'Five-dimensional' glass discs could store data for billions of years | Hacker News Reader

'Five-dimensional' glass discs could store data for billions of years

independent.co.uk

40 pointsaethertap34 comments

On the '5D' claim: "The information encoding inside the disc is realised in five dimensions - the size and orientation of the nanostructures, in additional to their three-dimensional positions inside the disc." Those nanostructures are generated by a laser which alters the polarization of the material.

Is that even five dimensional? Polarization is one more, but can each point in space contribute to an arbitrary (or even just large) number of different polarization? I do not see where the fifth might come from.

The billion year claim does not seem to be validated, unlike similar research which used tungsten and silicon carbide[1], i.e. two different materials. The million/billion year durability was verified by substituting long periods of time with higher temperatures, i.e. heating the storage medium, and then checking the error rate.

[1]: https://doc.utwente.nl/74827/1/Vries2010MME.pdf / https://www.utwente.nl/en/news/!/2013/10/141415/a-mega-to-gi... (not peer reviewed?)

As I read it: 1. size 2. orientation 3-5. the usual dimensions.

Reading it as "five measurements" might be clearer.

Orientation is at least 2 dimensions (see e.g. spherical coordinates).

Orientation of a vector is 2 dimensions (spherical coordinates), orientation of a 3D object is 3 dimensions (Euler angles). It all boils down to the fact that you can rotate the vector using itself as its axis and it stays the same, but you can't do that with, say, a cube.

Hence I said "at least" :)

In the image of the data being read it looks like the laser only hits the structures from one direction. I think the "orientation" parameter is actually the polarization angle for the ray, so only one dimension.

Could be true, but then those spatial dimensions should probably be 2 instead of 3 (laser coming from one direction means you lose that direction).

Yeah, it's a bit sensationalist. By that logic you could argue that writing on a piece of paper is 4-D: horizontal and vertical position, what letter/symbol is there, and color. You could add font and underlines/italics to encode even more data per symbol if you wanted.

Well yea because a dimension is infinitely divisiable. so a dimension can hold infinite dimensions within it.

like i.e numbers.

I guess the real problem with all such storage media is that, OK, you may encode your message in a certain way- but how do you know that anyone who comes across it will be able to read it?

How do you figure out that you need to look at nanosctructures inside a chunk of glass (or whatever) and that there's a message in there stored in a certain encoding?

As a way to preserve technological and scientific knowledge in particular, it's a bit pointless. You need to already possess the technology and the science to a) discover the message and b) decipher it.

I am sure that there's been a lot of research into that particular problem, but I would also challenge the basic assumption you seem to be making, that science would advance exactly as it did last time.

Take our plane, rocket and nuclear technology, would it be as advanced relative to other sciences if WW2/the cold war hadn't have happened? If a lot of money and manpower hadn't been ploughed into it relative to other sciences for 30 years?

Also this preservation technology is now available to us going forward, say a catastrophic collapse happened 200 years from now and 2000 years from now they get the technology to decode these things, they'd still leap forward 200 years in whatever they find.

>> I would also challenge the basic assumption you seem to be making, that science would advance exactly as it did last time.

I think I'm worried that science will _not_ advance exactly as it did last time. Right now we have two things: a) the technology to create the physical medium of the message and b) the technology to encode it in a way that is efficient in terms of its storage.

Who's to say that future generations will have both (a) and (b) at the time they come across the message? If they don't, then either they will not recognise it for what it is, or they will not be able to read it, or both.

>> they'd still leap forward 200 years in whatever they find.

Even if there's no collapse, I assume the future will like to know a few things about the past, but of course we can't know. Maybe the future is a boot stepping on a human face, forever and such things as history are not allowed. That would be a development entirely unrelated to technological means, but that would defeat any planning based on assumptions about future technologies. And unfortunately we can't really do much about it except, I don't know- form secret societies for the preservation of knowledge? Like in Nightfall [1]? :/

[1] https://en.wikipedia.org/wiki/Nightfall_%28Asimov_short_stor...

One way they could hint at how to read it could be thru the form factor. Imagine if the data were to be embedded in a glass skull like the crystal skulls[1] or something akin.

1. https://en.wikipedia.org/wiki/Crystal_skull

How does being in the form of a crystal skull show me how to read it?

It's a good question. Short answer: bootstrapping.

I happened to be over at the Long Now foundation on Thursday when this fellow stopped by for a chat; the Long Now wants to evaluate this storage technique for their language archive, the Rosetta Disk.

The Long Now's solution is to have information at multiple scales:

http://rosettaproject.org/disk/concept/

Note that there is human-readable text around the rim that gets rapidly smaller. That's a clue that you need to keep zooming in. A prototype container is part magnifying glass, and I think the plan is to have instructions at modest resolutions on how to build a 100x microscope, and then more instructions at that scale on how to build what you need to get to the next scale.

Presumably for this nanoscale disc you would apply the principle repeatedly. Each scale contains instructions for getting to the next scale. For the finest scales you'd need a lot of social and technological infrastructure. But there's another Long Now project to get people to think about what that requires:

http://blog.longnow.org/02014/02/06/manual-for-civilization-...

What if each observer hold its scale as a state (like a ear mark), so the thing is observer independent. Sounds like a universe to me.

>social and technological infrastructure

The question I am yet to find a good answer for, with regards to the Long Now proposal, is this: why?

Like, if you know you're going to have a situation where this immensely important information is going to have to be re-discovered, some distant time into the future, because .. after all, entropy is the natural state of things and who could expect we'll still speak English/C++ 100,000 short years from now .. so we'll teach them again, some day way in the future. But, why? Is it merely to fight the entropy, and say: no matter the light, here is the shadow we cast today, and millions of days, millions of years into the future?

Because that is some mighty high pretense.

However, perhaps you have a better answer, as a more coherent observer of the Long Now technological/social infrastructure? Is it really just to challenge time in a technological sense, or .. perhaps .. is this one area where secularism is becoming what it resists, and indeed such foundations are a new religion/cult/etc.?

Sorry, I'm not following. Why what? The Long Now Foundation itself? The Rosetta disk in particular? Some other project of theirs?

You can see what the founders say in their own words here: http://longnow.org/about/

But their basic goal as I see it is: "encourag[ing] the long view and the taking of long-term responsibility, where 'long-term' is measured at least in centuries".

The disk itself is an example of that. The name obviously hearkens back to the Rosetta Stone, which unlocked the translation of Egyptian hieroglyphics for us: https://en.wikipedia.org/wiki/Rosetta_Stone

With a copy of the Rosetta Disk, as long as some future finder understands any one language spoken today they'll be able to translate any other materials they find from our era.

Will it survive? Will it be useful? Will it even be necessary? More broadly, what does the future mean to us, and what effect do we want to have on it?

Those are all good questions. And my take is that the real purpose of the Long Now is not to answer them, but to get people asking them. So good work, your post helps fulfill the mission.

That is a great answer, thanks. :)

I remember reading about the Rosetta disk. It's a good start, the fact that it's human-readable makes it much more likely to be useful. The "bootstrapping" idea is really cool also.

There's still some assumptions- that future civilisations will have some knowledge of some of the languages we speak nowadays and so on, but they're probably safe to make, more or less.

There's always the chance that one of those things will only be found by some alien civilisation visiting the Earth in the far future, when there's no humanity left, let alone civilisation- but that's probably a very remote possibility to plan for. Although I guess Voyager 1 already has taken it into account.

One of the Rosetta disks ended up on the Rosetta probe. If things go according to the ESA's plan, the probe will finish on the comet, meaning that the disk could live a very long time indeed.

>How do you figure out that you need to look at nanosctructures inside a chunk of glass (or whatever) and that there's a message in there stored in a certain encoding?

I've asked myself this question numerous times, mostly while staring at the arcane crystalline structures available in meteorites, wondering if in fact there's some way that some parts of the universe aren't, indeed, just a message being left in the sand by a higher power.

Oh crap! Those holographic storage cartridges I've been expecting for a a dozen years have been obsoleted by a new storage tech that will never come to market either.

Hopefully ibm will soon start to market their 3D write with lasers in gel like substance that will replace hard drives in ten years as they announced circa 2002-2003 .

Maybe not, and this new media release is no different than the others breakthrough storage tech we hear on a regular basis but are never turned into a reality.

What good would a media that last billions of years when 50 years later there is not a device able to read those discs ? If a proprietary format then it will lack the documentation to be able to build a device to read the disc and interface with the computers of 2060. If an open format and standard then how to keep the specs along with the discs as storing it on the discs would defeat its purpose.

Hopefully some investors will burn some money in this tech that will never come to light for consumers and will not keep up to its overhyped claims.

At least there is proven technology today [1] that lets you archive for about a thousand years. Store all your stuff to that, (and, of course, the cloud) - and then when one of these holographic storage systems with good traction comes online, dupe your stuff to that.

[1] http://www.mdisc.com/mdisc-technology/

There's actually already a medium that can last hundreds, if not thousands of years which could contain the spec for this device. Let me see if I can drum up a link for it. Ah, here we go: https://en.m.wikipedia.org/wiki/Book

I've seen those used for all the things I could use a glass disc for.

None of these things are of any use if the user can't read, or hasn't sustained itself, biologically, for whatever its digestion period is ...

Several commenters here protest that the information might not be readable by any devices 50 years from now. But the application of the tech isn't a new media player. It's preserving the literature of our civilization across potential dark ages. If future civilizations discover these media, they'll easily discern that they contain encoded information, and crack the code. Imagine we were able to recover indestructible, encoded books from the library of Alexandria. Cryptographers would soon discover the code.

The intent isn't to illuminate the dark ages when it might be impossible to decode the texts. The intent is to preserve for a time beyond the dark ages, when knowledge and culture return.

I think it's more likely a hypothetical dark age would see them as interesting jewellery, and drill holes in them so they could be worn as necklaces.

Technology has an historical event horizon. If an item is on the wrong side of it, it's literally incomprehensible. Items like these are only useful to a culture that has already reached a similar technological level, give or take some small time delta.

That makes them a very bad way indeed to preserve information about science and technology.

It sounds nice. But I have been fooled before. During my training I found a paper on IBM joining some company to create a "HD-ROM", using gallium ions instead of a Laser, which supposedly would have stored some 150 GB on a medium the size of a CD or DVD, and that mediumd could been literally anything, rock, iron, a diamond, whatever, ... but apparently, it never came to pass.

I think it did sort of come to pass.

http://www.norsam.com/rosetta.html

https://en.wikipedia.org/wiki/HD-Rosetta

http://rosettaproject.org/disk/technology/

But it just wasn't very popular? Hard to tell.

Hi! Occasional Long Now volunteer here. I've seen the prototype disks, but my hazy recollection is that production cost kept this from being a practical technology for many uses, which of course keeps the production cost high. It turns out most people don't really care about archival quality. E.g., almost all my data is on media that I replace a couple times a decade.

How well does it withstand cosmic radiation?

Put it in a vault.

Remember inPhase holographic storage.

Reply on news.ycombinator.com