4 ms·
What Tracy Chou learned about online harassment while building app to solve it
- rwmj 6y agoIt's baffling to me that Reddit doesn't have a way to block users unless they actually message you. It's possible - there's even a Javascript hack to add block buttons[1] - so I assume there's some weird reason behind this (is it only available in a premium version)? Anyway I agree with the article that "personal blocking/filter tools" are shockingly poor or even lacking completely on every social media platform I've used. It badly needs innovation. [1] https://www.reddit.com/r/redditdev/comments/3nvrdi/deleting_post_replies_from_inbox_i_need_help_and/cvrttje/ https://www.reddit.com/r/redditdev/comments/3nvrdi/deleting_...
- livueta 6y agoIt still blows my mind that the most powerful "personal blocking/filter tool" I've ever encountered was some dude's greasemonkey script that was popular on 4chan in the late 00s. It had text filtering, obviously, which could be applied to name/tripcode/subject/filename fields as well as the body text, but would also take regexes. This is much more useful in practice, assuming you can write a regex. You could give it a list of image hashes to automatically filter. Once smartasses started perturbing images to change the hash, someone made a fork based on perceptual hashing. Recursive filtering (I don't want to see this, nor any conversation it spawned) works better with imageboard-style DAGs of posts because the mechanism is trivial, but is a powerful concept if it could be implemented more generally. I'm probably forgetting a lot more examples but the point is that the ideas are out there and that powerful tools of that sort are definitely possible - it's just that, for various reasons, the big social media players today don't see value in building something like that for their users. It doesn't help that modern social media's general antipathy to adversarial interop prevents a lot of the easier third-party implementations of tools like this.
- aaron695 6y agoThe Reddit AMA - https://www.reddit.com/r/IAmA/comments/icqpsm/i_made_silicon_valley_publish_its_diversity_data/ https://www.reddit.com/r/IAmA/comments/icqpsm/i_made_silicon... Blocking/individually managing other users is the future. It will however create bubbles. It's hard to know how bad (or good) they will be without implementing something. I'm not sure this is the company to be doing it, it depends if they can separate the tech from the rhetoric, but the Valley has ignored the issue so good on them.
- khartig 6y agoI perceive the effort to build this app as a positive and productive step in the correct direction to filter huge amounts of destructive and toxic rhetoric thrown at individuals. Ms Tracy Chou's personal experience with this, her talent, and intellect are a good fit to build tools like this. One of the downsides of personal filtering is the potential to block out disagreeable relevant truths and facts. Making arguments one-sided and exacerbating the problem of divisive thought in the world. However, this needed tool is a great and important contribution to the online world.
- jfengel 6y agoI've found that if an argument is worth making, somebody will find a way to make it politely. For people who genuinely wish to learn things, relevant facts will make their way to them. And for people who genuinely don't want to learn things, they have no trouble rejecting true things regardless of how they are written. People in filter bubbles always have been exposed to the truth at least once; they simply reject it. So I don't find any problems with blocking rude people. The author here wrote the Block button on Quora, and I use it vigorously. I would be unable to use the site otherwise, because it attracts a lot of very hostile people. It also has many smart people with a wide variety of perspectives, and they'll inform me when I'm missing something. All I lose out on is hostility.