Digital Permanence

Manton Reececited wrote some sensible words about the permanence of material on the Internet, and Dave Winercited followed suit shortly after (and then againcited). It’s an important topic.

We need places — secure, digital, permanent places — to store things. It’s not a particularly difficult problem, at least to attempt to solve. (Of course, we won’t really know if we’ve succeded for another few hundred years.) So we should try!

A couple of ideas that I’m using as a baseline these days:

  1. Store things in open formats, so we can continue to read them.
  2. Store things in a small number of large (and non-esoteric!) repositories (i.e. filesystems, or drives, or websites, or whatever), so they’re easy to migrate to other places.

The latter is, I think, important: it means that the data can be easily handed over to someone else.

Of course we can do it

I read a few posts this morning about the wonders of open software and how it can help ‘the 99%’. Nothing new there, in the techno-optimism, nor with the “yes, but” reactions to it. It’s heartening to read it, though; there’s far too many people harping on about the delights of the iPhone.

One thing that does strike me is that we really do already have the tools for this open paradigm: we don’t need some new crew of idealists to come along and build a Faceboo’ replacement, or invent some new way of storing files (or remove the necessity for doing so). We just need to stick to the open standards and apply a tiny amount of conservatism when it comes to choosing the next groovy technology that we’re to trust. Not very difficult, and yet people run to cast their lot in with corporations who give barely a nod to these ideas.

Perhaps Bruce Love is correct, and a practical distributed and decentralized net can be built trivially using a mix of explicit peering with rsync and rss for open peering. Seems simple to me; just needs to have a GUI that makes it look like an iPhone, I guess.