31–40 of 105 posts
Every few years or so I flirt with the idea of backing up to M-Disks. Then I remember how slow and low density they are. For recording and forgetting, I recently played with LTO, again. But I always end up at local HDD + remote AWS Glacier.
AWS Glacier is mostly just LTO done for you with exponentially less risk.
Verbatim, who owns M-Disc, doesn't even manufacture discs anymore. They just slap their brand on other peoples discs. The discs sold as M-Discs now are actually just MABL BD-R's. They used to have a 'stone' data layer. Now they use a titanium metal data layer. MABL discs were always designed to last for about 100 years, and are orders of magnitude cheaper. 100 years is long enough, and 1000 years was always a wild-as…
For things recorded today, nobody will care about refreshing for 75 heads, and if during 2100-2125 there are some budget constraints/wars/whatnot, the data will rot.
With 1000 years, you can just throw stuff into a bunker and all that it will take will be for someone in a period of 2800-3025 to take the time to refresh the library. 200 years should be enough to wait through a mild civ collapse or something.
Sure we’re talking about times beyond our lifetimes, but a lot if things we are sorrounded by were planned by people who knew they wouldn’t live to see the outcomes.
The drives to read them (Blu-rays) will be the problem. Pioneer stopped manufacturing them in 2024, only LG drives are available now. Coincidentally I just bought one, they’ve doubled in price now recently. Also bought some M-Discs to try. They’re half the price in Japan (Amazon) compared to Germany.
Even the largest 100GB disk seems too small nowadays
Verbatim, who owns M-Disc, doesn't even manufacture discs anymore. They just slap their brand on other peoples discs. The discs sold as M-Discs now are actually just MABL BD-R's. They used to have a 'stone' data layer. Now they use a titanium metal data layer. MABL discs were always designed to last for about 100 years, and are orders of magnitude cheaper. 100 years is long enough, and 1000 years was always a wild-as…
> 100 years is long enough, and 1000 years was always a wild-ass theoretical abstraction. The software will be an issue before the hardware anyway inevitably
Earlier quoted context omitted.
Maybe. The question is why are we needing to read these disks in a thousand years. I would guess right now that the most likely case is information we care about is archived on modern formats such as hard drives, which are constantly being copied to different drives as the old ones go out and so in a thousand years they'll still have all the data in a much more accessible format than a very obsolete disk. If, instead…
> The question is why are we needing to read these disks in a thousand years. The British Domesday Book from 1086 is still referenced by historians for all sorts of purposes: * https://en.wikipedia.org/wiki/Domesday_Book Not sure what kind of present-day data the historians of the future would be interested in. > If you want something archived for a thousand years, well, we know that clay tablets baked worked […] The…
Burrowing stuff sometimes works, however, a lot is lost. Even when we attempt to save things, it often doesn't work. Some years ago a town near me had a big unburying of a vault they had saved for 50 years. In there was a brand new car complete with gas in case gasoline was obsolete in 2000. The gas was stale and unusable. The car was completely rusted the point where it couldn't be restored. The only answer that you can be confident of storing something is if someone every few years personally looks at it and verifies it is still in good shape and if not repairs it. Archaeologists can learn a lot from garbage and trash that was buried with no attempt to save it. However, archaeologists are also very aware of the limitations of what they can get. There is a lot they know they don't know because, well, this would not have been saved or been preserved.
Earlier quoted context omitted.
> Issue with storage is you need shelf stability without ongoing electrical/mechanical work refreshing data time to time. For what? Nothing we do today will be even remotely important in 1,000 years. Any tiny fraction of a fraction that is important will have been copied thousands, millions, or billions of times into various formats. "I met a traveller from an antique land Who said: Two vast and trunkless legs of sto…
> For what? Nothing we do today will be even remotely important in 1,000 years. All sort of things done a thousand years ago still shape how we live today: * https://en.wikipedia.org/wiki/Magna_Carta * https://www.goodreads.com/en/book/show/2187985.The_Origins_o... * https://www.goodreads.com/book/show/3624887-the-medieval-ori... Who knows what the actions of a single, seemingly inconsequential, individual could trig…
Earlier quoted context omitted.
> Issue with storage is you need shelf stability without ongoing electrical/mechanical work refreshing data time to time. For what? Nothing we do today will be even remotely important in 1,000 years. Any tiny fraction of a fraction that is important will have been copied thousands, millions, or billions of times into various formats. "I met a traveller from an antique land Who said: Two vast and trunkless legs of sto…
> For what? Nothing we do today will be even remotely important in 1,000 years. All sort of things done a thousand years ago still shape how we live today: * https://en.wikipedia.org/wiki/Magna_Carta * https://www.goodreads.com/en/book/show/2187985.The_Origins_o... * https://www.goodreads.com/book/show/3624887-the-medieval-ori... Who knows what the actions of a single, seemingly inconsequential, individual could trig…
I don't disagree that a few things that happen today will be of consequence. I disagree that any single source of that data needs to last for a thousand years in isolation.
There are, I'm sure, one or two examples of thousand+ year old primary sources revealing otherwise lost information that was still relevant in some way. I suspect not many, other than as datum for academics to ponder.
Important information has always survived by copying.
Earlier quoted context omitted.
> For what? Nothing we do today will be even remotely important in 1,000 years. All sort of things done a thousand years ago still shape how we live today: * https://en.wikipedia.org/wiki/Magna_Carta * https://www.goodreads.com/en/book/show/2187985.The_Origins_o... * https://www.goodreads.com/book/show/3624887-the-medieval-ori... Who knows what the actions of a single, seemingly inconsequential, individual could trig…
Every single thing you cite has been, and will be, copied many times over and preserved that way. e.g. - We don't have the original Bible, we have eleventy-billion copies of it. I don't disagree that a few things that happen today will be of consequence. I disagree that any single source of that data needs to last for a thousand years in isolation. There are, I'm sure, one or two examples of thousand+ year old primar…
> Dating from the 3rd century BCE to the 1st century CE, the Dead Sea Scrolls include the oldest surviving manuscripts of entire books later included in the various Jewish and Christian biblical canons, including deuterocanonical manuscripts from late Second Temple Judaism, as well as extrabiblical books. As such, they cast new light on the emergence of Rabbinic Judaism and of Christianity, and their relationship with Jewish religious traditions.
One could similarly imagine a set of M-Discs from today that would be rediscovered 1000 years from now that could prove similarly useful to the people of the future thanks to data on them that has been preserved unaltered since today until then on that set of M-Discs.