3 ms·
And how much of that data is in outdated formats that it's now almost impossible to read? If I could still find my middle school papers written using WordPerfec
by twobyfour 9y ago
And how much of that data is in outdated formats that it's now almost impossible to read? If I could still find my middle school papers written using WordPerfect 5.1 or MS Write, and if the floppies were still legible, would I be able to open them in software that runs on my current operating system without emulation? How long would it take to find something to read them with?
And that's with very common software from only a quarter of a century ago! Heck, I've got files written in niche Mac OS X software from less than a decade ago that probably can no longer be read on a computer manufactured this year.
- ekianjo 9y agoThats precisely why we should only use Free and Open document formats. They have a much higher likelihood to make it in the long term.
- twobyfour 9y agoIn the long term of tens of years, MAYBE that helps. My guess is that ODF, for instance, will be harder to find readers for 50 years from now than DOCX will. In the long terms of tens of thousands of years, the chances anyone will be able to read any of them in any way are vanishingly low. Even if the media are intact (which is an incredibly low probability) and even if hardware were developed to read the media, all that's going to come off the media are ones and zeroes that represent a document format wrapped around an encoding that represents a dead language. OK, maybe plain text ASCII could be decoded by cryptographers highly familiar with Ancient English. You might look up the Long Now project and how they're thinking about communicating 10,000 years into the future.
- ekianjo 9y ago> In the long terms of tens of thousands of years, the chances anyone will be able to read any of them in any way are vanishingly low. Even if the media are intact (which is an incredibly low probability) and even if hardware were developed to read the media, all that's going to come off the media are ones and zeroes that represent a document format wrapped around an encoding that represents a dead language. Sorry, but if there is any kind of document specification left, it should be trivial for whoever is left then (as long as they can grasp our 21th century plain language) to recreate a reader for an old format. With proprietary documents, that is just never going to be possible.