In a blog post this week, Twitter announced a number of safety changes to continue the site’s campaign against online harassment.
Most notable (and most welcome among users) is the option to filter notifications from users with a default profile photo or users who haven’t yet verified their e-mail address and phone number. It may seem like small change, but given how easy it is to set up a new Twitter profile, many people create profiles solely for the task of harassing someone else. Forcing those users to select a profile photo and verify their account will hopefully make it more difficult for trolls to reach their target. If you run your own blog, you know how troublesome it can be to deal with multiple account spamming over and over.
This option is going to be available to ever Twitter user. A notification with instructions on how to use the feature would be included in the next update.
Also a part of Twitter’s safety changes is an expansion of the site’s “mute by keyword” feature. Users can now mute keywords, phrases, and even entire conversations and set a time limit for how long the mute last. This feature could potentially be popular with people who want to avoid spoilers of their favorite television show, or people who just want to mute an argument for a while to cool off.
Twitter’s vice president, Ed Ho, noted in the company’s blog post that these changes were two of the most often requested from site users.
Perhaps the biggest announcement though was that the social networking site is starting to use algorithms to determine potential abuse. Soon, Twitter will be able to “identify accounts as they’re engaging in abusive behavior, even if this behavior hasn’t been reported to us.” If Twitter’s algorithms detect abusive behavior, that user’s account will automatically be limited temporarily. This is likely in response to a big request from people on Twitter: for the service to become more proactive about abuse instead of waiting on reports from other users.
Of course, the system won’t be perfect, and users who aren’t being abusive will likely get undeserved time outs. Twitter’s intention, Ho said, was to “only act on accounts when we’re confident, based on our algorithms. But he added that the new system would inevitably make mistakes.