*You* can "pretty much" guarantee it? Ok good. Great. If I have issues in the next half-century finding a Blu-Ray compatible drive that will work with a then-modern computer, I should come find *you* and you'll take care of it? I should only have 3-4 exabytes of data for you to recover by then.
Let me give you a few exercises so you can demonstrate the quality of your 'guarantee'.
- Find me a cassette drive so that I can pull some software for an Apple II from it.
- Find me a working 8" floppy drive that I can plug into my system.
- Find me a punch-card reader that will interface with a Windows 8 laptop.
- Now find me some software that will run on my computer (under either OS X 10.8 or Windows 7, your choice) that will allow me to read and recover the data I have stored in a Microsoft Works word processor file from 1993.
These are the issues with what most people think of as 'long-term' archives. Changes to undocumented formats, changes to hardware so profound that you've got to chain 3 or more adapters in order to allow you to plug it in, and that doesn't even guarantee it'll actually do the job at that point. And I'm not even talking about spans of time as long as the life-spans claimed by 'archival quality' optical media today. (Most notably because we weren't using electronic, or even electromechanical, computers yet that long ago.)
The fact of the matter is you *can't* guarantee it. You can't even guarantee that the 'archival quality' disc will last more than a decade. In practice, many 'archival quality' CD-Rs, with manufacturer estimated life spans of 50+ years, failed inside of 10 years despite being stored according to the manufacturer's recommendations. Burnable DVDs are a bit better, but still have the same root issue. Why? Because those life spans are estimated (based on standard tests, yes, but those test give numbers that cannot be verified, because even *pressed* CDs haven't been around long verify a life-span that long). Even 'archival quality' discs commonly have an *unburned* shelf-life of 5-10 years, even the *best* 'archival quality' discs have dyes which will degrade in less than a month of exposure to sunlight.
No, you can't guarantee that *any* individual instance of media will last as long as its manufacturer claims, but no optical media labeled 'archival quality' by it's manufacturer has lived up to those claims in practice yet. (As you said, CD-Rs, one of the earliest wide-spread writable optical formats, are less than 30 years old, and the manufacturers claim terms longer than that for their life spans, though quite a large number of those discs have since failed.) Pressed discs, I'd give decent odds on that front, but pressed discs aren't used or archives. They're used for mass reproduction of the same data on thousands or millions of blanks.
Even with proper storage, the
National Archives site recommends that you test your optical archive media "at least every two years to assure your records are still readable", presumably so you can catch the 'bit-rot' in the dyes before it exceeds the capabilities of the format's error correction measures. And this is in their FAQ on how to deal with storing *Temporary Records*.
Professional archival groups are busy looking for *actual* long-term storage solutions, because none of what we have now actually fit the bill.