Twitter just launched, and killed, a new abuse fix in 2 hours

Twitter is trying to listen to users when it comes to abuse, apparently. And that means rolling out a fix and killing it in short order.

On Tuesday morning, the company announced it would no longer notify users when they were added to lists. Lists are often deployed as a tool of harassment -- adding people to a "top idiots of 2017" list, for example. But they're also used to organise communities or to follow certain topics.

Two hours later, after waves of mostly negative feedback, the company tweeted: "This was a misstep. We're rolling back the change and we'll keep listening."

"Reconsidered and reversing," tweeted Ed Ho, the company's Vice President of Engineering.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

The notification change was part of an ongoing effort to address the company's "broken windows" when it comes to abuse and harassment, a source familiar with the situation said.

The Twitter list move was a problematic "fix," as many online critics pointed out. Most importantly, those added to abusive lists would not know and therefore would not be able to take action, including blocking associated accounts.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

Twitter does want users to get list notifications, but not for lists that are abusive. It remains difficult to define the latter, however, so for the moment it chose to block such notifications all together. Until it didn't.

User customisation seems the obvious answer. It's preferable that users be able to opt in or out of being added to lists at all, as well as being able to report those using it as a tool of abuse.

But others just want Twitter to deal with its Nazi problem.

This Tweet is currently unavailable. It might be loading or has been removed.

While Twitter has updated its abuse and harassment policies, the complexity of abuse on the platform can only be dealt with through product fixes. "Safety by design," in other words.

In recent weeks, it has allowed users to report accounts even when blocked and introduced additional steps to prevent those who have been suspended from creating new accounts.

Using machine learning and detection methods that previously countered extreme content relating to terrorism and child abuse, the company is looking at a combination of factors that might indicate abusive accounts with a high degree of certainty: IP address that have previously had suspended accounts, similar handles, similar email addresses, and accounts that start @ replying the same people and using the same hashtags.

There's more in the works at Twitter, but the issue remains: The company gives everyone a public platform no matter their opinion (almost), so most technical tweaks are going to look ineffective and incremental compared to the size of the problem.

"We are aware that we are far behind the industry standard," the inside source said. "We have always erred on the side of freedom of speech and the unintended consequence is abuse."

And that's the almost impossible problem Twitter has set itself: How can you fix abuse without giving away your entire business model and mission?