Twitter’s latest crackdown on abuse is a damp squib

There is something fundamentally flawed in Twitter’s approach to the whole issue of Trolls and online abuse.


Twitter announces crackdown on abuse with new filter and tighter rules, reported The Guardian on 21 April 2015. For a company whose CEO had publicly repented its shortcomings in handling the evil trolls which cluttered its services with vitriol, it could have been a moment of reckoning. It should have been the time to show the world that they really did mean what they said. Alas! That was not to be.

Twitter had always been climbing the wrong tree even after identifying the ills that plague its platform--Trolls and trolling. It had after all, since its inception stuck to its argument that it will not censor content as it believes in free speech.  A position it has stubbornly tried to hold on to, to this date.

All that Twitter did back in 2015 was to introduce a few filters which would automatically prevent the users from seeing threatening messages. This was later followed up with options to mute accounts too. Nothing of help for the many, by then mentally drained, victims of abuse.

Which is what makes this week’s latest announcement all the more interesting. Ed Ho (@mrdonut) the VP of Engineering at Twitter, announced on February 7th 2017, the newest of announcements about Twitter’s Safety concerns and measures undertaken to make it safer.

So what has changed in the two years since Twitter’s chief executive acknowledged that the company “sucks at dealing with abuse and trolls on the platform, and we’ve sucked at it for years”?

The latest announcement included three changes: stopping the creation of new abusive accounts, bringing forward safer search results, and collapsing potentially abusive or low-quality Tweets.


1. Stopping the creation of new abusive accounts:

The vagueness as to how this was going to be achieved, would make even the great “Humphrey” of “Yes Minister” proud. Apparently Twitter is “taking steps to identify people who have been permanently suspended and stop them from creating new accounts. This focuses more effectively on some of the most prevalent and damaging forms of behavior, particularly accounts that are created only to abuse and harass others.”

How is such identification going to stop creation of new accounts? The new accounts as my humble mind tells me are yet to be created. Is this the same old story of the ip address based blocking? If so what is Twitter going to do with ip address conflicts, proxies and the like? Or is it going to be based on the use of phone numbers which these suspended users are presently required to punch in for revoking suspension? There seems to be no clarity. Hopefully something specific will be unveiled about this vague sounding measure in the coming days.

The rest of the updates are déjà vu. It seems Twitter can’t do much about the existing trolls even now other than hide them. Considering how difficult it is to get Twitter to suspend/delete accounts, this is disappointing.


2. Introducing safer search results:

Twitter, it was announced, was also working on ‘safe search’ which removes Tweets that contain potentially sensitive content and Tweets from blocked and muted accounts from search results. While this type of content will be discoverable if you want to find it, it won’t clutter search results any longer.


"Safe search is certainly not intended to stop abuse and harassment to begin with."


This is not different from Twitter’s earlier approach of putting a blanket over the potential victim to soften the punch while arming the abuser to the hilt. Safe search is certainly not intended to stop abuse and harassment to begin with. It only aims to make it look like it has disappeared once a user enables the safe search after logging in. Safe search only eliminates search results from accounts one has blocked or muted or which contains sensitive content.



3. Collapsing potentially abusive or low-quality Tweets:

The last safety feature from Twitter is the identification and collapsing of “potentially abusive and low-quality replies so the most relevant conversations are brought forward. These Tweet replies will still be accessible to those who seek them out.”

This last measure from Twitter follows the same old philosophy that the victims of abuse on Twitter are actually weaklings who do not have the heart to face criticism and public flak. It is to soothe those babies that Twitter is rolling out these features. As pointed out by this user.



What to make out of it?

Irrespective of the year of update or the flowery concern for victims of harassment and abuse, Twitter’s policy has never been pointed in the right direction. While it is the trolls who abuse and harass, Twitter has always been concerned about how to stop people from seeing things which would make them report. It expects its users to feel safe by adopting its suggested measure of hiding their heads in the sand while Twitter should have been preventing the abusers from going scot free.


"The new features doesn’t have anything against fake accounts such as these either, which is in direct violation of Twitter policies on many counts."


The new features doesn’t have anything against fake accounts such as these either, which is in direct violation of Twitter policies on many counts.




Egg and sockpuppet accounts is another problem with Twitter. A sockpuppet is an online identity used for purposes of deception. The term now includes other misleading uses of online identities, such as those created to praise, defend or support a person or organization, to manipulate public opinion, or to circumvent a suspension or ban from a website. “The egg”, is the default avatar that Twitter gives to every new account.  Many accounts, automated or not, never care to put up a picture.

There have been many suggestions such as automuting of unknown accounts or account deanonymization to counter egg and sockpuppet accounts, yet Twitter remains oblivious. Numerous egg and sockpuppet accounts are allowed to proliferate, propagate and spew hatred for whatever reasons, inspite of the  repeated cry for deletion of such accounts.



Twitter’s response to all this has been “We will let you filter them out” or as this user put it aptly;




There is something fundamentally flawed in Twitter’s approach to the whole issue of Trolls and online abuse. Even when someone reports an account which abuses repeatedly, unless its someone famous, no action is ever taken.

Going by what the latest update reveals, it could be a long time before Twitter takes effective penalising action to regulate the behaviour of its rude users. It needs to start by taking the reporting by users far more seriously and responding to them even when the user is not a celebrity who threatens to quit Twitter.  



The Hoot is the only not-for-profit initiative in India which does independent media monitoring. Your support is vital for this website. Click here to make a contribution.
Subscribe To The Newsletter
The back story of the huge apology notice published by the Hindustan Times on September 18 (see this Hoot brief) is to be found in the record of sittings of the Privileges Committee of the Lok Sabha. The apology was published three days after the last sitting to which the editor of HT was summoned. The notice given by  Andhra Pradesh MP Jithender Reddy was taken up five times by the Committee  between July end and September 15. This too has fed into the wide ranging speculation over the reason for the resignation of the current editor of the paper, Aparisim Ghosh.                       

Did it really take the Hindustan Times almost six months to figure out that it had got the figures on the attendance  in Parliament of certain MPs, wrong? Or is there more to why it carried a front page apology covering half the page on September 18? It said, "In the edition of March 24, 2017, we had, because of a technical glitch, erroneously reported the attendance in Parliament of certain MPs. Below are the accurate figures. Hindustan Times offers an unconditional apology, and deeply regrets any offence or inconvenience caused." Of the seven MPs whom it said had 100 per cent attendance  not one had it, the paper listed six other names for this statistic. And the list of those whom it said had the worst attendance in Parliament is headed by Abhijeet Mukherjee, the former President's son, who in fact has a figure of 97 per cent attendance.                                    

View More

The Washington Post  is rolling out Talk  a new commenting system that will allow the paper to better engage with readers who comment on its stories and help promote civil conversations. The software was developed by the Coral Project, a collaboration between The Post, the NYT and Mozilla, funded by a grant from the  Knight Foundation. The Post will integrate Talk with ModBot, its AI-powered comment moderation technology.                                                                         

Propublica has built a  Facebook bot which is a tiny computer program that automatically converses with you over Facebook Messenger to determine you experiences with reporting hate speech on Facebook. Its says its objective is to learn more about Facebook’s secret censorship rules and what the social media determines is hate speech. (Nieman Lab)                                       
View More