 Because this is the talk of the #internet again: you don't preserve digital media by stuffing the One True Version in an #archive. Make countless copies and scatter them to the wind. Make each a different format. You don't know which of them will still be readable next decade, so *don't try to guess*. That goes for analog media too. The Library of Alexandria contained copies. Many classic paintings only survived as copies. Copying is how life itself beats death. Embrace it. 23 comments
 @felix Today, I received in the mail a newspaper from the 1940s that is a reprint of a dime novel from the 1880s. The 1880s edition is not well preserved. Copies exist, but they're trashed. The 1940s reprint is plentiful enough that I paid $5 for mine. @morgan  @footsteps thank you. I love stuff like this. I put stuff like this on my #blog, fastestslowguy.blogspot.com. Working on some good blog posts now, they'll be interesting. @morgan @ajroach42 @felix one of my favorite books is Rain of Gold by Villasenor, so you got me interested. I just found a copy online and bought it. Will share with others after I enjoy it.  @morgan @ajroach42 @felix Seems like it would be an interesting companion read to "Sixty Years in Southern California, 1853-1913" by Harris Newmark @felix@wandering.shop @felix Print it. PDF it. Let The Internet Archive host it. No matter what, though, I guarantee that, in a century, computers of that era will still be able to open a .txt file. @colin @felix Or an ePub - the format is simple enough that it can be figured out from scratch if need be so long as you know what a ZIP file is, and those aren't going to be forgotten. But ePub includes metadata, which means the people in a century might actually be able to find the story within the haystack of old data.  No offense, but I think this take is ahistorical and borderline anti-intellectual. We actually can reason about data storage, formatting, indexing, etc., and we're obliged to reason about this seriously. We have precedent for this as well. Archival is a science with a history. I do agree with the sentiment about the fundamental importance of copying.
@felix Though I am finding you can't publish on a website any more. It's all Github or GTFO. If your program isn't up there, people just decide it has to be shady or outright malware. One important reddit forum even has it in the policy that no software may be linked to unless it's published on Github.  @felix As someone in the archives profession (though not digital), I would also request: Document what changes you you may have made in terms of format or whatnot in the metadata as best you can, so if something does go wrong with a copy, people can potentially figure out *what* went wrong. Actually, really my advice is "make documentation of what decisions you made." Because goodness knows in my worklife there are times I *wish* certain decisions by my predecessors had been documented.  @felix This is exactly the point of the fantastic (FANTASTIC!) book "How the Irish Saved Civilization" (which I just finished). @msbellows @felix One of the first books I read when I left Mormonism two decades ago. Unfortunately I loaned out my underlined copy. I will never see it again. (But someone will.)  @felix Yes. The One Giant Archive is for convenience of lookup, not for preservation. |
Gregory's Server
@felix
This is extremely good advice. Call it the plant based model of archiving.