3 ms·

To be fair most of the books written in Herodotus’ time are also gone! Many of them we only know about from Byzantine library catalogs or summaries, but surely

To be fair most of the books written in Herodotus’ time are also gone! Many of them we only know about from Byzantine library catalogs or summaries, but surely many more have just completely vanished from knowledge or existence. With the existence of things like the Internet Archive, it’s quite possible the public web content of today will be well preserved for centuries to come.

pjmlp 8y ago
Right now the Internet Archive and BitKeepers feel like Alexandria's library, a kind of single point of failure with a big loss to all of us if something happens. And we also need to keep running the applications that know how to deal with specific file formats.
- larkeith 8y ago
  To be fair, if we do manage to preserve the Internet Archive to a significant degree, information on html and other filetype standards should be readily available for future recreation.
- int_19h 8y ago
  But it's much easier to mirror the Internet Archive than it was to mirror Alexandria's library. In fact, it already has one official mirror (which is hosted by the modern library of Alexandria!). FWIW, with today's typical Internet speeds and storage prices, it's entirely feasible for individuals to mirror the entirety of Wikipedia and Project Gutenberg (done both myself, and a bunch of other stuff like TED talks). Throw in a mirror of the entire Debian source package repository, and there's your apps, too. How many copies are already around? Who knows, but I'd say we already have a lot more redundancy than at any point in the past. Of course, IA is to any of these like these are to what you could store locally circa 90s. I don't think we're going to be at the point where an individual can easily store 40 petabytes of data anytime soon. But for large entities - governments and corporations - this is feasible. And I suspect there are some private full IA mirrors out there already, in places like NSA's Utah data center. And then you can make distributed backups, too: http://iabak.archiveteam.org/ http://iabak.archiveteam.org/
  - pjmlp 8y ago
    Thanks for the heads up. Yeah I guess we need to trust that some of us will keep our recipe lists alive. :)

pjmlp 8y ago

Right now the Internet Archive and BitKeepers feel like Alexandria's library, a kind of single point of failure with a big loss to all of us if something happens. And we also need to keep running the applications that know how to deal with specific file formats.

larkeith 8y ago
To be fair, if we do manage to preserve the Internet Archive to a significant degree, information on html and other filetype standards should be readily available for future recreation.
int_19h 8y ago
But it's much easier to mirror the Internet Archive than it was to mirror Alexandria's library. In fact, it already has one official mirror (which is hosted by the modern library of Alexandria!). FWIW, with today's typical Internet speeds and storage prices, it's entirely feasible for individuals to mirror the entirety of Wikipedia and Project Gutenberg (done both myself, and a bunch of other stuff like TED talks). Throw in a mirror of the entire Debian source package repository, and there's your apps, too. How many copies are already around? Who knows, but I'd say we already have a lot more redundancy than at any point in the past. Of course, IA is to any of these like these are to what you could store locally circa 90s. I don't think we're going to be at the point where an individual can easily store 40 petabytes of data anytime soon. But for large entities - governments and corporations - this is feasible. And I suspect there are some private full IA mirrors out there already, in places like NSA's Utah data center. And then you can make distributed backups, too: http://iabak.archiveteam.org/ http://iabak.archiveteam.org/
- pjmlp 8y ago
  Thanks for the heads up. Yeah I guess we need to trust that some of us will keep our recipe lists alive. :)