How much data do you want to preserve?

I know it sounds “boring” but it is hard to beat a good acid-free archival paper printed with a good quality ink. Stored in the right circumstances (not too humid, dark place, not on fire) it should last half a millennia easily. It is also pretty much guaranteed that whatever happens we will have the technology to read it again.

Exotics like laser engraved metal plates or ceramics might be better if you have a lot of data and can’t guarantee your storage will be fire and flood free. If you don’t have a lot of data you can also think of engraving it into stone or press it into a clay tablet and fire it. These mediums are known to be very stable even in adverse circumstances.

First example I thought of is the titanium punch sheets and wordlists for crypto wallet recovery phrases https://shop.ledger.com/products/cryptotag-zeus and https://github.com/bitcoin/bips/blob/master/bip-0039/bip-003...

There is also Microsoft Project Silica which I recall seeing in person at their EBC playing back a movie from it https://www.microsoft.com/en-us/research/project/project-sil...

Then there is m-disc which uses the DVD+R and BD-R/BDXL standards but writing to a sort of glass rather than to the traditional medium. These are cool as they play back n regular players too. https://www.mdisc.com/

I run a business that makes 3d printed braille molds that are used to repeatably emboss paper. I haven't considered the molds being offline storage, but I suppose they are. I mostly operate with the assumption of a shelf life of 10 years for the PETG molds, but ink free, embossed paper has excellent lifespan if stored correctly.

I guess you could consider it an "offline datastore as a service." It would be a pretty good offline storage of keys with a way to request a paper copy. Certainly issues of trust and physical security but wrapping it with encryption would be easy. Also benefit from your government's legal protections for mail. There might actually be a usecase here.

Couple fast facts:

- Current 26 * 32 = 832 cells * 6 dot braille = 4992 bits/mold/page

- Possible 28 * 34 = 952 cells * 6 dot braille = 5712 bits/mold/page

- Maybe some more headroom, but that's what is possible with current spacings

Related - this insane project to build the "most 3D printed" 3D printer [0]. The author has 3D printed mechanical components, motor windings, and a complicated electomechanical "motor controller".

The whole thing is intended to work from 3D printed punch cards.

I'm not sure how much of this project actually works, but it is interesting.

0 - https://clickprintchem.wordpress.com/2022/05/31/self-replica...

I'd probably also pick some kind of "punched" card; a print of some kind with holes in one place and not in another. The presence of a hole could be checked with a bed probe if everything was lined up correctly. I doubt the plastic would be intact for centuries like you want, but it would certainly outlast the printer you use to read it. A paper card would probably last longer though.

For high density, your best option is probably microfilm.

Seems like low data density. I recall some people talking about laser-engraving. IMHO engraving on a solid chemically-inert surface is more likely to be useful.

There are M-Disks. These are CD/DVD/BluRay disks which use a drive with a higher power laser and work by ablating a metal layer, rather than a photosensitive dye as in the lower-powered disks. Regular drives will read both kinds.

For a small amount of data (crypto keys?), consider deep laser engraving on stainless steel. That's very durable. Or even engrave text into stainless steel with a small CNC mill.

You can engrave QR codes, bar codes, etc. But there's a lot to be said for engraving plain text.

You can print QR code with different heights representing white or black. That piece of plastic would last a long time, and you could encode.....something. You could pint a version 40 (177x177) grid with a Prusa XL easily, and that would last for a few thousand years and be waterproof and fairly resilient to a lot of things. You could probably print a 177x177 grid on a regular Prusa MK4 and just have a color change layer. E.g. layer 1 white, and then layer 2 black. This would be 2,953 bytes.

You could even get a binder, and then print the QR codes sheets as 'pages'. You could print these thin enough to do 1mm sheets. That means ~ 120 of these in a 5 inch binder. So, a 3 ring binder of 1mm sheets of 177x177 QR codes would hold about 346.2 KiB. You could store encryption keys in this way. You can probably push the density well past 177x177 even on a Mk4s pretty safely. I may try this later today.

Hey all! Wild to see my silly question pop up on the front page, but very cool! I hope it was because of a spike in interest for some reason, but...

Please forgive me, this is the first time I've ever been on the front page, but I'm seeing something odd? I posted this topic days ago and it was well received and then kind of done. But now it is showing both on the front page and on the ask page as having been posted only 8 hours ago. Is that normal?

I did see a topic the other day about HN being flooded with content, where some mods responded that they were doing work in that area. If I hadn't seen that, I probably would have just assumed that any ask that gets bumped up to the front page might get re-zeroed for posting time on all the existing comments. But given the other topic, I thought I'd pop back in and ask!

That aside, I do appreciate everyone's comments! I can't respond to them all, but hopefully I've hit all the main suggestions so far. Thanks again for the breadth of consideration!

why not print onto paper and put it in a safety deposit box?

your X/Y resolution isnt going to be fantastic woth filament, but your Z is gonna really suck.

you could instead print(emboss?) like a barcode on some paper, and encase that in resin. the big benefit being that you can read it non-destructively. keep it out of the sun so whatever ink doesnt fade and you're set

Wouldn't that be just QR codes (and equivalents)? I suppose 3D printers can be used to etch/print them onto a durable material and then have it read back using the measuring tools you mention, but at that point I think you would be better off just 3D-printing out something like a a vinyl disc maker/reader and using that.

Modern chips made via EUV lithography can be thought of as 3d printed. Like the most mind bogglingly complicated 3d printing imaginable.

I dont think TSMC make ROM chips using their 5nm process but I think they _could_, if needed. So that would be a very compact way of storing info.

To satisfy the "disk" condition, would a vinyl record suit your needs?

In the abstract, I don't see an obvious practical advantage of 3d steganography over 2d because the tools for 2d (printers and cameras) are more readily accessible and the processes of creation and reading are faster...particularly reading.

There might be corner cases where 3d makes sense...but it is hard to compete with decals/stickers, and 2d steganography can also use color and saturation as additional data dimensions.

But I could be wrong.

3D printing (additive manufacturing) usually involves layers that tend to be a weak point.

If you want something truly long lasting you might find substractive manufacturing techniques like CNC milling better. Indeed there are solutions from companies like Cryptotag for punching cryptographic seed data into plates / blocks of titanium. Something like this could be automated with a CNC router to store a bunch more data.

You could use letters for blind people with the amount of different letters you decide to use as base of a number system.

A "vinyl" record is probably your cleanest option. Record players are pretty common so there's much less risk of losing the ability to turn it back into a digital file. (physical object to .wav file is commodity, .wav to you your original binary data is left as an exercise for the reader.)

3d printing filaments generally aren't very heat resistant, by the nature of the process. There are ways around that, like 3d printing metal or clay. But if we are allowed attachments to 3d printers I'd choose a laser or a pen, then use optical equipment for reading back the data

Easiest way (I'm sure it's been done) would be to encode as QR codes and print that. (Multi-filament for two colours, or raise one of them enough for shadow or filament change.)

Best choice will be thick metal medium, such as stainless steel.

Probably use a 177x177 QR code with Q or H error correction engraved (deeply). I would probably compress the data and encode it as Hex so it can be stored in alphanumeric mode on the QR. You might need a series of QRs for larger data.

This was solved with M-DISC. The data is literally glassy carbon pits. http://www.microscopy-uk.org.uk/mag/artsep16/mol-mdisc-revie...

For long term data storage, you might consider using a UV laser to etch inside a crystal cube. This should last for thousands of years I would think. You could also read back the bits using optics which is non destructive.

Some sort of wire recorder? The teeth of the 3d printer ‘driver’ (sorry lost on the terminology here) already make an imprint on the filament, maybe it could be made intentional

you could maybe do the inverse of what a vinyl does or something. the problem would likely be reading the data back efficiently for whatever it represents more than some encodingscheme to put data onto a surface (just looking at it from an encoding issue without problems of running out of surface so to say :p)

You could print an old-school punch card, bring it full circle.

There are special companys, like for ex. Axnum, the make needle printers for steel.

Depending on how you define 3D printing: braille.

Do you want to build a time-capsule?

I've experimented with this using one of the commercial print-on-demand services. Not filament, a binder-jet sintering marine stainless steel type of process, IIRC. Slightly modified Noto Sans Extracondensed font with... Let me check the files... 0.1mm line width, 0.45mm lower-case "m" width, dense letter and line spacing, gets 3KB ASCII per square inch. ...Engrave, don't emboss, or the letters will come off with impact and corrosion!

Filling the engraved letters with black wax makes it optically legible under a good handheld magnifying glass. Then rub with vaseline/mineral oil/wax/ACF-50, mount on a brass holder (check relative galvanic corrosion order!), store pressed against aluminium or zinc sheet sacrificial anodes Just In Case(TM), inside polyamide or ABS case. Should last basically forever.

I designed it for manual reading, so the holder doubles as the "reader". Basically just a spindle going through a hole in the disk, holding a margin in the edge and center of the disk. The text is a spiral, so you can spin the disk to help keep the viewed region aligned under a higher-power microscope.

For home printing, probably the way to go is Formlabs wax resin (or equivalent), then either learn brass casting yourself or hire a jeweller to do it. Though you won't get as good resolution/density as the metal process I tried. And really, laser engraving'll probably be cheaper and better. Consider if good-quality paper and ink, maybe laminated, inside a Pelican case'll be more practical.

I wouldn't personally trust any of the common plastic printing materials to hold up for important data under oxygen, UV, fatigue, heat cycles. Are you sure the resin's not overcured or undercured, the filament fused correctly and won't delaminate, it won't reach the glass transition temperature during summer? And bacteria is already evolving to eat plastics. Maybe SLS/MJF polyamide's okay, but in that case, I'm not sure I'd trust the sintered structure for small details (and they don't have great resolution anyway).

> The assumption was that using common 3D printer measurement tools (like for bed-leveling) would provide a way to read back whatever data was encoded onto the surface.

> And then, obviously, if that proof of concept exists, I'd wonder about some kind of advanced version that used specialized equipment for the reading (and possibly the writing/printing).

At these small amounts of data, specialized read hardware just adds risk IMO. Plain text can be read manually or with OCR. QR codes can automate reading with a standard flatbed scanner or smartphone camera.

Consider ideas for encoding: Plain text, B64, hex, Reed-Solomon codes... 1-bit depth structure turns most of 3D printing's storage density into redundancy, but anything truly 3D adds read risk. If you insist on automating reading, QR codes will get you error correction, encode/decode software, and COTS hardware for "free". Personally I think human readability is a big advantage too. Maybe OCR text, and put error codes in a QR beside the text for byte-perfect computer input?

Compare: Memory of Mankind uses ceramic tablets in a desert cave. Arch Mission Foundation project uses holographic glass. Long Now Rosetta Disk engraves with electron beam on nickel, IIRC was/is also commercially available for personallized jewellery. M-Disc and Bluray (HTL?) have modern digital storage density, good stability, work with commodity hardware. ...See design considerations of prior art for digital storage in 2D, naïve 3D version is to just use these as a heightmap:

- https://github.com/cyphar/paperback - https://github.com/za3k/qr-backup/ - https://github.com/colindean/optar - https://github.com/Sjlver/psst - https://github.com/schroeding/paperstorage - https://github.com/intra2net/paperbackup

Also, don't sleep on the centuries of work done by archivists and historians! The top comment is right; acid-free archival paper has very good overall cost, density, stability.

Disney has a patent on this