The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
The Gambia v. Myanmar & Facebook
The opinion, handed down yesterday by Magistrate Judge Zia Faruqui (D.D.C.) is too complex for me to summarize, but it's very interesting, and I thought I'd pass it along. Here's a brief excerpt from a story at Reuters (Poppy Mcpherson):
A U.S. federal judge has ordered Facebook … to release records of accounts connected to anti-Rohingya violence in Myanmar that the social media giant had shut down, rejecting its argument about protecting privacy as "rich with irony"….
Facebook had refused to release the data, saying it would violate a U.S. law barring electronic communication services from disclosing users' communications.
But the judge said the posts, which were deleted, would not be covered under the law and not sharing the content would "compound the tragedy that has befallen the Rohingya".
Here's the introduction from the opinion, which is denominated In re Application Pursuant to 28 U.S.C. § 1782 of The Republic of the Gambia:
By Facebook's own admission it was "too slow to respond to the concerns raised" about its role in the genocide of the Rohingya, an ethnic and religious minority in Myanmar. In 2018, roughly six years into the genocide, Facebook began deleting accounts and other content from its platform used by Myanmar government agents that sparked the genocide….
The Republic of The Gambia … seeks the content Facebook deleted for use in The Gambia's litigation against the Republic of the Union of Myanmar … at the International Court of Justice …. The Gambia seeks these records for "evidence of genocidal intent necessary to support a finding of responsibility for genocide" of the Rohingya. Facebook argues that The Gambia's request: (1) violates the Stored Communications Act ("SCA"), and (2) is unduly burdensome…. The Court GRANTS The Gambia's application in part and DENIES it in part.
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
Having one guy spend 10 minutes pulling this from a database, then using the remaining 9 to walk down to the Starbuck's in the cafeteria is unduly burdensome to a company that could have funded half of World War II all by itself.
Having been one of the guys who had to pull stuff like this from databases, it's not nearly that easy. In this case, the parties are asking for long-deleted content that now exists only on backup servers. Backups are primarily designed to efficiently recover the entire environment in the event of hardware (or sometimes software) failures. Backups do not retain the data segregations used by the live environment. Conducting electronic discovery against a backup requires either a restore of the then-current environment - which means either overwriting your current version of the data or cloning the entire data center necessary to support the restore - or the use of highly specialized and expensive software to parse the backup files into usable data.
When considering these costs, consider that Facebook is not actually a party to this litigation. They are being compelled to spend all this time and money solely as a third-party subpoena respondent. (Okay, not quite but the international litigation equivalent.) The point is that neither side in this suit is claiming that Facebook did anything wrong. Yet Facebook is being compelled to spend rather a lot of its own money to dig this information out their backups.
I can see the legal argument that third parties may have to cooperate with legal investigations. However, I have trouble with the argument that the people demanding the information don't have to pay for it and can so easily offload those costs onto often-innocent bystanders.
That was a really useful description of those limitations, thanks!
I appreciate your point, but at the time the information was requested the relevant backups would have been almost 2 years old. I assume other backups in the intervening 2 years were not all incremental, so their reason for keeping the old stuff wasn't so they could play back a full two years worth in order to recover. Why then would Facebook retain this information if not for selective retrieval?
I don't know for sure but I'll make two guesses. Either:
a) Facebook was over-retaining. DASD is cheap and finding the time/attention to purge old data is a lower priority that fixing whatever's broken today or
b) Facebook had the backups preserved in response to some other litigation which had not (and maybe still has not) gotten to the point where actual restoration of the data is appropriate. Large companies often suffer from rolling litigation orders. Risk-averse law firms require the company to preserve the data even as they fight to show why the company should not be required to process or disclose it. If new case B comes in before you've successfully fought off case A, the things you held for case A roll forward.