Mehdi Hasan Dismantles the Entire Foundation of the Twitter Files
techdirt.com
techdirt.com
> But, more importantly, regarding the small number of tweets that EIP did report to Twitter… it never suggested what Twitter should do about them
Someone should read Chomsky’s “Manufacturing Consent.” You don’t need express collusion for the result to resemble express collusion.
The broader issue here is trust, or lack thereof. Do you agree with Hasan here because you trust the organizations and people involved, or because the structure of the system provides assurance that you don’t need to trust them? For example, if the actors involved were faith based organizations rather than non-profits, would you feel the same way?
Zoom all the way out here: Twitter is partnering with organizations (universities, non-profits) where 90%+ of members support one party to help monitor “election information.” You cannot defuse that problem by saying that we should just trust those people to be impartial. It’s like computer security: look at the system assuming everyone is a malicious actor. How does this Twitter’s system look from that vantage point?
As to the error—I don’t think anything is “trivial” here. In the EIP study, there were 181 tickets, with almost 6 million tweets “directly referenced in a ticket.”
If you flag a tweet and it’s removed, what happens to the associated tweets? I believe retweets and replies are removed too, right? If so, I don’t see how Hassan’s characterization is accurate either. EIP is shooting at the roots of information trees that branch out. They don’t need to shoot at all the leaves.
And as noted above—Hassan’s rebuttal seems equally if not more mistaken. Taibbi is being sloppy with the definition of “flag.” But the end result—in terms of what happens if a flagged tweet is removed—is closer to Taibbi’s characterization.