[huffman]

Maybe the medium will last longer, but isn't "If you have a computer with the right software and an optical microscope hooked up to it" a big part of the data decay problem on floppy disks etc.? Even if the data is there, it's useless without the right software/hardware to read it.

[rsaarelm]

If it's actual visual dots on the glass, it sounds like you'd just need to figure out how to get a good enough photograph of the thing to see the dots in and spend an afternoon whipping up an image processing script in Python to convert the dots to binary code. This sounds a lot simpler than whatever would be involved in setting up the apparatus to read bits of volatile magnetic media where you'd have to interact more closely with the media, be more mindful of not accidentally destroying the stored data and rely on much less intuitive physics than looking at a thing through a microscope for the recovery process.

[huffman]

Those are good point. I guess I was more curious about how the data is stored — say someone 100 years from now picks one of these up and wants to read the data. Will the format in which it's stored prevent them from doing so?

[rsaarelm]

A very simple format would be regular fixed dot patterns (⣿) corresponding to bytes, spaced a apart a bit. Then it's the user's responsibility to be sensible and encode stuff without too clever compression. This wouldn't guard against trying to read the image sideways or upside down though.

If you get a disc with ASCII text and can read the bits a 100 years from now, even if ASCII coding is lost to history, as long as you can figure out the thing is supposed to contain writing that uses 8 bits per letter, and still know English, a trivial pattern analysis gets you the code table you can use to read the text.

The plaintext can then be used to describe the formats for whatever other data the disk may contain.