Authors:
(1) Anh V. Vu, University of Cambridge, Cambridge Cyberry Center ([email protected]);
(2) Alice Hutchings, University of Cambridge, Cambridge Cyberry Center ([email protected]);
(3) Ross Anderson, University of Cambridge and Edinburgh University ([email protected]).).
Table
Summary and 1 Introduction
2. Exhaustion and influences
2.1. Related to work
2.2. Interruption of Kiwi farms
3. Methods, data sets and ethics and 3.1. Forum and Discussions of IMAGEBOARD
3.2. Telegram conversations and 3.3. Web traffic and search trend analytics
3.4. Web community and 3.5 tweets. Licensing data
3.6. Ethical considerations
4. Impact on forum activity and traffic and 4.1. The effect of major disorders
4.2. Shifting the platform
4.3. Fragmentation
5. Impact on the stakeholders concerned and 5.1. The community that started the campaign
5.2. The answers of the industry
5.3. Forum operators
5.4. Forum members
6. Tensions, challenges and consequences and 6.1. The effectiveness of the alarm
6.2. Censorship versus freedom of expression
6.3. The role of the industry in moderating the content
6.4. Political influence
6.5. Restrictions and future work
7. Conclusion, recognition and references
Add A.
The alarm campaign began on Twitter on August 22, 2022 with the tweets posted down the Hashtag #droppkiwifarts. We collected the main tweets plus related metadata, such as posting time and reactions (eg answers, retwets, likes and quotes) using SNSCRAape, open source Python framework for social networking companions.[11] Because they use Twitter APIs as a basic method, the data is likely to be completed. We collected 3,886 users of 11,076 tweets, covering the entire campaign period. This information helps us to understand the community's reaction throughout the campaign when the industry took and the forum recovers. Without the hashtag #dropkkiwifarts, there may be more related tweets that we do not know, but it is impossible to scan the entire Twitter space. It is likely that the measured trend of our collection is representative as the campaign gathered around the hashtag.
3.5. Licensing data
Our data sets and data collection and analysis scripts are available to academics as well as an interactive web portal to help those who do not have the technical skills to access our data [67]To. But as members of both scientists and actors such as a forum may be exposed to risk and damage [68]We refuse to make our data publicly accessible. Our regular practice is at the Cambridge Cybercrime Center to require our licensers to sign to prevent abuse to ensure that data is addressed appropriately and keep us up to date on the results of research [69]To. The history of sharing such sensitive data and sharing specific procedures is carefully created with legal academics, university lawyers and a specialized external counselor to allow data to be shared in a number of jurisdictions.
3.6. Ethical considerations
Our work officially approved our institutional ethics overview Council (ERB) to collect and analyze data. Our data sets are collected from publicly available forums and channels that are accessible to everyone. We collected the forum when it was hosted in the US; 2022. According to the US case, scraping public data is legal [70]To. Our scraping method does not violate any regulation or have negative consequences for targeted websites, such as bandwidth congestion or disabling the service. It would be impractical to send thousands of messages to obtain consent from all members of the forum and telegram; We assume that they are aware that their activities in public web sites are widely available.
Contrary to some of the earlier work on web forums, we name the forums studied in this thesis. Pseudonyming for the forum name is meaningless due to the high -level campaign being studied. Thus, we avoid the pretense that the forum is not identifiable and that we focus on the potential damage to our research scientists and those involved. We designed our analysis for ethically and joint action by presenting only a redundant behavior to avoid the private and sensitive information of the people inferred. This is in line with the British Criminology Society about ethics [71]To.
Scientists may be in danger and may result in various elevated digital dangers when working on sensitive resources [68]To do, [72]To. Learning extreme forums can create a higher risk of countermeasures than other forums resulting in mental or physical damage. We have taken actions
Minimize possible damage to scientists and actors involved when they do their studies with human subjects and populations at risk [73]To do, [74]To. For example, we are considering options for anonymous names of authors or using pseudonyms for project -related publications, including this document if necessary. We also refrain from looking at the direct media, which can cause emotional damage; Thus, our scrapers collect the text only by throwing pictures and videos. Although all data sets are widely available and the public can collect them, we refrain from scraping private and protected posts due to the problems of safety and legality.
