Hacker News Reader

Comment by lapcat | Hacker News Reader

A lot of people here in the comments are getting hung up on semantics, the definition of "troll", but the article author addresses this directly:

> Trolling can refer to a variety of online behaviour. In some circumstances, the intent of the trolling behaviour may even be to amuse and entertain. However, in my research, I have explored trolling as a malevolent behaviour, where the troll wants to hurt their online victim.

You can argue that "troll" isn't the right term, but then what is the right term for this? I think it's reasonable to use and adapt the term for this purpose, for lack of another more specific term. The article is describing a subset of trolls.

View on HN

I welcome authors who lay out their assumptions and definitions before they construct their argument. We see exactly the same in mathematics and we're obligated to generalize the same standard and flexibility to other sciences.

I can't count the number of times I've read a math paper which says to the effect, "If the axiom of choice is true then such and such follows." In some ways math has even greater liberty in terms of conjectures - statements assumed true until proven.

All I'm asking is for logical consistency where math and other sciences are held to the same standards regardless of what they are.

I agree that laying out assumptions are essential and foundational and should be demanded of all serious academic work.

That said, different disciplines have different standards and practices about what should be surfaced.

Two examples:

1. Anthropologists describe the context and behaviors of a society. They can aspire to neutrality, but they are biased observers, in terms of what they notice and comment on. This is why it is especially useful for the observer to reflect on their preconceived notions and be transparent about them. This allows the reader to better understand what the "lens" observes.

2. I believe that all papers using statistical models should be fully transparent about, at least: (1) preconceived notions about what associations the authors thought would exist; and (2) how many functional forms were tried and discarded before arriving at the models used for the published results.

The right term is "griefing".

Semantics do matter. If you read a medical article that said "the lungs are the organ of the body that pump blood around, other authors may use the term 'heart'", you wouldn't pay much attention to it, right?

> The right term is "griefing".

I've never heard that one before, despite being online for decades.

It seems to be specific to online gaming, which I don't participate in.

Griefing is usually in the context of a game, and griefing usually involves ruining the fun someone is having in that game. It's generally more benign than anti-semitism or promoting racism online.

As semantics matter, in your sentence "Semantics do matter." it would have been more better to say "Word choice matters." because semantics is the meaning of the word used (or the study thereof), not the actual word used.

You are not objecting, I believe, to the author's intended meaning--you seem to have clearly understood that and believe the different word would have been better used.

Cyberbully?

Yeah, this seems like the correct term for what the paper is describing. Or heck, just 'bully' in general, since that's what it'd be referred to if offline.

As much as the meaning seems to have drifted, trolling and bullying are two different issues, not different names for the same one.

I interpret cyberbullying as more targeted harassment, where you single out someone specifically, likely someone known by the harasser, whereas trolling would be more untargeted harassment involving strangers.

It's like trolling for fish is just trying to catch some fish but not any specific fish a la Moby-Dick (which admittedly is not a fish but rather a mammal, of course).

Reply on news.ycombinator.com