In the aftermath of the chaotic coup-that-wasn’t incited by a Brady Bunch of QAnon conspiracists, Proud Boys, and everyday American Trumpists on January 6, 2021, their newfound home, alternative (read: white supremacist-friendly) social platform Parler is in trouble. Amazon has dropped the controversial platform from its AWS hosting service starting Sunday night, January 10. Apple and Google also removed Parler from the App Store and the Play Store respectively, and a host of others have terminated their business with Parler. While Parler’s CEO and founder, John Matze, is currently crying censorship and eschewing responsibility for the Capitol riot, it’s worth revisiting what he told CNBC when inquired about the presence of bad actors on the platform, stating that they represent “a minute percentage” of the app’s user base and that they won’t be a “long-term problem.”
RELEASE: Every Parler post made during the 06/01/2021 US Capitol riots. https://t.co/YUl8CtFPw8 (batches of 100k URLs, for archival purposes)
— crash override (@donk_enby) January 9, 2021
In the wake of these events, an Austrian hacker and researcher, who goes by @donk_enby on Twitter, has managed to scrape over 80TB of unprocessed data from Parler’s servers, in the hours before the take down, including 1,098,552 video URLs. “These are the original, unprocessed, raw files as uploaded to Parler with all associated metadata,” she tweeted on Sunday, including the GPS coordinates of users at the time of filming these videos. She describes the event as “a bunch of people running into a burning building trying to grab as many things as we can”, which at the moment feels very apt, adding that “[t]hings will be available in a more accessible form later.”
— crash override (@donk_enby) January 10, 2021
Parler, with its user base of Trumpists, extremists, and QAnon truthers and pro-Trump financiers (like Republican donor Rebekah Mercer), is said to be among the platforms in the “alternative” digital ecosystem favored by the far-right mob that stormed Capitol Hill, in a farcical effort to overturn the results of the 2020 election, which led to five reported deaths. Through an exploit of Parler’s API, @donk_enby began archiving all posts from the day of the riot.
@donk_enby’s effort ultimately led to the capture of almost the entirety of the platform’s content from “Wednesday, most of Thursday and all of 3 days prior by the end of it”, successfully aggravating digital defenders of “free speech”, whose definition of the term I suppose doesn’t include archiving for posterity. Such a leak would have interesting implications for several online subcultures prone to doxxing and harassing their political opponents, but, given the incriminating posts on the platform, this is hardly an attack on free speech.
For context, the only way one could previously verify their account on Parler was to give their social security number to the platform. Parler had apparently been using a free trial of identity and access management company Okta’s software, until the latter was publicly informed of this and promptly terminated Parler’s access, thus disabling the email and phone verification needed to create an account and allowing anyone to directly create multiple accounts. It seems that Parler, rather than protecting their users’ expression, screwed them over quite spectacularly.
This isn’t the first time a hacktivist has dug into Parler. In November, Kirtner, credited with founding Anonymous, claimed he acquired over six gigabytes of Parler user data from an unsecured AWS server. The following month, Kirtner was suspended from Twitter for posting, “I’m killing Parler and its fucking glorious”; his account remains suspended. Trump’s deplatforming caused much hoopla regarding platforms and censorship, with critics questioning the sinister implications of platforms intervening in such a way. Similarly, Parler, which created a reputation for itself as a sort of free speech utopia, but its ecosystem is hardly organic; its business model, per Matze, is premised on influencers attracting ad revenue.
Anyone with experience scraping online data is no stranger to its precarity and contingency; @donk_enby’s work is invaluable for researchers working on mis- and disinformation. Her important efforts are documented here.