TW: I will be talking about abuse and posting and discussing fairly terrible examples.
This is a baby sloth gif to make sure content starts out further down.
It’s no secret that twitter has an abuse problem. Well, let’s be honest pretty much any tool that allows people to communicate with one another has an abuse problem. To try to remedy this situation Twitter is adding a report abuse button to every tweet in order to make it easier to handle abusive users in response to this petition.
We need to consider when we implement anti-abuse tools that it is protecting the people that need to be protected. I would make the obvious argument that those being abused are the ones that should be protected, however, the definition of abuse varies. The following tweet by Anita Sarkeesian shows that though it is clearly abuse by my personal definition of the word, those reviewing the complaint found no issue with this sort of abusive behavior.
Twitter says "We have found the reported account is currently not in violation of the Twitter Rules at this time" pic.twitter.com/HnJ6gueb3u— Feminist Frequency (@femfreq) July 28, 2013
This is often what happens when users are reported and reviewed by hand. A lot of bad things slip through and those users go on to abuse another day.
So, enter the “report abuse” button and since Anita Sarkeesian is no stranger to abuse we’ll use her as an example again. YouTube’s flag feature is the general pattern for most report abuse systems. The button or flag gets hit a certain number of times and then the content is removed and set to be reviewed. This gives detractors a perfect silencing tool and it’s used as just that. YouTube’s flag feature was used to take down Sarkeesian’s Tropes vs Women in Games video.
The core of the problem is that all of these systems are built on faulty assumptions. The first major faulty assumption being: if a large enough group of people think that something is abusive it must be abusive. This might be true in a egalitarian society and since technology is a “meritocracy” no wonder this is seen as a useful approach, but when discussing oppression as a member of an oppressed group the masses have the numbers and the ability to silence those trying to get the word out. Being silenced when all you have is your voice is extremely disenfranchising.
I don't trust twitter staff who will be reviewing these 'abusive' tweets to be pro trans* and/or pro sex work. Let alone intersectional.— Sophia (@sophiaphotos) July 30, 2013
The biggest groups concerned about the twitter abuse button are the most marginalized and rightly so. They have the numbers stacked against them in this game and since tech is not as diverse as its user base there’s no guarantee that someone reading the report and deciding if something is actually abuse or not is an ally or can understand an oppressed individual.
The second major faulty assumption is that there is only one definition for abuse. There clearly isn’t. You can’t rely on a dictionary definition always and in the end usually someone with a poor understanding of the constitution will end up screaming about their right to free speech.
The issues are complicated and I honestly do not expect a privileged homogenous team to actually be able to come up with a solution, because privileged groups tend come up with solutions that are best for privileged groups. Better than nothing is no good when that solution will be used to actively harm those groups of people already persecuted on a daily basis.
See a typo? Submit a pull request