Authors:
(1) Anh V. Vu, University of Cambridge, Cambridge Cybercrime Center ([email protected]));
(2) Alice Hutchings, University of Cambridge, Cambridge Cybercrime Center ([email protected]));
(3) Ross Anderson, University of Cambridge and University of Edinburgh ([email protected]).
Ties
Summary and 1 Introduction
2. Travel and impact
2.1. Related work
2.2. Kiwi farmers
3. Methods, data and ethical sets, and 3.1. Forum and image table discussions
3.2. Telegrams cats and 3.3. Traffical and research on web traffic and analyzes
3.4. Tweets made by the online community and 3.5. Data license
3.6. Ethical considerations
4. The impact on forum activity and traffic, and 4.1. The impact of major disturbances
4.2. Displacement of the platform
4.3. Traffic fragmentation
5. The impacts on the stakeholders concerned and 5.1. The community that launched the campaign
5.2. Industry responses
5.3. Forum operators
5.4. Forum members
6. Tensions, challenges and implications and 6.1. The effectiveness of the disturbance
6.2. Censorship against freedom of expression
6.3. The role of industry in content moderation
6.4. Political implications
6.5. Limitations and future work
7. Conclusion, thanks and references
Annex A.
5. The impacts on the relevant stakeholders
We have examined the impacts of the disturbance on the farms of Kiwi itself. This section examines the effects on the relevant stakeholders, in particular the harassed victim, the community who runs the campaign, the industry, the forum operators and the users of the active forum who have displayed at least once. As our ethical approval does not allow the study of individuals, all measures are made collectively on user subsets. In addition to quantitative evidence, we also examine the declarations made by technological companies on the incident qualitatively.
There were 3,886 users in the online community involved in the start of the campaign. Among these, 1,670 users (42.97%) were responsible for around 80% of tweets. There was a sharp increase in tweets and reactions at the start (see Figure 6). The first peak took place on August 25, 2022 with nearly 900 tweets of around 600 users. However, this quickly dropped less than 100 a day after a few weeks when Cloudflare and Ddos-Guard took action, and almost zero two weeks later. The number of tweets specifically mentioning Cloudflare (such as their official account, as well as those for jobs, aid and developers) was around 200 at the start but decreased over time, and fell to zero after taking measures. It lasted about a month until after recovering the forum: we see about 400 tweets mentioning Cloudflare, twice the previous peak and representing almost all these tweets that day. However, these tweets seem to be mainly associated with another campaign counted by the hashtag #Stopdoghate, suggesting a short -lived aberrant value instead of a real peak.
The trans activist who launched the campaign was engaged at first but has become much less active in the publication of new tweets, although she always answered people. Its publication volume was, however, trivial compared to global figures: it only made four tweets the day the campaign started, the number then dropped quickly to one on September 4, 2022 after Cloudflare took measures and zero thereafter. This suggests that although she has triggered the campaign, she may not be the main manager.
We do not see any notable tweets after closing the forum, suggesting a clear loss of interest in continuing the campaign, both of people displaying tweets and people reacting to tweets. The community seemed to be bored quickly after a few weeks when they seemed to have obtained what they wanted – “Kiwi Farms died, and I move on to the next campaign,” tweeted the activist.
5.2. Industry responses
Unlike the measurement of the activity of the forum, there are no such quantitative data to cover the impact on industry players, so we move on to the qualitative analysis of public declarations made by those who have directly tried to terminate the forum. We first comply a list of technological companies involved seen from withdrawal incidents, then examine their websites, news and blogs to identify their official declarations if available. We have repeated the search regularly and the final list consists of four companies: Cloudflare, DDOS-Guard, Diamwall and Harica. We then adopted a deductive approach to understand (1) their accommodation policy, (2) their point of view on Kiwi farms and (3) their reactions to community pressure.
Cloudflare declared their abuse policies on August 31, 2022 without directly mentioning the Twitter campaign [76]. In summary, the company offers a circulation proxy and DDOS protection at many sites (mainly unpaid), regardless of the content hosted, including Kiwi farms. The cabinet maintains that abusive content is not a problem alone, and the forum – although Immoral – always deserves the same protection as other customers, as long as it does not violate US law. Although Cloudflare has the right to refuse Kiwi Farms affairs, they initially considered that the fact because of its content would create a bad precedent, leading to unexpected consequences on content regulations and would make things more difficult for Cloudflare. This could affect the entire Internet, because Cloudflare manages a large part of the network traffic. They did not want to get involved in the online content of the police, but if they were to do so, they prefer to do so in response to a court order instead of a popular opinion. The company had previously abandoned the daily neonazi website [12] and the Extremist 8CHAN card [13] Due to their links with terrorist attacks and mass murders, and a false assertion on the secret support of Cloudflare. They also claimed that the Kiwi farms delete Farms would not delete the contents of hatred, but would only slow it down for a while.
Nevertheless, Cloudflare did a tour in the U a few days later on September 3, 2022, announcing that they would end the service for Kiwi Farms [24]. They explained that climbing the pressure campaign allowed users to be more aggressive, which could lead to crime. They contacted the police in several jurisdictions concerning potential criminal acts, but as the legal process was too slow in relation to the climbing of the threat, they made the decision alone [24]. They have always said that following a legal process would be correct and denied that the decision was the direct result of community pressure. The action of Cloudflare also inadvertently led to the end of a group of neonazi in New Zealand, because it was organized by the same company as the forum [77].
Ddos-Guard's statements on the incident told a similar story [25]. Although they can restrict access to their customers if they violate the acceptable use policy, the moderation of the content is not their obligation (except under a court order) so that they do not need to determine if each site that they protect the law. Diamwall has taken the same line; They said they were not responsible and are unable to moderate the content hosted on websites [26]. They also argued that the termination of services in response to public pressure is not a good policy, but the case of Kiwi farms was exceptional because of its “revolting” content. They also noted that their actions could only delay things but not repair the deep cause, because the forum could find another supplier. The Diamwall declaration was deleted later without any clear explanation; It is now only accessible through online archives.
Unlike the three companies above, Harica – a Greek authority providing certificates for sites .onion – has taken a different line. They confirmed their support for freedom of expression and declared that they would not censure any website, but are obliged to investigate complaints concerning websites violating the law, certificate policy (CP) and the certificate declaration of practice (CPS). After an examination process, on May 15, 2023, they announced that they would revoke the certificates. Onion issued to Kiwi Farms due to concerns about harassment linked to suicides. After having set a period of 3 days for Kiwi Farms to seek a new authority, their support team was targeted by various threats and harassment, but on the basis of a friendly and polished message, stressing that Harica is one of the only two authorities who have issued certificates. [78].
Whether or not it is to block kiwi farms, it is understandable that infrastructure and certificate providers do not want to get involved in content regulation like Facebook and Google must, because moderation is complex, difficult, controversial and costly [79].
