Hacker News Reader

Comment by crazygringo | Hacker News Reader

> The only sure-fire way to guarantee that important information can be recovered in the event of a catastrophic event, is to back it up with specialized archival media that are designed to survive and be recoverable, even without access to computers and electronics.

I mean, I guess you don't need a computer to read this, but you sure do need a good microscope.

I guess I'm not really entirely sure of the point. If you're really trying to preserve basic knowledge to restart civilization, I'd think you'd want libraries of full-size readable books spread across the world. Not something that requires you to first build a microscope.

While if you're trying to store as much historical and technological data as possible, it seems like you'd want to encode it digitally with error correction codes, not as optically readable text. The encoding/decoding instructions can easily fit on a single cover page, optically encoded.

View on HN

You're focusing on how hard it will be to read but the central thrust is whether it is recoverable at all.

Can't read what no longer exists, by any means.

Paper disintegrates without maintenance. Tape disintegrates. Hard drives lose magnetization. SSDs lose their charge states. Stone erodes. Several of these won't survive fire, water, magnetic fields, or radiation.

Also, I don't see why you couldn't store digital data on this, for example as pits and lands, or base 64 text, or some other scheme, with error correction of your choice. (But if it's as durable as they say, it's not as necessary.) But, that requires additional effort to decode.

I enjoy research into highly stable means of storing data. I think Microsoft Research was developing a technique of storing data in a 3D arrangement in glass. Very fascinating.

> Paper disintegrates without maintenance. Tape disintegrates. Hard drives lose magnetization. SSDs lose their charge states. Stone erodes. Several of these won't survive fire, water, magnetic fields, or radiation.

Computers, peripherals, microscopes, and any other devices needed to access some incredibly durable archival format will also degrade. The information needed to produce new ones will need to be archived, and that material will degrade and need archiving.

Do you know how much information is necessary to describe all of the details relevant to producing an extremely good microscope from scratch? It's not just the parts and materials, it's the entirety of the supply chain: the mining, refining, electronics, power production, and everything else. All of that information needs to be accessibly preserved.

So it begs the question, is it really easier to make the prerequisite for the information's use be a working precision instrument or is it easier to just actively maintain lots of copies of less durable materials like books?

If the sky falls tomorrow, how many years is some archive really useful for in any format? Most of a sturdy, well-designed library of books will probably survive for a decade. A good percentage will probably survive a century. On the other hand, a high precision microscope probably won't "just work" after a decade or two of sitting around disused in unideal conditions. A working power supply would probably last even less time.

We must accept that there's some ongoing upkeep of an archive regardless of how durable its contents are. And while a book isn't as durable as some quartz or other fancy material, you can pretty easily make books to replace old ones even in pretty miserable conditions (we did it for centuries, albeit at great cost). There's a remarkably high cost to making sure there's a working microscope around, and keeping the information and resources necessary to make new ones and repair existing ones available.

We've been able to manage to make very good microscopes for quite a while. It won't take much more than industrial revolution level tech for this to work

If you have industrial revolution tech, you can already maintain and mass produce libraries of books. Probably quite easily.

What about all the information that was lost in the meantime, because you weren't maintaining it then?

The counterpoint is the same: what happens to all the nanofiche disks next to a pile of crushed microscopes that were destroyed when the roof fell in? In one case, the information which mattered to the people at the time persisted. In the other, no information persisted.

The whole point is that the operation persists for longer than a human lifespan. If we have to rediscover 100 years of making lasers and rockets, who cares? If we have to rediscover 2000 years of agriculture and medicine, that's a problem.

I think it would be reasonable to store a fixed-focus compound microscope as part of your archive. That's basically a tube and a couple of lenses, and it gets you to 500x right off the bat.

But if you want to be doubly sure, include instructions for making a hand lens at 1x, and instructions for making a microscope at 5x.

Maybe it would be possible to make the archive in "stages"? Like the first stage is readable without any equipment and describes how to make equipment to read the next stage. Repeat until you have a high-precision microscrope and can read the nanofiches.

> We must accept that there's some ongoing upkeep of an archive regardless of how durable its contents are.

That's a good point too. Archiving is a living process.

I love the "theory" that optical media, especially the golden DVDs will be readable in "100 years" or whatever hilariously optimistic length of time claimed. Sure, DVDs and even CDs i have carefully stored in a humidity controlled environment worked the last time i checked, there's always the risk that the next time i pull them out of storage they won't. I think blu-ray is claimed to be even longer than DVD for storage, and i guess it makes sense in theory, the QC/QM needed to pack 120GB on a 12cm disc must be wild.

I have books on my bookshelf that are older than 100 years old. I don't think 100 years is even that much to ask, seeing as i know who William Shakesman and Chaucer are.

I'm in a unique position, though - my music is on an SD card traveling in space right now, thanks to some attentive friends who petitioned to get me on there. Or so i'm told. How would i know, my spaceship is in the shop.

plastic media optical disks don’t last 100 years, but glass media optical disks do. They exist. My family’s small business used to sell, refurbish, and maintain big jukeboxes for 12 inch Write-Once, Read-Many glass optical disks (in these big minidisc-like cartridges) for customers like big banks that needed guaranteed 50 year archives.

have you ever read a disk that was 50 years old?

The tech isn’t yet 50 years old, although accelerated aging is a well-understood field and the margins are good for glass media. However, the tech is 30-40 years old and the early disks ARE still readable.

You're understating how long good-quality paper will last. Proper archival paper lasts hundreds of years without fading or discolouration—we have no idea what the upper limit on its lifetime is.

And, in practice, it doesn't matter whether the storage medium can't be read because it's damaged or because the machine to read it doesn't exist anymore. Either you can read it or you can't.

Of course it matters a great deal - you can build the machine again

Not necessarily. You can't just build a high-powered microscope out of twigs and leaves. You need sophisticated manufacturing tools. Mineral resources. Where are we going to encode the knowledge of how to build or acquire these things, if already having them is a prerequisite for accessing any of our records? Either you can read it—i.e. it is intact and you have the means to access the information stored inside—or you can't.

> Where are we going to encode the knowledge of how to build or acquire these things, if already having them is a prerequisite for accessing any of our records?

It's not a prerequisite, you can have a smaller batch of knowledge on how to build the tools to access all the knowledge in various other more accessible systems.

Besides, you presuppose that just because there is no machine, there is also no knowledge re. how to build a machine and you only have twigs and leaves. That's a false universal assumption, you could just as well be in a situation where you don't have the machine, but you have tools and resources to build it. This is yet another difference to the "the medium is broken and can't be used" situation

"I think Microsoft Research was developing a technique of storing data in a 3D arrangement in glass. Very fascinating."

a-la Superman ?

If all you're trying to preserve is text, then a digital format would make sense. Just add "analog" images of unicode tables covering all used symbols, and any used error correction format.

But if this is trying to entice libraries with "replace your microfiche with this easier to maintain format, and do something for history while you are at it" then you need to store scans of newspapers, magazines, etc. Which probably means storing PNGs or similar. At that point, I'm not sure the capacity improvement over "analog" optically readable images is really worth the much more complex decoding setup.

Both for the library use case and for the "a civilization that might not speak our language wants to decode this" use case the complexity of the decoding setup is a major factor.

Since this is using halftoning and discrete dots, it's technically "digital". You could add some columns for error correction if you wanted.

Strictly speaking you only need books enough to teach the reader to build up to an optical 100x microscope and then use that to access the rest. Eventually you teach them how to build blue lasers and rockets to scan the silica bricks you engraved and stashed in a moon cave for "everything else".

How useful are books about building an optical 100x+ microscope from scratch to people who have survived a global catastrophic event? Relearning how to make blue lasers is probably extremely low on the list of priorities compared to, say, farming or desalinization. You'd be relying on a fairly niche group of apocalypse survivors to actively preserve the knowledge of microscope production so that someday in the not-very-near future someone might want to learn about quarks or sixteenth century music.

It makes a lot of sense to make the most useful information for apocalypse survivors the most accessible information. Just make books about that information and keep them maintained. Then there's at least incentive to keep the books protected and maintained.

Edit: it sounds like you'd probably need a 1000x to 10000x microscope to really read the nanofiche described in this article. So a potter's wheel is probably not sufficient. Plus, you'd need the materials and tools for working with glass, which might not even be feasible in an apocalypse.

An apocalypse is not a weird time reset. You still have giant glass furnaces and propane and even industrial control systems.

The tech that survives is probably an odd amalgamation, but chemistry is basically the thing you really want for society - antibiotics, explosives, and fuel.

How can you keep propane production, giant glass furnaces, and industrial control systems working but you can't maintain a few hundred warehouses of "normal" archival material? Or just continuing to print books?

The trick is that you would need to keep the archival material “continously” while the microscope tech can be lost, forgotten, rediscovered, and then used to read the material.

That is if you have a thousand year gap in your archival material warehouse maintenance the data is lost. If you have a thousand year gap in your ability to make microscopes the data will wait around patiently.

The point is that the time scale of "lost, forgotten, rediscovered" is perhaps orders of magnitude longer than it is to be useful for much of humanity. And you similarly need to cross your fingers and hope the facility isn't burned to the ground or blasted apart or whatever. An indestructible nanofiche disk at the bottom of a forgotten pile of rubble is going to do little good to anyone.

You can have a whole bunch of caches of preservation materials in a variety of formats. Clay tablets, nanofiche, laser etched silicon, etc.

> And you similarly need to cross your fingers and hope the facility isn't burned to the ground or blasted apart or whatever.

Except you don’t need a “facility”. A lot of lot of knowledge fits on a bank card size nanofiche. You make ten thousand of them, and people put them away as best as they can. Some will melt in a fire, or get blasted apart, some will be burried and forgotten forever. But some will be more lucky, by just chance they will end up in an atick or crawlspace where nobody disturbs them.

Not counting that we can hide some of them in places where they are unlikely to be disturbed, such as place a few on the moon or place a copy on every single geo and lagrange point destined satelite. Heck, burry a few on a glacier and they will naturally re-appear in a few thousand years.

I think its fine, microscopes are not that hard to make. If you have a potters' wheel and some abrasive you could work up to it. Small price for access to the old worlds' knowledge, certainly easier than building a record player like the voyager probe suggests xD

If you know what you're making, sure. Europeans kind of lucked out historically, since there was a lot of glass technology in the lead up to optics and then microscopes. Historical China got kind of fucked in that regard, since they did have mirrors, but porcelain was so good they never had to get into the glass part of the tech tree.

Oh… that’s interesting. I’ve never thought of that aspect of why maybe the industrial Revolution took off in Europe but not China. Lenses allowed the development of the telescope, reading glasses (extending working years of scientists and engineers, etc, enabling finer work on clockwork mechanisms), the sextant, the barometer (and thus the study of vacuum that led to steam engines), and the microscope. All that predated the industrial Revolution, laying the scientific groundwork for all that followed.

(Obviously things like that are multi-causal, but I had never heard that idea before, that the amazing porcelain of China kind of held back the development of glass and all its many, many applications.)

Yup. I've heard that the lack of easily navigable rivers and no coal to speak of caused more of an impact on china hitting industrial revolution before europe.

> If you're really trying to preserve basic knowledge to restart civilization, I'd think you'd want libraries of full-size readable books spread across the world. Not something that requires you to first build a microscope.

It's not an either/or though. Optimistically we'll be able to store all our collective knowledge in all accessible formats, digital / internet accessible, libraries, etc.

But nobody can see the future. Having this as a backup sounds great. Archive all of e.g. Wikipedia, put it in a time capsule along with a microscope and/or readable instructions on how to make one, make thousands of them and put them in landmarks, caves, monuments, dumb sattelites that are designed to land on ground in 10.000 years, the moon, around the sun with a homing signal (e.g. reflective surface since electronics won't work after 1000 years), etc.

A library can catch fire, the earth may get glassed in WW3, but surely some of these would then survive for future generations or alien archeologists in the millennia to come.

They're storing it on the moon. I think if you can manage to get to the moon, you'll be able to get a good enough microscope.

Plot twist: when they get to the location, they find a similar plate already there from the last time earthlings climbed out of the mud.

…but the reading device (not a microscope but some unrecognizable tech) did not survive.

Any decently-designed library like this is going to have some type of easily-read instruction sheet that shows how the bulk of the data is encoded and how to read it, perhaps with plans for a device to do so.

Yes but it is very possible such a README does not survive, is lost, or separated from the library.

> not a microscope but some unrecognizable tech

But that is the beauty of a microscope. It is not some “unrecognizable tech”. Seeing tiny things in detail is a basic skill many new civilisations can re-discover on their own. It doesn’t matter if it is our human descendants, smart octopuses or the silurians. They will be interested in seeing tiny things and thus they will construct their equivalent of microscopes.

The important thing is sign-posting. How will they know where to look?

Why would octopuses be interested in seeing tiny things?

If they develop a civilization, they'll want to develop technology that requires tiny parts (e.g. computers).

is it a tall, black obelisk?

And after decoding and translating it starts with "We're no strangers to love..."

When we reached the Moon in 1969, could we have read it?

Yes, extremely easily.

Microscopes are old tech. The first ones were constructed in around 1600.

In a "humanity collapse" scenario, for surviving humans it would be much much much more valuable to have 'reboot instructions' for hunter-gatherer to industrial society, than beyond.

Of course you'd want to do both (early steps, and advanced knowledge).

And there will be various kinds of low-hanging fruit, to propel newly-industrialized society into the space faring internet age. Skipping resource-wasting, fossil burning stages entirely.

As for culture (novels, music, games, software, movies, etc): who cares, whatever makes it through the bottleneck in whatever way, shape or form.

The "Zones of Thought" series explored this. It was well-understood that societies could easily collapse and would need to start over from scratch repeatedly, particularly in the slow zones of the galaxy where advanced AI and FTL weren't possible.

I could imagine a safe the size of a gun safe completely containing a nano-fiche machine and all of the information on Wikipedia today. This isn’t a hard problem.

The hard problem is meting through the information. Fast, contextual search is probably the most important development we have today and you can’t really replace it with anything but tons of compute and digitization. Big ass physical libraries don’t solve this either.

Wikipedia doesn't nearly store as much information as you might think. It's basically the intro 101 on all subjects, but it barely scratches the surface of serious scientific fields. They have a million articles. A large science publisher has more than that for its journals, and there are many large publishers.

On their site they have a photo of

> Wikipedia printed on a nickel book with each leaflet holding 8,000 pages

It's tiny...

https://nanofiche.com/

> The hard problem is meting through the information.

It doesn’t have to be an instanteous thing. Just think of how much time schollars spend on deciphering papyrus fragments. If we would have a nanofiche of the contents of the library of Alexandria multiple departments would spend multiple lifetimes reading and contextualising it.

Reply on news.ycombinator.com