An Archival Conundrum

[Christoph C. Feldhaim]

Maybe, after all, film wasn't that bad ...

[NikoJorj]

Maybe, after all, film wasn't that bad ...

No possibility to duplicate an original without quality loss : "If there is no solution, it is because there is no problem." 

[Christoph C. Feldhaim]

True - once you cannot read your data anymore you can sleep quiet again and think for a new thing to try out, like collecting stamps or learning Didgeridoo ...

[bradleygibson]

Interesting! Do you got an order of magnitude of cost?

Unfortunately not--I've never embarked on a 25-year mission for myself or for clients, so I've not done any pricing.

The good news is the technology is ubiquitous--all the commercial optical media (music CD's and movie DVD's) is done this way--this implies some economy of scale.

The bad news is the commercial optical media are done at rather high production runs...    I'm sure there are duplication houses that offer pressed media, presumably in very small production runs, but as I say, I've never priced it.

[tived]

True - once you cannot read your data anymore you can sleep quiet again and think for a new thing to try out, like collecting stamps or learning Didgeridoo ...

  LOL, that just made my day 
i just hope i never experience it

henrik

[sty]

I would just get a cheap off-the-shelf NAS that can do zfs, put the disks in raidz1. Copy stuff there, shut down and then move the box to your summer cottage or something.

Run the scrub every 6 months or so to detect if cosmic rays/bitshift/aliens is affecting the data in any way (zfs doesn't trust hardware, only checksums so it'll recover your bits).

Modern drives can and will take easily that 50 startups your're going to need for that 25 years archival time. If a drive fails to start, well that's why you bought the drives from different batches, had one spare on the side and were running raidz1.

[jduncan]

My two cents:
1. An other vote for magnetic tape if you can control huminity.  The key will be to select old technology that have proved to last.
2. Flash technology sound good but is unproved and could be subject to all sorts of quantum tunneling effects. I will not recommend this one. But if some one needs to do it , peak less dense [old] technology.
3. This is cheeting but what about the web? it could be a excellent secondary /
4. Print all the stuff in metal ?they said it will last, but i am not sure about it.
5. Just for fun: place an old camera in front of the display and snap a  glass negative of each one