The publishing of and exposure to offensive or sensitive tweets is a problem that has plagued Twitter since its inception. If you use the official Twitter Search or any other Twitter based search tool including Social Bearing, there’s a chance you will see a tweet with offensive text, image or video.
In 2011, Twitter added a ‘possibly sensitive’ flag to each tweet. This allows tweets to be reported through Twitter by users if they contain sensitive content. The ‘possibly sensitive’ flag is a true/false value that is returned through the Twitter API and Social Bearing has implemented a ‘hide possibly sensitive tweets’ option to all streams and timelines by default. To see these tweets, you will need to uncheck this option.
This method of reporting tweets by Twitter however only goes part way into addressing these kinds of issues. I have seen examples of offensive tweets that are clearly not flagged as sensitive when they should be. The problem is particularly apparent when searching for sensitive keywords (as you might expect) or with the very latest tweets; where Twitter users are less likely to have seen and reported the tweet as being offensive.
Twitter are constantly trying to address the issue of online abuse and have made several improvements in recent months including making the reporting process easier, improvements to blocking and new filters and tighter rules.
Other than implementing the sensitive option, Social Bearing does not filter tweets any further based on abusive language or any other factors. While the problem of sensitive Tweets is something that should largely be addressed by Twitter, we are open to any ideas or feedback regarding how this site can tackle the issue of offensive content more directly.