I replaced the link with the official Amnesty International blog post, since the original link was mostly just quoting/re-wording it (and with an even more sensationalized title).
I replaced the link with the official Amnesty International blog post, since the original link was mostly just quoting/re-wording it (and with an even more sensationalized title).
Off topic: The donation popup is nasty. It took me to a donation pagr twice because I could not pinpoint the X (the actual letter, not the square its corners constitute). On topic: I know the...
Off topic: The donation popup is nasty. It took me to a donation pagr twice because I could not pinpoint the X (the actual letter, not the square its corners constitute).
On topic: I know the abuse on internet out of experience reading thru thread on Twitter etc., so while I (intuitively and anectodally, to be precise) agree the results, I nevertheless can't take research seriously when it is a survey of volunteering surveyees. Still, it at least can draw people's attention to the issue, so possibly useful in that regard.
TBH I followed the link because I thought it was funded via crowdsourcing. That might be an interesting thing to explore, but there might be negative aspect to it like the funders trying to affect the results or the researchers catering to their crowds trading off scientific honesty. Do we have something like the FSF for scientific research?
The survey was from an older study conducted by Amnesty international. From what I can tell, this one used volunteers to help train an AI to analyze a larger dataset. My biggest issue with the...
The survey was from an older study conducted by Amnesty international. From what I can tell, this one used volunteers to help train an AI to analyze a larger dataset.
My biggest issue with the study is the same as @bel - there's no control group to compare it to. We know that women (and men) experience online abuse. In my mind the more compelling questions are 1. Is the type and severity of abuse different? and 2. Are the effects different? (e.g., are women less likely to actively participate in an online community after experiencing abuse than men?). The most interesting finding in the study is the difference between abuse experienced between women of different races because the comparison allows you to see how severe the problem is for women of colour.
It's good they consider the difference by race, but what about men? Are these numbers significantly different than male journalists and politicians? I imagine they are higher, but not notably so.
It's good they consider the difference by race, but what about men? Are these numbers significantly different than male journalists and politicians? I imagine they are higher, but not notably so.
Eh. They obviously didn't get the results they were hoping for (not the way to do science) with just abusive comments, so made a handwavy definition for 'problematic' to pad the numbers into a...
Eh. They obviously didn't get the results they were hoping for (not the way to do science) with just abusive comments, so made a handwavy definition for 'problematic' to pad the numbers into a story.
I'm confused by your statement, do you disagree with the definition of problematic they used? Or do you disagree with the idea that the problematic statements are in fact problematic and highlight...
I'm confused by your statement, do you disagree with the definition of problematic they used?
Problematic tweets contain hurtful or hostile content, especially if repeated to an individual on multiple or
occasions, but do not necessarily meet the threshold of abuse. Problematic tweets can reinforce negative
or harmful stereotypes against a group of individuals (e.g. negative stereotypes about a race or people
who follow a certain religion). Such tweets may still have the effect of silencing an individual or groups of
individuals.
It’s important to acknowledge that problematic tweets may qualify as legitimate speech and would not
necessarily be subject to removal from the platform. We included problematic tweets because it is
important to highlight the breadth and depth of toxicity on Twitter in its various forms and to recognize the
cumulative effect that problematic content may have on the ability of women to freely expressing
themselves on the platform.
Or do you disagree with the idea that the problematic statements are in fact problematic and highlight toxicity of a platform? And what's your criteria for "the way to do science"?
"More than 6,500 digital volunteers from around the world then took part in Troll Patrol, analysing 288,000
unique tweets to create a labelled dataset of abusive or problematic content. The volunteers were shown
an anonymized tweet mentioning one of the women in our study, then were asked simple questions about
whether the tweets were abusive or problematic, and if so, whether they revealed misogynistic,
homophobic or racist abuse, or other types of violence. Each tweet was analysed by multiple people. The
volunteers were given a tutorial and definitions and examples of abusive and problematic content, as well
as an online forum where they could discuss the tweets with each other and with Amnesty International’s
researchers."
What do you think is the problem with this methodology?
I haven't read the article so IDK if I agree this or not yet, but I observe labels being used as downvotes here. This is on topic, not a joke, not noise, and not malice either, even if it is not...
I haven't read the article so IDK if I agree this or not yet, but I observe labels being used as downvotes here. This is on topic, not a joke, not noise, and not malice either, even if it is not valid as criticism.
That's one of the things they're designed to be, a "downvote with a reason": https://docs.tildes.net/mechanics#comment-labels And regarding the comment being labeled as noise: Whether or not...
but I observe labels being used as downvotes here
That's one of the things they're designed to be, a "downvote with a reason":
Comment labels serve multiple purposes overall. Tildes has no downvoting, but some labels can effectively act as "downvote with a reason". Labels will also make it possible to support various methods of filtering comment threads, such as both "show this thread without jokes" and "show only jokes from this thread".
And regarding the comment being labeled as noise:
Noise - Comments that don't add anything to the discussion. This includes obvious non-contributing comments like "lol", "I agree", and responses to the headline like "finally!", but can also cover anything where the comment's presence doesn't add anything meaningful.
Whether or not vakieh's comment counts as noise is probably up for debate, however I would argue that it's pretty borderline IMO. It really doesn't add anything meaningful the discussion since it was essentially just a pronouncement with absolutely nothing to back up their claims or offhanded dismissal of the study. They also haven't bothered to respond to the reply critical of their comment either which doesn't exactly reflect well on them, and if they're just here to make pronouncements but not willing to engage when challenged that reduces the value of their contributions even further. Had they responded to the criticism by further expanding on what they meant and backing up their claims I might agree that it shouldn't be labeled as noise... but they didn't.
I replaced the link with the official Amnesty International blog post, since the original link was mostly just quoting/re-wording it (and with an even more sensationalized title).
Off topic: The donation popup is nasty. It took me to a donation pagr twice because I could not pinpoint the X (the actual letter, not the square its corners constitute).
On topic: I know the abuse on internet out of experience reading thru thread on Twitter etc., so while I (intuitively and anectodally, to be precise) agree the results, I nevertheless can't take research seriously when it is a survey of volunteering surveyees. Still, it at least can draw people's attention to the issue, so possibly useful in that regard.
TBH I followed the link because I thought it was funded via crowdsourcing. That might be an interesting thing to explore, but there might be negative aspect to it like the funders trying to affect the results or the researchers catering to their crowds trading off scientific honesty. Do we have something like the FSF for scientific research?
The survey was from an older study conducted by Amnesty international. From what I can tell, this one used volunteers to help train an AI to analyze a larger dataset.
My biggest issue with the study is the same as @bel - there's no control group to compare it to. We know that women (and men) experience online abuse. In my mind the more compelling questions are 1. Is the type and severity of abuse different? and 2. Are the effects different? (e.g., are women less likely to actively participate in an online community after experiencing abuse than men?). The most interesting finding in the study is the difference between abuse experienced between women of different races because the comparison allows you to see how severe the problem is for women of colour.
It's good they consider the difference by race, but what about men? Are these numbers significantly different than male journalists and politicians? I imagine they are higher, but not notably so.
Eh. They obviously didn't get the results they were hoping for (not the way to do science) with just abusive comments, so made a handwavy definition for 'problematic' to pad the numbers into a story.
Next.
I'm confused by your statement, do you disagree with the definition of problematic they used?
Or do you disagree with the idea that the problematic statements are in fact problematic and highlight toxicity of a platform? And what's your criteria for "the way to do science"?
What do you think is the problem with this methodology?
I haven't read the article so IDK if I agree this or not yet, but I observe labels being used as downvotes here. This is on topic, not a joke, not noise, and not malice either, even if it is not valid as criticism.
Cc: @Deimos
That's one of the things they're designed to be, a "downvote with a reason":
https://docs.tildes.net/mechanics#comment-labels
And regarding the comment being labeled as noise:
Whether or not vakieh's comment counts as noise is probably up for debate, however I would argue that it's pretty borderline IMO. It really doesn't add anything meaningful the discussion since it was essentially just a pronouncement with absolutely nothing to back up their claims or offhanded dismissal of the study. They also haven't bothered to respond to the reply critical of their comment either which doesn't exactly reflect well on them, and if they're just here to make pronouncements but not willing to engage when challenged that reduces the value of their contributions even further. Had they responded to the criticism by further expanding on what they meant and backing up their claims I might agree that it shouldn't be labeled as noise... but they didn't.