Trolling through Twitter hashtags
Andrei Coste

Context
Internet Trolling has been around ever since there were means that users could communicate with each other over web protocols. It’s deeply rooted in the internet culture and is often considered to be a form of bullying (cyberbullying). The website Know Your Meme says that “the contemporary use of the term is alleged to have appeared on the Internet in the late 1980s, but the earliest known mention of the word “troll” on record can be found in a post on the Usenet newsgroup alt.folklore.urban from December 14th, 1992. The term continued to grow popular in the early 1990s through its usage in the Usenet group alt.folklore.urban and by the late 1990s, the alt.folklore.urban newsgroup had such heavy traffic and participation that acts of trolling became frowned upon. Prior to the widespread use of the term “trolling,” similar behaviors have been observed through “griefing,” the act of intentionally causing distress to other players in an online game, since the days of Multi-User Dungeons (MUDs) in the late 1980s and “flaming,” the act of instigating hostility or unpleasant exchanges in online forums, which also emerged through Usenet newsgroup discussions”. They also state that In the late 1980s, Internet users adopted the word “troll” to denote someone who intentionally disrupts online communities. Early trolling was relatively innocuous, taking place inside of small, single-topic Usenet groups. The trolls employed what the M.I.T. professor Judith Donath calls a “pseudo-naïve” tactic, asking stupid questions and seeing who would rise to the bait. The game was to find out who would see through this stereotypical newbie behavior, and who would fall for it. As one guide to trolldom puts it, “If you don’t fall for the joke, you get to be in on it.” Even though you can find traces of trolling on niche sites such as 4chan or reddit, for more widespread appeal, trolls started using some of the more mainstream channels to spread it. Facebook and Twitter topics have for some time been two of the most widespread communication channels, so it shouldn’t be a surprise if we see a lot of internet trolling going on in these places.

What is a troll?
In Internet slang, a troll is someone who posts inflammatory, extraneous, or offtopic messages in an online community, such as a forum, chat room, or blog, with the primary intent of provoking readers into an emotional response or of otherwise disrupting normal ontopic discussion. The noun troll may also refer to the provocative message itself, as in: "That was an excellent troll you posted." While the word troll and its associated verb trolling are associated with Internet discourse, media attention in recent years has made such labels subjective, with trolling describing intentionally provocative actions and harassment outside of an online context. For example, mass media has used troll to describe "a person who defaces Internet tribute sites with the aim of causing grief to families." The most likely derivation of the word troll can be found in the phrase "trolling for newbies", popularized in the early 1990s in the Usenet group, alt.folklore.urban (AFU). Commonly, what is meant is a relatively gentle inside joke by veteran users, presenting questions or topics that had been so overdone that only a new user would respond to them earnestly. For example, a veteran of the group might make a post on the common misconception that glass flows over time. Long-time readers would both recognize the poster's name and know that the topic had been discussed a lot, but new subscribers to the group would not realize, and would thus respond. These types of trolls served as a practice to identify group insiders. This definition of trolling, considerably narrower than the modern understanding of the term, was considered a positive contribution. One of the most notorious AFU trollers, David Mikkelson, went on to create the urban folklore website Snopes.com. By the late 1990s, alt.folklore.urban had such heavy traffic and participation that trolling of this sort was frowned upon. Others expanded the term to include the practice of playing a seriously misinformed or deluded user, even in newsgroups where one was not a regular; these were often attempts at humor rather than provocation. In such contexts, the noun troll usually referred to an act of trolling, rather than to the author. Early incidents of trolling were considered to be the same as flaming, but this has changed with modern usage by the news media to refer to the creation of any content that

targets another person. The Internet dictionary NetLingo suggests there are four grades of trolling: playtime trolling, tactical trolling, strategic trolling, and domination trolling. The relationship between trolling and flaming was observed in open-access forums in California, on a series of modem-linked computers in the 1970s, like CommuniTree which when accessed by high school teenagers became a ground for trashing and abuse. Some psychologists have suggested that flaming would be caused by deindividuation or decreased self-evaluation: the anonymity of online postings would lead to disinhibition amongst individuals Others have suggested that although flaming and trolling is often unpleasant, it may be a form of normative behavior that expresses the social identity of a certain user group. According to Tom Postmes, a professor of social and organisational psychology at the universities of Exeter, England, and Groningen, The Netherlands, and the author ofIndividuality and the Group, who has studied online behavior for 20 years, "Trolls aspire to violence, to the level of trouble they can cause in an environment. They want it to kick off. They want to promote antipathetic emotions of disgust and outrage, which morbidly gives them a sense of pleasure." In academic literature, the practice of trolling was first documented by Judith Donath (1999). Donath's paper outlines the ambiguity of identity in a disembodied "virtual community" such as Usenet: In the physical world there is an inherent unity to the self, for the body provides a compelling and convenient definition of identity. The norm is: one body, one identity ... The virtual world is different. It is composed of information rather than matter. Donath provides a concise overview of identity deception games which trade on the confusion between physical and epistemic community: Trolling is a game about identity deception, albeit one that is played without the consent of most of the players. The troll attempts to pass as a legitimate participant, sharing the group's common interests and concerns; the newsgroups members, if they are cognizant of trolls and other identity deceptions, attempt to both distinguish real from trolling postings, and upon judging a poster a troll, make the offending poster leave the group. Their success at the former depends on how well they – and the troll – understand identity cues; their success at the latter depends on whether the troll's enjoyment is sufficiently diminished or outweighed by the costs imposed by the group. Trolls can be costly in several ways. A troll can disrupt the discussion on a newsgroup, disseminate bad advice, and damage the feeling of trust in the newsgroup community. Furthermore, in a group that has become sensitized to trolling – where the rate of deception is high – many honestly naïve questions may be quickly rejected as trollings. This can be quite off-putting to

the new user who upon venturing a first posting is immediately bombarded with angry accusations. Even if the accusation is unfounded, being branded a troll is quite damaging to one's online reputation. In an effort to reduce uncivil behavior by increasing accountability, many web sites (e.g. Reuters, Facebook, and Gizmodo) now require commenters to register their names and e-mail addresses.

What is Twitter?
Twitter is an online social networking service and microblogging service that enables its users to send and read text-based messages of up to 140 characters, known as "tweets". Twitter was created in March 2006 by Jack Dorsey and by July, the social networking site was launched. The service rapidly gained worldwide popularity, with over 500 million registered users as of 2012, generating over 340 million tweets daily and handling over 1.6 billion search queries per day. Since its launch, Twitter has become one of the ten most visited websites on the Internet, and has been described as "the SMS of the Internet." Unregistered users can read tweets, while registered users can post tweets through the website interface, SMS, or a range of apps for mobile devices. Tweets The Twitter account page for Wikipedia, demonstrating the account-customized timeline view showing tweets in reverse chronological order. Tweets are publicly visible by default, but senders can restrict message delivery to just their followers. Users can tweet via the Twitter website, compatible external applications (such as for smartphones), or by Short Message Service (SMS) available in certain countries. While the service is free, accessing it through SMS may incur phone service provider fees. Users may subscribe to other users' tweets – this is known as following and subscribers are known as followersor tweeps (Twitter + peeps). The users can also check the people who are un-subscribing them on Twitter better known as unfollowing via various services. Twitter allows users the ability to update their profile by using their mobile phone either by text messaging or by apps released for certain smartphones and tablets. Twitter has been compared to a web-based Internet Relay Chat (IRC) client. In a 2009 Time essay, technology author Steven Johnson described the basic mechanics of Twitter as "remarkably simple".

As a social network, Twitter revolves around the principle of followers. When you choose to follow another Twitter user, that user's tweets appear in reverse chronological order on your main Twitter page. If you follow 20 people, you'll see a mix of tweets scrolling down the page: breakfast-cereal updates, interesting new links, music recommendations, even musings on the future of education. Users can group posts together by topic or type by use of hashtags – words or phrases prefixed with a "#" sign. Similarly, the "@" sign followed by a username is used for mentioning or replying to other users. To repost a message from another Twitter user, and share it with one's own followers, the retweet function is symbolized by "RT" in the message. In late 2009, the "Twitter Lists" feature was added, making it possible for users to follow (as well as mention and reply to) ad-hoc lists of authors instead of individual authors. The tweets were initially set to a 140-character limit for compatibility with SMS messaging, introducing the shorthand notation and slang commonly used in SMS messages. The 140-character limit has also increased the usage of URL shortening services such as bit.ly, goo.gl, and tr.im, and content-hosting services, such as Twitpic, memozu.com and NotePub to accommodate multimedia content and text longer than 140 characters. Twitter uses its own t.co domain for automatic shortening of all URLs posted on its website. Trending topics A word, phrase or topic that is tagged at a greater rate than other tags is said to be a trending topic. Trending topics become popular either through a concerted effort by users or because of an event that prompts people to talk about one specific topic. These topics help Twitter and their users to understand what is happening in the world. Trending topics are sometimes the result of concerted efforts by fans of certain celebrities or cultural phenomena, particularly Lady Gaga (known as Monsters), Justin Bieber (Beliebers) and fans of the Twilight (Twihards) and Harry Potter (Potterheads) novels. Twitter have altered the trend algorithm in the past to prevent manipulation of this type. Twitter's March 30, 2010 blog post announced that the hottest Twitter trending topics will scroll across the Twitter homepage. Users will also be able to find out how a specific topic became a trending topic.

There have been controversies surrounding Twitter trending topics: Twitter has censored hashtags that other users found offensive. Twitter censored the #Thatsafrican and the #thingsdarkiessay hashtags after users complained that they found the hashtags offensive. Verified accounts In June 2008, Twitter launched a verification program, allowing celebrities to get their accounts verified. Originally intended to help users verify which celebrity accounts were created by the celebrities themselves (and therefore are not fake), they have since been used to verify accounts of businesses and accounts for public figures who may not actually tweet but still wish to maintain control over the account that bears their name - for example, the Dalai Lama.

Research Design
Context Ever since the rise of social media, internet trolling has been there. It’s a pretty weird sociological phenomenon that has been extensively present on image boards like 4chan, Facebook, Twitter and even IRC. Some even consider it as being a form of cyber bullying and, in the last couple of months, it has been extremely present on twitter hashtags. Goal The goal of this research is to find out how trolls and flamers communicate on twitter, how can a hashtag gain huge popularity, even becoming a trending topic on twitter’s main page, even though it started out as a way to make fun of someone or something. Objectives Identifying user typologies Calculating how long does it take before people realize the hashtag was a joke

Research Questions What importance do flamers have in the escalation of a twitter topic? How fast does a hashtag made for trolling become a trending topic? How many users fall into the trolls trap? To what extent do these hashtags stay alive?

Methodology In order for me to complete my research, I will be using The Archivist, a tool developed for downloading and analyzing twitter queries, as well as Tropes, a software for discourse analysis. The hashtags that I will study are #cutforbieber and #cut4bieber. I will download the tweets through The Archivist, export them as an excel compatible text file for crunching the numbers, as well as importing it into tropes in order to see the most relevant keywords and when did they first appear. The period in which I collected my data has been the 7th of January 2013 to the 4th of February 2013.

Limits and Advantages Even though Twitter’s API blocks users from downloading everything that has ever been written on a hashtag, if you leave The Archivist open ever since the twitter topic began, you can get all the info you need without too much hassle. One of the main advantages about using Tropes and The Archivists is that they do automated data analysis, which speeds up the times needed in order to generate results. Anticipated results I believe my research will show some of the typologies found on the internet, a couple of patterns used in order to troll and that hashtags that are made exclusively for trolling don’t last too long.

Analysis
The Importance of flamers in the escalation of a twitter topic #Cut4Bieber or #Cutforbieber was a Twitter troll that was originally organized by 4chan's /b/. /B/tards started posting images of people cutting their arms and wrists on twitter, together with the hashtag #cutforbieber. The troll was intended to let real fans believe they had to cut wrists as a sign of protest against Bieber's pot smoking behaviour (which leaked on the internet earlier, courtesy of TMZ). The Gay Nigger Association of America (which is a trolling organization) decided to step in and take the lead. And it was with success, because shortly after GNAA's member @dutchminati joined in, the whole thing took off, trending on the first spot in the United States and eventually Worldwide in just a couple of minutes. I downloaded the data of both the hashtags, #cut4bieber, respectively #cutforbieber and combined them (because they were both, alternatively, number one on twitter’s trend lists for a couple of days). As it was described above, the main flamer was user @dutchminati, with this tweet: “#cut4bieber please stop smoking weed Justin... i love you so much!”. Some other members of the GNAA started doing their thing and, eventually, this led to these two hashtags becoming the world’s most used for a couple of days. He has been retweeted 40.371 times over the two days that this hashtag was trending

During the period in which I analyzed the tweets, #cutforbieber had 780.178 tweets sent and #cut4bieber had 976.542 tweets sent. Because of their similar names, the hashtags can easily be combined and turned into a single database, so if we were to sum things up, it would give us a number of 1.756.720 tweets sent on both of them combined. Before user @dutchminati tweeted, there were merely 200 tweets on both the hashtags combined. After that, it escalated to about 200.000 in an hour. Other notable users that generated this amount of traffic cannot be identified, because the number of retweets they got was extremely small compared to @dutchminati’s approximately 40.000 retweets. The road to becoming a trending topic Because of the massive amount of tweets that were being sent after @dutchminati’s tweet, the hashtag became a trending topic in about 20 minutes, and became the biggest in the US. In about an hour it became the world’s biggest trending topic, above classics like #explaintomewhy and #30thingsaboutme, both with about 30 tweets per minute. This shows us that a couple of smart trolls with a big enough database of followers can generate huge amounts of traffic on a topic. “Beliebers” that have been trolled Because the main theme of the hashtag was about Justin Bieber, we’re going to refer to the ones that got trolled as “beliebers”, a term used by them to describe themselves. Even though a huge amount of people realized that this hashtag was for trolling purposes was for trolling purposes only, using images that can be easily found on the internet with people cutting themselves, there was a very large number of people that believed it to be true and started self-mutilating and posting pictures of that on twitter. Even though a complete analysis of this would be impossible to do, after watching the pictures that were posted by the top 40 twitter users on these hashtags, you can see that about 60% of them fell into the trap the trolls prepared for them.

The life of a trolling hashtag Even though its peaks were in the first two days since its emergence, the hashtag is slowly dying. Right now, the combined number of tweets fell from approximately 800.000 in its first two days to about 40 per day. This may be a result of the strong media coverage it received after it peaked the trending topics list and the fact that, usually, a trolling becomes rather stale after two days, because a large amount of users appear that unmask this activity. In internet lingo, this is usually referred to as the fail or failing of a troll. We can actually calculate the fail factor of a trolling if we find out when the first unmasker, whom shall be referred from now on as a white knight (term used in 4chan’s /b/ image board), tweeted about the fact that the hashtag was for laughing purposes only and how fast does the number of tweets fall after that. The main perpetrator here was user @saracourtney10, whom after tweeting: “OMG, can’t you retards see this is a joke #cutforbieber #cut4bieber?” on the 8th of January, generated a lot of negative responses on the hashtag, thus the number of tweets started to drop at a fast pace. So fast, that it went from 74.5 tweets per minute to a mere 20 tweets per minute. One of the more interesting things is that the hashtag didn’t die, it just had a smaller number of tweets, over the course of the next 10 days, after it’s unmasking. After those ten days, it literally went to about 40 tweets per day, which is a pretty small number compared to other hashtags that topped the trending list. Another thing that is pretty unusual for a hashtag is the fact that users started posting totally unrelated stuff on it, without any correlation to cutting yourself for Justin Bieber. So, after having all this data, we could easily say that this hashtag had a fail component of about 70%, because the unmasking of the trolls came the second day the hashtag was topping the chart. As a note, it should be considered that, although there were other white knights before @saracourtney10, this user was the first to actually generate reactions from the others, thus earning the title of supreme white knight. Another thing that’s worth mentioning, is the fact that, although the fail factor was pretty big on this hashtag, it’s one of the best that could have been scored. Other twitter trollings tend to die after only a couple of hours after reaching their full potential. This one beat two days, so it’s one of the most successful ones yet. Another important thing that needs to be noted is that although Justin Bieber posted on these hashtags in order to stop his fans

from cutting themselves, he generated fewer retweets than user @saracourtney10. This could also have a thing to do with his fans’ demographic. Types of users Considering the fact that Tropes identified a couple of keywords used by the users that posted in these hashtags, we can place them in the following categories: flamers, trolls, helpless victims, white knights and unrelated. Flamers These can be seen as the generators of all the commotion surrounding a hashtag. These are the initiators, the people that start the hashtag and make it big. We can identify them pretty easily because they used other hashtags besides #cutforbieber or #cut4bieber, mainly #justinbieber and #belibers. They used them in order to raise awareness of their own hashtags, thus creating a butterfly effect. They amount to 4% of the users that posted on this hashtag Trolls They represent the users that continue the work of the flamers, posting fake content that seems to be real. They’re constant use of the hashtag along with links to photos hosted either on twitter’s services or instagram (or any other image sharing site) help the hashtag grow and raise awareness for the potential victims of the trolling. They generally tend to have texts that are similar to those posted by the helpless victims group but, at a closer look, you can easily observe that the pictures they post are downloaded from google images (I managed to find them with a simple google image search using the keywords “cutting” and “selfmutilation”). They amount to 20% of the users posting on this hashtag Helpless Victims These are the ones that actually believe this hashtag is serious and get caught in. They are the ones that post real picture of them cutting themselves. They’re pretty similar to the trolls text wise, but their photos can’t be found anywhere else on the internet. They amount to 63% of the users that posted on this hashtag.

White Knights These users can be called the “party poopers” of the hashtag. They are the ones that try to unmask the hashtag for what it really is and are the primary generators of the “fail” factor. They amount to 5% of the users on this hashtag. Unrelated These can be considered the users that have absolutely nothing to do with the hashtag but still post under it. One good example is user @Zeeboii’s tweet:
“Finally completed a Rubik's Cube in under 60 seconds!!!!#AsianLife #Cut4bieber”

They are generally seen as the only inhabitants of the hashtag after all the trolling had been done. They are the only ones that stay on the hashtag after it has been proven to be a joke and continue to populate it with totally unrelated tweets, considering that they are funny.

Conclusions
As I’ve analyzed these hashtags, I can say that I’ve found mostly the results I had anticipated. There are five categories of users who use these hashtags, the most important being the flamer, the one that gets everything started. Even though there is no user’s manual to trolling, I can easily say that there are patterns as to how this thing works. First we have the flamer that gets things started, usually referencing something that has its roots in pop culture. Then we have the trolls that start spreading the info just like a plague. Then we have the victims of trolling, those that actually believe that the content that the other users generated is true. Then we have the white knights, who prove that the information the victims believed to be true is fake and finally, the unrelated users, who just post things for the lulz. After looking at these users, we can easily see that following the order that one of them comes to prominence is directly correlated to the life of a twitter hashtag used for trolling. Basically, what this means is that the faster a white knight appears, the bigger are the chances that such a topic could fail really fast, afterwards being populated by the unrelated users.