CRICOS No.00213J
Ten Years of Uninterrupted Debate:
The #auspol Hashtag Community,
2014-2023
Axel Bruns and Anand Badola
Digital Media Research Centre
Queensland University of Technology
Brisbane, Australia
a.bruns@qut.edu.au
@snurb_dot_info | Mastodon: @snurb@aoir.social | Bluesky: @snurb.info
CRICOS No.00213J
2026
SUMMER
SCHOOL
DIGITAL MEDIA RESEARCH CENTRE
MON2FEB-FRI6FEB2026
QUEENSLANDUNIVERSITYOFTECHNOLOGY,BRISBANE
Seehereformoredetails
CRICOS No.00213J
https://mediashift.org/2010/03/the-spill-effect-twitter-
hashtag-upends-australian-political-journalism061/
“I can’t tell you how many times I heard
journos admit they ‘better get into this Twitter
thing,’ that fortnight … It was the only service
providing minute-by-minute updates of the very
fluid situation,” Bourke said.
Prehistory:
#spill 2009
Image: Midjourney
#auspol: In Memoriam
CRICOS No.00213J
Introducing #auspol
• Hashtag. Community?
• Amongst the biggest and most persistent hashtags in Australia
• December 2013 to July 2023: 57m tweets, 1.25m unique accounts
• Strong core of highly active accounts: 44.6m tweets (78%) by top 1% most
active accounts
 Indicative of genuine community structures?
• Dataset:
• Continuous data collection (TCAT, TweetQuery, Academic API )
🪦
• Focus on top 1% of accounts by tweet count (12,452 unique accounts) /
top 0.1% of accounts (1,236 unique accounts)
(https://www.smh.com.au/technology/tony-abbott-dismisses-social-media-as-electronic-graffiti-again-20150126-12yg26.html)
CRICOS No.00213J
http://votecompass.com/2013/07/25/are-you-among-australias-most-influential-political-tweeters-votecompass-maps-the-auspol-twittersphere/
CRICOS No.00213J
CRICOS No.00213J
Theoretical maximum:
3,468 posting days
 span from first to last posting day  ↕ count of days with active posting ↕
CRICOS No.00213J
(Images: https://www.abc.net.au/news/2024-01-29/tony-abbott-eats-an-onion/103394716; https://www.gq.com.au/success/career/our-exclusive-interview-with-malcolm-turnbull/news-story/abdba2bbf5fc5e182ec720b5938ef8d2; https://independentaustralia.net/politics/politics-display/scott-morrisons-not-so-good-year-of-2020,14632; https://www.theguardian.com/australia-news/article/2024/jul/11/anthony-albanese-appeals-to-western-sydney-amid-muslim-voting-campaigns-on-gaza-war)
Morrison’s appointment as PM
to start of 2019 election campaign:
Organised agitation against Morrison,
later moving from #auspol to #ausvotes?
federal
election
2016
federal
election
2019
federal
election
2022
enXittification
🤬
Voice referendum
campaign?
CRICOS No.00213J
CRICOS No.00213J
• Assessing similarities between accounts in:
• @mentioning
• Retweeting
• Secondary (non-#auspol) hashtag use
• Domain sharing
• Top 1% accounts, further filtered for:
• Accounts engaging in all four practices
• At least 100 domains, 1000 hashtags,
1000 retweets, 500 @mentions posted
Practice Mapping
Image: Midjourney
Nodes: Twitter accounts
Node colours: Louvain modularity detection
Edges: combined similarities between accounts:
domains + hashtags + retweets + @mentions
CRICOS No.00213J
federal
election
2016
federal
election
2019
federal
election
2022
CRICOS No.00213J
Community?
CRICOS No.00213J
Finding Community
• What makes participants central to a community?
• Engagement: addressing and responding to other participants through @mentions
• Balance between number of sent @mention and received @mention tweets
• Balance between number of accounts @mentioned and accounts @mentioning
• Consistency: maintaining such balance even while posting a lot of @mentions
• Operationalisation:
• Engagement – Sent-Received Indices:
• Tweets: (# sent @mentions - # received @mentions) / (# sent @mentions + # received @mentions)
• Accounts: (# @mentioned accounts - # @mentioning accounts) / (# @mentioned accounts + # @mentioning accounts)
+1 (@mentioning without any response) 0 (balanced) -1 (not responding to @mentions)
• Posting volume:
• Consistency – Balanced Mention Tweets: # total @mentions × (1 - | Sent-Received Index for Tweets |)
CRICOS No.00213J
Finding Community
• #auspol patterns:
• Sent-Received Indices for tweets and
accounts largely move together
• Consistency highest for 0.1% group
• Key groups emerge:
• Institutional accounts: many @mentions
received, but very little reciprocal
engagement
• Peripheral participants: many attempts
to engage, but few responses received
• In-group cliques: very actively
@mentioning, but selective engagement
with a subset of accounts
• Community core: broadly balanced
engagement (slightly more sent than
received)
Lots of @mentions
sent, few responses
received
Lots of @mentions
received, few
responses sent
Many outgoing
@mentions, but only
to a few accounts
Many accounts
@mentioned, but
engaging with few
responses
News,
journalists,
politicians
Peripheral
participants
Rare –
automated /
procedural?
Rare –
spammers?
Community
core
CRICOS No.00213J
Finding Community
• #auspol patterns:
• Sent-Received Indices for tweets and
accounts largely move together
• Consistency highest for 0.1% group
• Key groups emerge:
• Institutional accounts: many @mentions
received, but very little reciprocal
engagement
• Peripheral participants: many attempts
to engage, but few responses received
• In-group cliques: very actively
@mentioning, but selective engagement
with a subset of accounts
• Community core: broadly balanced
engagement (slightly more sent than
received)
CRICOS No.00213J
Observations
CRICOS No.00213J
Observations and Outlook
• Community evolution:
• Strongly committed and very active core of participants
• Genuine community – highly interconnected and interactive
• Gradual churn in community core
• Especially pronounced as PM Morrison comes to power, declining after 2022 election
– indicative of orchestrated opposition campaign? Or genuine reaction to Morrison’s style of politics?
• Next steps:
• Further analysis of evolution in themes and topics over the years
• Practice mapping analysis for selected periods during the ten-year timespan
• Changes in toxicity and other indicators of communicative dysfunction
• Information sourcing and sharing practices (reliable URL data available from Aug. 2016)
CRICOS No.00213J
Thank you!
Image: Midjourney (prompt: #auspol)
MON2FEB- FRI6FEB2025
QUTKELVINGROVECAMPUS
2026
SUMMER
SCHOOL
DIGITAL MEDIA RESEARCH CENTRE
Register now to
attend the
Open to research students in Master or Doctorate programs, as
well as to early career researchers. Join us for a week of inspiring
and thought-provoking training by world-leading DMRC
researchers.
The Summer School focusses on theories, approaches, methods
and digital research skills, and will allow you to build career-long
connections with digital media researchers from around the
world. Delegate registrations open on Monday
13th October
. Please formally register to
attendbynolaterthan5pm AESTFriday
19December2025.
CRICOS No.00213J
This research is supported by the Australian Research Council through the
Australian Laureate Fellowship project Determining the Dynamics of
Partisanship and Polarisation in Online Public Debate and the Australian Future
Fellowship Understanding Intermedia Information Flows in the Australian Online
Public Sphere.
Acknowledgments

Ten Years of Uninterrupted Debate: The #auspol Hashtag Community, 2014-2023

  • 1.
    CRICOS No.00213J Ten Yearsof Uninterrupted Debate: The #auspol Hashtag Community, 2014-2023 Axel Bruns and Anand Badola Digital Media Research Centre Queensland University of Technology Brisbane, Australia a.bruns@qut.edu.au @snurb_dot_info | Mastodon: @snurb@aoir.social | Bluesky: @snurb.info
  • 2.
  • 3.
    2026 SUMMER SCHOOL DIGITAL MEDIA RESEARCHCENTRE MON2FEB-FRI6FEB2026 QUEENSLANDUNIVERSITYOFTECHNOLOGY,BRISBANE Seehereformoredetails
  • 4.
    CRICOS No.00213J https://mediashift.org/2010/03/the-spill-effect-twitter- hashtag-upends-australian-political-journalism061/ “I can’ttell you how many times I heard journos admit they ‘better get into this Twitter thing,’ that fortnight … It was the only service providing minute-by-minute updates of the very fluid situation,” Bourke said. Prehistory: #spill 2009
  • 5.
  • 6.
    CRICOS No.00213J Introducing #auspol •Hashtag. Community? • Amongst the biggest and most persistent hashtags in Australia • December 2013 to July 2023: 57m tweets, 1.25m unique accounts • Strong core of highly active accounts: 44.6m tweets (78%) by top 1% most active accounts  Indicative of genuine community structures? • Dataset: • Continuous data collection (TCAT, TweetQuery, Academic API ) 🪦 • Focus on top 1% of accounts by tweet count (12,452 unique accounts) / top 0.1% of accounts (1,236 unique accounts) (https://www.smh.com.au/technology/tony-abbott-dismisses-social-media-as-electronic-graffiti-again-20150126-12yg26.html)
  • 7.
  • 8.
  • 9.
    CRICOS No.00213J Theoretical maximum: 3,468posting days  span from first to last posting day  ↕ count of days with active posting ↕
  • 10.
    CRICOS No.00213J (Images: https://www.abc.net.au/news/2024-01-29/tony-abbott-eats-an-onion/103394716;https://www.gq.com.au/success/career/our-exclusive-interview-with-malcolm-turnbull/news-story/abdba2bbf5fc5e182ec720b5938ef8d2; https://independentaustralia.net/politics/politics-display/scott-morrisons-not-so-good-year-of-2020,14632; https://www.theguardian.com/australia-news/article/2024/jul/11/anthony-albanese-appeals-to-western-sydney-amid-muslim-voting-campaigns-on-gaza-war) Morrison’s appointment as PM to start of 2019 election campaign: Organised agitation against Morrison, later moving from #auspol to #ausvotes? federal election 2016 federal election 2019 federal election 2022 enXittification 🤬 Voice referendum campaign?
  • 11.
  • 13.
    CRICOS No.00213J • Assessingsimilarities between accounts in: • @mentioning • Retweeting • Secondary (non-#auspol) hashtag use • Domain sharing • Top 1% accounts, further filtered for: • Accounts engaging in all four practices • At least 100 domains, 1000 hashtags, 1000 retweets, 500 @mentions posted Practice Mapping Image: Midjourney
  • 14.
    Nodes: Twitter accounts Nodecolours: Louvain modularity detection Edges: combined similarities between accounts: domains + hashtags + retweets + @mentions
  • 15.
  • 16.
  • 17.
    CRICOS No.00213J Finding Community •What makes participants central to a community? • Engagement: addressing and responding to other participants through @mentions • Balance between number of sent @mention and received @mention tweets • Balance between number of accounts @mentioned and accounts @mentioning • Consistency: maintaining such balance even while posting a lot of @mentions • Operationalisation: • Engagement – Sent-Received Indices: • Tweets: (# sent @mentions - # received @mentions) / (# sent @mentions + # received @mentions) • Accounts: (# @mentioned accounts - # @mentioning accounts) / (# @mentioned accounts + # @mentioning accounts) +1 (@mentioning without any response) 0 (balanced) -1 (not responding to @mentions) • Posting volume: • Consistency – Balanced Mention Tweets: # total @mentions × (1 - | Sent-Received Index for Tweets |)
  • 18.
    CRICOS No.00213J Finding Community •#auspol patterns: • Sent-Received Indices for tweets and accounts largely move together • Consistency highest for 0.1% group • Key groups emerge: • Institutional accounts: many @mentions received, but very little reciprocal engagement • Peripheral participants: many attempts to engage, but few responses received • In-group cliques: very actively @mentioning, but selective engagement with a subset of accounts • Community core: broadly balanced engagement (slightly more sent than received) Lots of @mentions sent, few responses received Lots of @mentions received, few responses sent Many outgoing @mentions, but only to a few accounts Many accounts @mentioned, but engaging with few responses News, journalists, politicians Peripheral participants Rare – automated / procedural? Rare – spammers? Community core
  • 19.
    CRICOS No.00213J Finding Community •#auspol patterns: • Sent-Received Indices for tweets and accounts largely move together • Consistency highest for 0.1% group • Key groups emerge: • Institutional accounts: many @mentions received, but very little reciprocal engagement • Peripheral participants: many attempts to engage, but few responses received • In-group cliques: very actively @mentioning, but selective engagement with a subset of accounts • Community core: broadly balanced engagement (slightly more sent than received)
  • 20.
  • 21.
    CRICOS No.00213J Observations andOutlook • Community evolution: • Strongly committed and very active core of participants • Genuine community – highly interconnected and interactive • Gradual churn in community core • Especially pronounced as PM Morrison comes to power, declining after 2022 election – indicative of orchestrated opposition campaign? Or genuine reaction to Morrison’s style of politics? • Next steps: • Further analysis of evolution in themes and topics over the years • Practice mapping analysis for selected periods during the ten-year timespan • Changes in toxicity and other indicators of communicative dysfunction • Information sourcing and sharing practices (reliable URL data available from Aug. 2016)
  • 22.
    CRICOS No.00213J Thank you! Image:Midjourney (prompt: #auspol)
  • 23.
    MON2FEB- FRI6FEB2025 QUTKELVINGROVECAMPUS 2026 SUMMER SCHOOL DIGITAL MEDIARESEARCH CENTRE Register now to attend the Open to research students in Master or Doctorate programs, as well as to early career researchers. Join us for a week of inspiring and thought-provoking training by world-leading DMRC researchers. The Summer School focusses on theories, approaches, methods and digital research skills, and will allow you to build career-long connections with digital media researchers from around the world. Delegate registrations open on Monday 13th October . Please formally register to attendbynolaterthan5pm AESTFriday 19December2025.
  • 24.
    CRICOS No.00213J This researchis supported by the Australian Research Council through the Australian Laureate Fellowship project Determining the Dynamics of Partisanship and Polarisation in Online Public Debate and the Australian Future Fellowship Understanding Intermedia Information Flows in the Australian Online Public Sphere. Acknowledgments