/edu/ - Education

Opening question:
>"Do not feed the trolls"

Does ignoring work? Can ignoring work? When is this effective, and when isn't it effective?

The Gentleperson's Guide To Forum Spies is a widely-circulated pasta, so its hard to find the best source:
https://web.archive.org/web/20231126215612/https://cryptome.org/2012/07/gent-forum-spies.htm

>>21141
Of course it works, trolls feed on reaction, if there is no reaction they will simply move on.

>how should we handle trolls?
I'll do you one better: what causes trolls? Is it as simple as the Hobbesian theory that the opportunity to be a dipshit without consequences unleashes our innate dipshit tendencies? Personally, this seems very iffy. For my money, it's a function of the incentive structure, which depends on multiple factors. One factor is how people interact with each other on a platform, e.g. (You)s or upvotes. Another is the overall "culture" of the platform's users - over time the major tendencies seem to become magnified if the structure has any positive feedback loop or vicious cycle built into it. If we understand how trolling is encouraged and cultivated as an online culture, we can learn to intentionally design the platform in a way that discourages trolling.

>>21147
Sure, and this is easy to ensure on an individual level or a disciplined community. However, can it scale? Or will there always be someone ignorant enough to take the bait?
How can a culture of bait-avoidance be fostered? Maybe staff enforcement (temp-banning people feeding trolls) or collective education/bullying of troll feeders? The second strategy is vulnerable to sockpuppet false flagging, where a troll pretends to feed themself, so hopefully such a community can recognize that instead of repeatedly taking the bait.

>>21149
>I'll do you one better: what causes trolls?
There is more than one obvious cause, so I can't settle for a simple Internet Fuckwad Theory. Consider political trolls, who are motivated to disrupt a forum out of righteousness rather than the typical attention-seeking or sadism motives.
>One factor is how people interact with each other on a platform, e.g. (You)s or upvotes.
This is a structural element I haven't seen discussed much. Twitter's format encourages short dunks and accessible, populist platitudes for maximum reach and attention. Someone recently linked a philosophy lecture about social media gamification which pointed out moral outrage is basically the ideal way to gain attention on the site.
On the other hand, a voting system, like reddit, penalizes controversy or going against a popular stance. The most agreeable post is rewarded, creating a filter bubble in some contexts.
A forum or chan imageboard, on the other hand, only rewards bumps (for all intents and purposes, the sage feature doesn't really matter). Not only are (You)s a visible rewarding symbol, but even the structure of the platform rewards getting replies, regardless of quality or agreement. Controversy keeps a thread at the top of the catalog, giving more eyes on it and getting more replies. An aggressive troll post, so long as it's not removed by staff or ignored by users, is most likely going to stay at the top of an imageboard. In this sense, it's the opposite of reddit: disagreeable posts are rewarded, edginess rises to the forefront, instead of quickly being buried and never seen again.

>If we understand how trolling is encouraged and cultivated as an online culture, we can learn to intentionally design the platform in a way that discourages trolling.
I think I agree, although I can't help but think of how sites like reddit, which very clearly penalise trolling by giving downvoted users a lower reputation score ('karma') and strongly lowering a disagreeable comment's visibility to the point of hiding posts below a threshold, still have trolling pretty widespread.

>>21151 *
An account's karma score can matter. I haven't posted on there in maybe 5 years so I don't know how normal this is, but many communities only allow accounts with over a certain total karma score to post.
That said, if they are fine in some unrelated communities but trolls in others, their overall score will be positive so the negative karma has no effect on their ability to post, plus a stricter community would just ban them for their troll post anyway, assuming their moderation team works properly.

>>21141
Works in theory but not in practice
just like communism

>>21150
Just do it like wikipedia.

>>21164
That doesn't really talk about trolls, so much as getting rid of fringe idiots and idiot opinions.

The first tactic, the strict quality rules, aren't really applicable to most forums or social media/chat sites. Obviously it's great for those which are explicitly strictly quality-controlled.
As for the second (censorship, in a word), I suppose this is easiest for really polarizing issues, when all the staff are on the same side. Of course, it can also lead to crisis if a rogue staff oversteps and censors something they think is self-evident but other staff or a large number of users don't. It also encourages a filter bubble and limits conversation (that can be a positive for some sites, but an outright contradiction of purpose for others, such as /leftypol/ for an obvious example)

>>21171
>(that can be a positive for some sites, but an outright contradiction of purpose for others, such as /leftypol/ for an obvious example)
All conversation on the internet is already limited or it would be drowned in spam and such.

>>21177
What's your point?

Of course I'm not pretending there's a binary choice between censorship and freeze peach. Even 4chan /b/ was a blatantly censored place, even beyond anti-CP and anti-bot/flood measures. They had censorship measures in-line with their community values, like wordfilters (esp. camgirl->cumdumpster after camgirls flooded in) and arbitrary public banning.
All conversations worth having should be limited, but not all to the same degree.

this might be of interest to you op

>>21147
consensus crack tho

>>21317
That image you keep posting doesn't make sense.

I'm not saying I disagree, I'm saying it's semantically weird.

1) The term 'CONSENSUS CRACKING' (emphasis in original, just like in the image) seems to originate only from The Gentleperson's Guide To Forum Spies >>21142
which claims a certain method is used to develop a crack, a method which doesn't involve posting bait or trolling at all. In fact, it would be smartest to intentionally avoid seeming like either, because the point is to attack your weak planted argument with evidence which appears convincing and widely supported (i.e. the consensus) to the uninformed reader. The rigged argument results in a pre-determined break of consensus being reached in the thread, because one side was intentionally introduced with a weak premise and the other side is artificially inflated with fake accounts. If the cracker were trolling or baiting for reactions, they wouldn't convince the uninformed reader nor be able to fake an anti-consensus, plus it would encourage other forum members to be adverse and then discredit the cracking attempt with real counter-arguments rather than only a rigged one, ruining the consensus crack.
Trolls posting bait is not a consensus cracking attempt, as they do not attempt to plant a conversation which reaches a rigged anti-consensus. If anything, they strive for the opposite - universal opposition to their posts. They reinforce the consensus by making an inflammatory opposition to it for the consensus to unite around, while consensus cracking attempts to manufacture a positive opposition to the consensus.

2) The images disregards that and implicitly reinterprets a 'consensus crack' as a shift in the Overton window of acceptable ideas, so let's be fair and work with that.

But even then, ignoring a shitty unwanted post has the same effect as the regular users themselves forum sliding that unwanted post! It doesn't create any impression that the post's ideas are accepted (let alone consensus!!) if it is completely ignored. Nor does it create that impression if, rather than taking the bait, users refuse to dignify it with a response and simply post laughing anime girls.

Rather, to take bait and pretend it has a right to conversation is to fall for a disruption tactic, and simultaneously draw more attention to dumb trash, and simultaneously validate it as an argument and not a ploy.

3) It's not a 'known COINTEL tactic'.
COINTEL == COINTELPRO. And the one document that links 'consensus cracking' with the word 'COINTELPRO' is that same Gentleperson's Guide. It's right at the top, where consensus cracking is listed as the second in the list of 'COINTELPRO Techniques for dilution, misdirection and control of a internet forum.'
(Obviously, and as implied later in the document, the author isn't referring literally to COINTELPRO, as it officially ended long before internet forums existed, but rather they're using the name COINTELPRO generically to refer to the FBI's ongoing political counter-intelligence operations).
The author of the guide provides no sources for anything anywhere, and it seems like there is no source for what they say at all. It's just a list of trolling techniques someone codified, with the claim that it's being used by 'feds' and 'spooks'. And hey, maybe they are using those tactics, it could be logical for them to, but that's a far cry from it being a 'known tactic'.
If it were a confirmed tactic, it would be simple to prove by giving a single source of 'consensus cracking' which doesn't stem from that Gentleperson's Guide.

*
>>21314
I've seen it before, it's a good watch and I recommend! Thanks for sharing.
The Fediverse, from what I've gathered, has normalized medium/high barriers of entry in large numbers of popular communities (as opposed to high-barrier communities being on the fringes, as opposed to capitalist-oriented platforms making barriers extremely low to aid getting as many users as possible), and being federated encourages people to find their own spaces. Contrast again with typical imageboards, with one of the lowest barriers of entry (no registration) and often leaning towards liberal rules if it's not a specialised community. The Mastodon approach, with federated safe spaces, collaborative moderation, and rapid staff responses, makes trolling more time consuming and less rewarding, I'd assume.