How exactly would you check if a webpage in the Wayback Machine precisely matches a thing that used to be? You would also have to trust whoever checks them.
An interesting long play hack would be to compromise a site but only feed false material to the archive’s crawler. Then you could actually pull off what Joy Ann Reid pretended happened[1] when people found some old posts she wanted to disavow.
I assume this is in reference to Elon Musk's recent claims that somebody's "uncle owns the Wayback machine & you’re saying he deleted information for nepotism"..
Blockchain with Proof of Authority and distribution of said authority across geographic and (to a degree) professional boundaries might help. If the institutions with authority are, for example, every legal depository, W3C, the IA, a UN agency, and selected non-Western institutions, performing a 51% attack would be very difficult. It'd be more centralized than people think of with blockchain, but it would be decentralized among institutions and geography which means that if there was a war in Canada/the US in 50-100 years, we wouldn't have to worry about losing our digital heritage. It also means that if the US or Canada do turn into authoritarian states in the next ~100-200 years that they can't start scrubbing the archive.
I respect the IA greatly, but it's currently a single point of failure and vulnerable to geopolitical disruptions on a long enough scale. They're also playing a dangerous game in mixing their archival work and political activism. (That I agree with said activism doesn't keep it from being dangerous to the continued existence of the Archive).
TinyURL.com was accused of something similar. They offered a "link shortening" service. But if you linked to a page on Amazon, they'd wait a few years and then swap in their own referrer code in place of yours.
When confronted, they claimed this was the nefarious work of a third-party to whom they'd outsourced their link-redirecting...
15 comments
[ 3.1 ms ] story [ 42.5 ms ] threadI've seen downloads on regular IA with hashes posted.
To what purpose? And why would their integrity be different from other archival services, like libraries and museums?
[1]https://blog.archive.org/2018/04/24/addressing-recent-claims...
Claims that have been very refuted.
https://twitter.com/brewster_kahle/status/165928339375300608...
I respect the IA greatly, but it's currently a single point of failure and vulnerable to geopolitical disruptions on a long enough scale. They're also playing a dangerous game in mixing their archival work and political activism. (That I agree with said activism doesn't keep it from being dangerous to the continued existence of the Archive).
When confronted, they claimed this was the nefarious work of a third-party to whom they'd outsourced their link-redirecting...