for film lovers); bol.com (a popular web store for books, music,
and movies); and peterjoosten.net (a micro-celebrity and public speaker calling himself a
“biohacker and DIY futurist”).
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 20
Experts. The Experts community includes a mix of professional media about germline
modification, genetic modification, as well as science and technology in a broader sense, scientific
journals, and media focused on ethical dilemmas and technology. Potential collaboration
partners include synbio.arnoschrauwers.nl (a blog about biotechnology by a science journalist);
motherboard.vice.com (an online magazine about technology, especially popular among younger
audiences); and c2w.nl (a magazine for professionals in chemistry and life sciences).
Health. The Health community comprises a mix of websites from hospitals; health
organizations focusing on fertility, gynecology, cancer, and clinical genetics; platforms to access
documents such as announcements or laws from the Dutch government; as well as alternative
news websites and satire. In addition to hospitals and health organizations, potential
collaboration partners include biomaatschappij.nl (a website about biotechnology and society)
and speld.nl (the Dutch equivalent of theonion.com).
Theatre and Reference Work. The Theatre and Reference Work community includes a mix
of websites of theaters, concert venues and night clubs, health organizations, and online
reference works for students. The theatres, concerts venues, and night clubs were mostly found
through key words and phrases relating to popular media and current affairs: for example, many
theatres host performances of Brave New World, or public lectures by experts on genetic
technology. In addition to the theatres, concert venues, and night clubs, potential collaboration
partners include ensie.nl and studeersnel.nl (both of which are reference works for students
where information about genetic modification could be linked with the DNA Dialogue).
Study Material. The Study Material community comprises a mix of platforms for students
as well as websites and blogs about science and society. Potential collaboration partners include
(but are not limited to) nl.wikipedia.org and scholieren.com (both of which are reference works
for students where information about genetic modification can be updated or linked with the
DNA Dialogue).
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 21
Ego Network (Twitter)
Table 4 shows the main characteristics of the communities in the issue network; Figure 5
shows a visualization of the ego network, where nodes are colored along the community they
belong to. Figure 6 only shows the authors, distinguishing between authors of tweets returned
by our first search (red) and authors of tweets in the conversational context (orange). The lines
between the nodes signify interactions such as replies, retweets, and quotes.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 22
Table 4
Main Characteristics of the Communities in the Ego Network (Twitter)
Nodes
Engaged
Top Words
high frequency and/or salience
Users
Actively tweeting
Profile texts
Politics & Media
771
28%
279
36%
politics, parliament, media, economics,
cda (christian democatic party, education
Anti-Establishment
722
26%
180
25%
anti, pvv, fvd (right-wing populist
parties), columnist, islam, right-wing,
critical
Flanders
483
18%
239
49%
own name, flanders, belgium, science,
media, expert, university
Health
472
17%
222
47%
health care, research, medical,
netherlands, science,
webcare, hospital
Agriculture
297
11%
163
55%
agriculture, food, environment,
sustainable, bio, green,
innovation, research
Other
35
0%
35
100%
Users without followers and not following any users.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 23
Figure 5
Ego Network
Note. The dots represent users, colored by the community they belong to. Grey dots represent users that did not
tweet, and thus represent interests or audiences that the authors have in common. The communities are labeled
based on analysis of the profile texts of the users in each community. Note that non-tweeting users are colored grey,
and that all nodes are sized along their eigenvector centrality scores—a statistic typically used to express influence.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 24
Figure 6
Interactions in the Ego Network
Note. The same network as Figure 5, but only including the authors and using lines to highlight interactions between
users. Red dots are users included in our core set of tweets, while yellow dots represent the conversational context.
Communities
Politics and Media. Although the Politics and Media community is the largest and most
central, it is also one of the least engaged communities with HGM or other genetic technologies.
In addition to public broadcasting organizations, newspapers, and online news magazines,
potential collaboration partners include (but are not limited to) the chief technology officer (CTO)
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 25
of the city of Amsterdam, several ex-MPs from the Dutch Green party, as well as the Dutch
Christian Democrats.
Anti-Establishment. While the Anti-Establishment is the second largest community, it is
also the least engaged community in the network. Most of their tweets show little relation to the
topics that we searched, mostly concerning replies or quotes—short messages sent to react to
another message—used to push a different agenda. In addition to well-known columnists and
MPs from the anti-establishment parties, potential collaboration partners include the editors of
the online news magazine OpinieZ.nl.
Flanders. The Flanders community is the third largest community and the second most
engaged. It includes Twitter users from Flanders, the Dutch-speaking part of Belgium, a country
which was beyond the scope of this project. A majority of the active users in this community had
Dutch profile texts but conversed in French with co-nationals from the French-speaking part of
Belgium (see the dense cluster of interactions on the top-left in Figure 6).
Health. The Health community is characterized by those with a medical background and
mostly shares news and insights from research. The community mainly uses Twitter to send
messages, as only a small proportion have responded to other Twitter users. In addition to the
VWS, hospitals, public health organizations, and other consortium partners we found among this
community, potential collaboration partners include several online magazines focused on the
health care sector, a wide variety of MDs, as well as medical philosophers and ethicists, medical
universities, and consumer TV shows.
Agriculture. The Agriculture is the smallest, yet most engaged community, sharing and
discussing news and insights about the application of genetic technologies in agriculture.
Potential collaboration partners include (but are not limited to) online magazines about
agriculture, livestock, the environment and sustainability; a well-known documentary producer;
independent journalists and bloggers; as well as members of public advocacy groups focusing on
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 26
agriculture and sustainability, a former forester, and Christian Democratic MPs specialized in
agriculture.
Recommender Networks (YouTube)
Table 5 shows the main characteristics of the communities in the recommender network;
Figure 7 and Figure 8 include visualizations of the recommender network, where nodes are
colored along the community to which they belong.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 27
Table 5
Main Characteristics of the Communities in the Recommender Network (YouTube)
Nodes
Engaged
Top Words
high frequency and/or salience
Videos
Language /
Orientation
Most salient words in video titles
HGM
215
25%
English /
International
crispr-cas, gene, human, babies,
embryos, technology,
jennifer doudna
Gentech
170
20%
Dutch / the
Netherlands
manipulation, genetic, food, gmo,
monsanto, embryo, world,
protest, gentech
Biotech Startups
99
11%
English /
International
biotech, startup, pitch
Fertility
95
11%
English /
International
ivf, embryo, transfer, baby, tube,
process, frozen, infertility, journey
Gentech and Ethics
89
10%
Dutch
engineering humans, genes, think,
science, technology, impact,
bio, ethics
HGM, Popular Media
& Conspiracy
21
2%
English /
International
crispr-cas, gattaca, mosquitoes,
blueprint, immunity,
biohack, chimera
HGM and Ethics
15
2%
Germany
crispr-cas, trans genetic, secret, genetic
manipulation, gene
Gentech & Fertility
11
1%
Dutch / Belgium
A Belgian MP, medical, contraception,
fertility, gattaca
Other
152
18%
Nodes from communities spanning less than 1.5% of the network, mostly concerning small clusters that are not
connected to the giant component.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 28
Figure 7
Related Video Network
Note. Related video network with node size expressing view count. The communities are labeled along their contents,
with flags signifying the contents’ origin or language.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 29
Figure 8
Related Video Network.
Note. The same network as Figure 7, but with node size expressing eigenvector centrality, expressing which videos are
most likely being referred to.
Communities
Human Germline Modification. The HGM community includes videos in English that are
mostly about how CRISPR-Cas9 technology can be applied to edit human genes. This community
is linked to and plays a central role in most of the communities in the recommender network.
Some of the videos are among the most watched in the network. However, as these videos are
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 30
mostly in English and published by channels targeting international audiences, they are beyond
the scope of the project reported in this chapter.
Gentech. The Gentech community includes videos in Dutch that explain genetic
technology in a light way, while a smaller subset wields a more sensationalistic approach (i.e.,
conspiracy theories). Some videos are linked to the (international) HGM and Gentech & Fertility
communities. Being in Dutch, the videos are among the least-watched in the network. Within
this community, several videos have relatively high PageRank scores, meaning that they are most
likely referred to by other well-connected videos (Xing & Ghorbani, 2004). Potential collaboration
partners include (but are not limited to) the YouTube channels of various Dutch TV shows, a series
of biology classes for high school students, an online science magazine, a futurist/technologist
and public speaker, and the YouTube channels of a few newspapers.
Biotech Startups. The Biotech and Startups community includes videos in English about
biotech startups, such as show reels, reports, and short documentaries. This community is not
linked with the other communities in the network. Although this community does not include
Dutch channels, it includes an online magazine covering the European Biotech Industry as a
potential collaboration partner.
Fertility. The Fertility community includes videos in English about fertility treatments
such as IVF as well as video dairies by couples undergoing IVF. This community is not linked with
other communities in the network and its videos are among the most-watched and most-linked.
Without any Dutch channels, this community is beyond the scope of the project reported in this
chapter.
Gentech and Ethics. The Gentech and Ethics community includes videos in Dutch,
including speeches, reports, and short informational videos highlighting the ethical aspects of
genetic technology (e.g., the altering of human species). The videos are published by Dutch and
Belgian channels, and some videos are linked to the HGM community. Being in Dutch, the videos
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 31
are not commonly watched and do not have high PageRank scores. Potential collaboration
partners include (but are not limited to) public speakers who specialize in science and society,
online science magazines, public speaking events, theatres, and public debate centers.
HGM, Popular Media & Conspiracy. The HGM, Popular Media & Conspiracy community
is a small cluster with videos in English about CRISPR-Cas9 technology, of which are loosely
connected to the (international) HGM and (Dutch) Gentech communities. Most videos are
characterized by a sensationalistic approach, lean toward conspiracy theories, and refer to
popular media (e.g., ‘gattaca’).
HGM and Ethics. The HGM and Ethics community is a cluster of videos in German about
CRISPR-Cas9 technology, of which some are connected to the HGM community. This community
does not include any Dutch channels.
Gentech & Fertility. The Gentech and Fertility community is a small cluster including
interviews with a Belgian member of the European Parliament. This community does not include
any Dutch channels.
Discussion
In this chapter, we explored online networks to find new avenues and potential
collaboration partners for a wide and inclusive societal debate on HGM in the Netherlands. Our
results include unexpected options for online collaborations as well as collaborations beyond the
Internet. Although we expected to find blogs, vlogs, and websites of micro-celebrities, our results
also include the websites; YouTube channels; as well as Twitter, Instagram, and Facebook
accounts of health and public advocacy organizations, businesses, hospitals, libraries, and news
and science magazines. These comprehensive results could be explained by the fact that many of
these organizations, businesses, and institutions use multiple channels to represent themselves
in the virtual domain.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 32
Content formats and partnership arrangements
To collaborate with potential partners, the DDC must establish partnership arrangements
and develop media formats that stimulate the exchange of expertise and perspectives around
HGM. To shed light on a wide range of real-world perspectives, the consortium has studied real-
world scenarios that reflect risk and safety issues as well as ethical aspects of the prospective
application of HGM (van Baalen et al., 2019). This has resulted in several scenarios and
‘vignettes’, connecting real-world perspectives to expertise and providing angles to produce
media content as well as the organization of public discussion events.
In collaborations with micro-influencers, the scenarios and vignettes can serve as
dialogue starters—micro-celebrities may sketch or enact a scenario, share their views on its
implications, or ask audiences for their opinion, doubts, and questions (Lutkenhaus et al., 2019c).
Moreover, micro-celebrities may follow up on their audiences’ responses by sharing their views
on these opinions or by responding to doubt and questions, ideally in collaboration with topical
experts or other stakeholders in order to approach the topic from numerous angles.
To prevent the spread of misinformation, it is important to establish partnership
arrangements that enable the DDC to introduce topical expertise—either by double- checking for
factuality or prospectively providing briefing documents and source materials. Micro-celebrities
have built their social capital based on what their peers appreciate (Frobenius, 2014; Senft,
2009; Usher, 2018; Van Eldik et al., 2019); as such, they should be responsible for creating the
media content.
In addition to micro-celebrities and online magazines, our results included channels
spanning organizations, businesses, and institutions. Some point to or offer specific products,
events, and (public) services (e.g., online bookstores, science festivals, and reference works). From
a content strategy perspective, this offers various opportunities for content formats and
partnership arrangements.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 33
For example, in the Popular Media community, we identified web stores selling books,
music, and movies. Web stores seem unlikely collaboration partners in the organization of a
societal debate. However, if we see them as the digital equivalent of bookstores—also known to
frequently host public discussion events—web stores become interesting venues to extend the
DNA Dialogue. This can be achieved in various ways. For example, the web store might ‘tag’ their
books and movies about genetic technology and gather them on a ‘landing page’. When
customers search for books and movies about genetic technologies, they would be directed to this
page, from where the web store could provide links to digital extensions of the DNA Dialogue,
adding contextual entry points.
Additionally, the DDC could offer their expertise to the editors of each web store’s content
marketing teams managing their digital extensions (e.g., Facebook pages, YouTube channels, and
online magazines). This would enable the editors to write interesting features about, for example,
which fictional genetic technologies have become (or are about to become) reality, thereby
providing links to products in their web store and to digital extensions of the DNA Dialogue. In
doing so, the web store would position itself as a ‘hub of knowledge’, which is interesting from a
branding perspective, while also contributing to its corporate social responsibility profile. In these
ways, collaborations would not only use each collaborations partners channels to diffuse
information, but also strike a balance between the interests of the collaboration partners and the
objectives of the societal debate.
Similarly, the Popular Media community includes multiple online magazines about
cinema. In these magazines as well as their digital extensions, content formats can tap into the
public consciousness of cinema lovers by using movies or movie scenes about genetic technology
as conversation starters. For example, in a scene from Jurassic Park, Dr. Hammond, the park’s
owner, and Dr. Malcolm, a visiting scientist who Dr. Hammond wants to affiliate with the park,
have a fierce discussion about whether it was morally right to ‘resurrect dinosaurs’. Dr. Hammond
(“How can we stand in the light of science, and not act?”) and Dr. Malcolm (“You wield genetic
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 34
power like a kid that has just found its dads gun”) represent two poles in the public debate about
HGM. Accordingly, the scene can be used to incite a debate on genetic modification among
cinema lovers.
On YouTube and Twitter, we found various micro-celebrities who wield a sensationalistic
tone, with a minority sharing myths or misinformation. It is easy to dismiss these micro-
celebrities as unreliable partners; yet in doing so, we would exclude their audiences—people
consuming their content on YouTube—from the DNA Dialogue. Instead, the DDC could approach
these micro-influencers and seek collaboration by offering its expertise to address myths and
misinformation or by affording channel owners the opportunity to interviews topical experts, in
which experts can respond to the channel owners’ doubts and questions.
Finally, offline collaborations might be needed to boost the salience of the DNA Dialogue
in some communities. For example, in the Theatre and Reference Work community, we found
performances and public discussion events on the website calendar pages of various theatres and
public libraries. This is an important insight in itself, as the consortium has not previously
considered organizing performances or public discussion events with theatres. Moreover, by
doing so, it is likely the DNA Dialogue becomes more prominent in this community through
calendar pages as well as digital extensions of theatres and public libraries.
Methodological limitations
Building in room for reflexivity allowed us to adapt to the circumstances, ultimately
providing insights that mattered most to the DDC. During the process, several issues emerged
that may require attention in future research.
First, our goal was to identify websites, micro-celebrities, and other potential
collaboration partners in online networks using a comprehensive list of key phrases. During the
earliest phase of this study, the members of the DDC were asked to share ideas for themes, topics,
and angles that ultimately led to our list of keywords and phrases. However, by creating a list of
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 35
key phrases ourselves, we only searched in areas that the DDC was already aware of, and the
resulting media content may have only spanned a small part of all possible real-world views and
associations.
The Rathenau Institute, also part of the DDC, studied real-world scenarios that reflect
risk and safety issues as well as ethical aspects of the prospective application of HGM (van Baalen
et al., 2019). This report became available a few weeks after we presented our final results to the
DDC , and, in retrospect, provides additional input for the list of keywords and phrases. For future
societal debates, we recommend to thoroughly study the real-world scenarios associated with
the debate’s topic before generating the keywords and phrases.
Second, by using Google Search for key phrases, we strongly relied on their search
algorithm—even though we disabled personalization, safe search, and searched from an IP
address with no search history. Nonetheless, Google is commonly used by a large majority of
Dutch Internet users, and thereby provides a shared interface to navigate the Internet. In our
study, we approached these search results as a snapshot of this common interface, being aware
that the results might vary due to algorithmic personalization and the high level of activity in the
networks. Especially for formative studies (Bouman, 1999), we believe this snapshot is
meaningful as it represents how the Internet actually looks to its end users.
Third, we used a condensed list of key phrases to search for content on Twitter and
YouTube. For each theme and level, we selected the key phrases for which Google returned the
most relevant results. Consequently, the results from Twitter and YouTube appeared to be equally
relevant and did not require us to do any manual filtering. This addressed an important problem,
as there was no time to manually filter three separate data sets with the media strategist waiting
for our inputs. However, a narrower set of keywords may have led to a smaller set of results.
Lastly, links between web pages do not always signify meaningful pathways of
information but can also be used as a tactic to boost web pages’ significance in Google’s search
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 36
results. This practice is called “link building” (Moogan, 2020), and it is likely that the issue
network includes noise as a result. We have aimed to diminish the noise by filtering nodes with a
small number of connections—a rather crude approach. Future researchers could explore the
practical use of various filtering approaches (Waldherr et al., 2017).
Conclusion
In recent years, experts have called for societal debates about HGM. Societal debates such as the
Dutch DNA Dialogue do not aim to reach unequivocal societal agreement on specific
implementations; rather, they aim to propel the exchange of expertise and a wide range of
stakeholder perspectives. As such, the DNA Dialogue could enhance the conditions for informed,
widely supported consensus decisions that can be implemented through quality standards and
legislation.
Currently, new media platforms can be used to extend the scope of societal debates by
reaching out to online communities and to create interactive media formats that enable and
invite audiences to engage with each other on the topic of HGM. To achieve this, there is a crucial
role to be played by media strategies. Our case study provides insights and suggestions for the
media strategy of the Dutch DNA Dialogue, helping to widen and deepen the societal debate in
three specific ways: (a) we identified online communities as avenues for public dialogues, often
beyond the reach of traditional mass media; (b) our results contribute to a deeper understanding
of how communities relate to HGM, helping the DDC to tailor media formats to their preferences;
and (c) we provide a wide variety of potential collaboration partners such as micro-celebrities,
magazines, web stores, fan communities, and reference works.
This chapter shows that formative analysis of online networks around socio-
technological issues is relevant, important, and ready to be applied in future societal debates.
This is especially evident nowadays, as the organization of societal debates seems to fit within a
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 37
larger trend of public engagement, playing an integral part in innovation trajectories in Europe,
Australia, and North America (Fisher et al., 2015).
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 38
References
Abidin, C. (2017). #Familygoals: Family influencers, calibrated amateurism, and justifying young
digital labor. Social Media and Society. https://doi.org/10.1177/2056305117707191
Alleyne, B. (2015). Narrative Networks. Storied Approaches in a Digital Age. SAGE Publications
Ltd.
Arnaboldi, V., Conti, M., Passarella, A., & Pezzoni, F. (2013). Ego networks in Twitter: An
experimental analysis. 2013 Proceedings IEEE INFOCOM, 3459–3464.
https://doi.org/10.1109/INFCOM.2013.6567181
Barabási, A.-L. (2016). Network Science (3rd printi). Cambridge University Press.
Barberá, P., Wang, N., Bonneau, R., Jost, J. T., Nagler, J., Tucker, J., & González-Bailón, S. (2015).
The Critical Periphery in the Growth of Social Protests. PLOS ONE, 10(11), e0143611.
https://doi.org/10.1371/journal.pone.0143611
Bastian, M., Heymann, S., & Jacomy, M. (2009). Gephi: An Open Source Software for Exploring
and Manipulating Networks.
http://www.aaai.org/ocs/index.php/ICWSM/09/paper/view/154
Baylis, F. (2017). Human germline genome editing and broad societal consensus. Nature Human
Behaviour, 1(6), 0103. https://doi.org/10.1038/s41562-017-0103
Beatty, J. (2018). Consensus: Sometimes it doesn’t add up. In S. B. Gissis, E. Lamm, & A. Shavit
(Eds.), Landscapes of Collectivity in the Life Sciences (pp. 179–198). The MIT Press.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 39
Blondel, V. D., Guillaume, J.-L., Lambiotte, R., & Lefebvre, E. (2008). Fast unfolding of
communities in large networks. https://doi.org/10.1088/1742-5468/2008/10/P10008
Bouman, M. (1999). The Turtle and the Peacock: Collaboration for prosocial change. The
Entertainment-Education strategy on television [Landbouwuniversiteit Wageningen].
https://library.wur.nl/WebQuery/wurpubs/fulltext/164907
Boynton, G., & Richardson, G. W. (2016). Agenda setting in the twenty-first century. New Media
& Society, 18(9), 1916–1934. https://doi.org/10.1177/1461444815616226
Burgess, J., & Green, J. (2018). YouTube: Online Video and Participatory Culture (2nd ed.). Polity
Press.
Castells, M. (2008). The New Public Sphere: Global Civil Society, Communication Networks, and
Global Governance. The ANNALS of the American Academy of Political and Social Science,
616(1), 78–93. https://doi.org/10.1177/0002716207311877
CBS. (2016). Internet; toegang, gebruik en faciliteiten. StatLine.
http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=83429ned&D1=0%2C2-
5%2C7%2C12-14%2C29-30%2C32%2C44&D2=0-
2%2C6&D3=0&D4=l&HDR=T&STB=G1%2CG2%2CG3&VW=T
Couldry, N. (2008). Mediatization or Mediation? Alternative Understandings of the Emergent
Space of Digital Storytelling. New Media & Society, 10(3), 373–391.
https://doi.org/10.1177/1461444808089414
Csardi, G., & Nepusz, T. (2006). The igraph software package for complex network research.
InterJournal, Complex Sy, 1695. http://igraph.org
de Best, R. (2019). Market share distribution of search engines in the Netherlands from 2009 to
2018. Statista. https://cdn.statista.com/statistics/1040219/market-shares-of-search-
engines-in-the-netherlands/
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 40
Dutch Ministry of Health Welfare and Sports, Erasmus MC, Erfocentrum, NEMO Kennislink, NPV,
& Instituut, R. (2020). DNA Dialoog. Over het genetisch aanpassen van embryo’s.
https://dnadialoog.nl/
Erfocentrum. (2018). Kick-Off DNA Dialoog.
Fisher, E., O’Rourke, M., Evans, R., Kennedy, E. B., Gorman, M. E., & Seager, T. P. (2015). Mapping
the integrative field: taking stock of socio-technical collaborations. Journal of Responsible
Innovation, 2(1), 39–61. https://doi.org/10.1080/23299460.2014.1001671
Frobenius, M. (2014). Audience design in monologues : How vloggers involve their viewers.
Journal of Pragmatics, 72, 59–72. https://doi.org/10.1016/j.pragma.2014.02.008
Hearn, A., & Schoenhoff, S. (2016). A Companion to Celebrity. In P. D. Marshall & S. Redmond
(Eds.), A Companion to Celebrity (pp. 194–211). John Wiley & Sons, Inc.
https://doi.org/10.1002/9781118475089
Helmond, A. (2015). The Platformization of the Web: Making Web Data Platform Ready. Social
Media and Society, 1(2). https://doi.org/10.1177/2056305115603080
Jenkins, H., Ford, S., & Green, J. (2013). Spreadable Media: Creating Value And Meaning In A
Networked Culture. New York University Press.
Katz, E., & Lazarsfeld, P. F. (2006). Personal influence: the part played by people in the flow of
mass communications (2nd editio). Transaction Publisher.
Kearney, M. W. (2017). rtweet: Collecting Twitter Data. https://cran.r-
project.org/package=rtweet
Kligler-Vilenchik, N., & Thorson, K. (2015). Good citizenship as a frame contest: Kony2012,
memes, and critiques of the networked citizen. New Media & Society, 18(9), 1993–2011.
https://doi.org/10.1177/1461444815575311
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 41
Krabbenborg, L. (2012). The Potential of National Public Engagement Exercises: Evaluating the
Case of the Recent Dutch Societal Dialogue on Nanotechnology. International Journal of
Emerging Technologies and Society, 10, 27–44.
Krabbenborg, L. (2016). Creating Inquiry Between Technology Developers and Civil Society
Actors: Learning from Experiences Around Nanotechnology. Science and Engineering Ethics,
22(3), 907–922. https://doi.org/10.1007/s11948-015-9660-2
Krabbenborg, L., & Mulder, H. A. J. (2015). Upstream Public Engagement in Nanotechnology.
Science Communication, 37(4), 452–484. https://doi.org/10.1177/1075547015588601
Krimsky, S. (2019). Ten ways in which He Jiankui violated ethics. Nature Biotechnology, 37(1),
19–20. https://doi.org/10.1038/nbt.4337
Lutkenhaus, R. O. (2019). Mapping Media Networks. RPubs. https://rpubs.com/roel-
sbcc/mapping-media-networks
Lutkenhaus, R. O., & Bouman, M. P. A. (2019). De medianetwerken rond kiembaanmodificatie.
https://www.dropbox.com/s/756qrknjeb1racy/De medianetwerken rond
kiembaanmodificatie - Centrum Media %26 Gezondheid.pdf
Lutkenhaus, R. O., Jansz, J., & Bouman, M. P. (2019a). Tailoring in the digital era: Stimulating
dialogues on health topics in collaboration with social media influencers. Digital Health, 5.
https://doi.org/10.1177/2055207618821521
Lutkenhaus, R. O., Jansz, J., & Bouman, M. P. A. (2019b). OUP accepted manuscript. Health
Promotion International. https://doi.org/10.1093/heapro/daz104
Lutkenhaus, R. O., Jansz, J., & Bouman, M. P. A. (2019c). Towards spreadable entertainment-
education: leveraging social influence in online networks. Health Promotion International.
https://doi.org/10.1093/heapro/daz104
Marchione, M. (2018). Chinese researcher claims first gene-edited babies. AP News.
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 42
https://www.apnews.com/4997bb7aa36c45449b488e19ac83e86d
Marres, N. (2017). Digital Sociology. Polity Press.
Moogan, P. (2020). The Beginner’s Guide to Link Building. MOZ. https://moz.com/beginners-
guide-to-link-building
Mukerjee, S., Majó-Vázquez, S., & González-Bailón, S. (2018). Networks of Audience Overlap in
the Consumption of Digital News. Journal of Communication, 68(1), 26–50.
https://doi.org/10.1093/joc/jqx007
Myers, S. A., & Leskovec, J. (2014). The bursty dynamics of the Twitter information network.
Proceedings of the 23rd International Conference on World Wide Web - WWW ’14, 913–924.
https://doi.org/10.1145/2566486.2568043
Normile, D. (2018). Shock greets claim of CRISPR-edited babies. Science, 362(6418), 978–979.
https://doi.org/10.1126/science.362.6418.978
Olson, S. (2015). International Summit on Human Gene Editing: A Global Discussion. In The
National Academis of Sciences, Engineering, and Medicine.
http://go.nature.com/2pwNV7w
Ramos, J. (2003). Using TF-IDF to Determine Word Relevance in Document Queries.
https://www.cs.rutgers.edu/~mlittman/courses/ml03/iCML03/papers/ramos.pdf
Regalado, A. (2018). EXCLUSIVE: Chinese scientists are creating CRISPR babies. MIT Technology
Review. https://www.technologyreview.com/s/612458/exclusive-chinese-scientists-are-
creating-crispr-babies/
Rieder, B. (2015). Introducing the YouTube Data Tools. The Politics of Systems. Thoughts on
Software, Power, and Digital Method. http://thepoliticsofsystems.net/2015/05/exploring-
youtube/
STIMULATING CONVERSATIONS ABOUT HGM VERSION: 9 APRIL 2020
DRAFT: PLEASE DO NOT CIRCULATE OR CITE WITHOUT PERMISSION OF THE AUTHOR 43
Rieder, B., Matamoros-Fernández, A., & Coromina, Ò. (2018). From ranking algorithms to
‘ranking cultures.’ Convergence: The International Journal of Research into New Media
Technologies, 24(1), 50–68. https://doi.org/10.1177/1354856517736982
Senft, T. M. (2009). Camgirls: Celebrity and Community in the Age of Social Networks. Camgirls.
https://doi.org/10.3983/twc.v3i0.120
Toepfl, F., & Piwoni, E. (2015). Public Spheres in Interaction: Comment Sections of News Websites
as Counterpublic Spaces. Journal of Communication, 65(3), 465–488.
https://doi.org/10.1111/jcom.12156
Usher, B. (2018). Rethinking microcelebrity: key points in practice, performance and purpose.
Celebrity Studies, 1–18. https://doi.org/10.1080/19392397.2018.1536558
van Baalen, S., Gouman, J., & Verhoef, P. (2019). In gesprek over het aanpassen van erfelijk DNA
van embry’s. Lessen voor een maatschappelijke dialoog. (p. 165). Rathenau Instituut.
https://www.rathenau.nl/sites/default/files/2019-10/RAPPORT In gesprek over het
aanpassen van erfelijk DNA van embryo%27s_0.pdf
Van Eldik, A. K., Kneer, J., & Jansz, J. (2019). Urban & Online: Social Media Use among
Adolescents and Sense of Belonging to a Super-Diverse City. Media and Communication,
7(2), 242. https://doi.org/10.17645/mac.v7i2.1879
Waldherr, A., Maier, D., Miltner, P., & Günther, E. (2017). Big Data, Big Noise. Social Science
Computer Review, 35(4), 427–443. https://doi.org/10.1177/0894439316643050
Women’s Encampment for a Future of Peace and Justice. (1983). Resource Handbook.
Xing, W., & Ghorbani, A. (2004). Weighted PageRank algorithm. Proceedings. Second Annual
Conference on Communication Networks and Services Research, 2004., 305–314.
https://doi.org/10.1109/DNSR.2004.1344743