Winter time capsule, retrieved.

Over one year ago, in a thread where people were joking about having to bury hidden stashes of taboo data outlawed by a future authoritarian government, stored on thumbrives in tin coffee cans, I was inspired to test that out and actually see if various media could actually hold up under the harsh treatment of mother nature when all they might be protected by was discarded food containers. So that's exactly what I did. If you may remember my thread from a year ago, you guys gave me a bunch of memes and I took random files of different types from my /downloads/ folder and put them on a thumbdrive, a SD card, a CF card, a CD-ROM, and a floppy disk. These storage devices had nothing to protect them but a tin can, a plastic jar, and a plastic rice container. Last week, I went and dug it up.

I walked back to the GPS coordinates I had written down a year ago. I was surprised to see that some sticks I stabbed into the ground were still upright.

I started to dig. A year ago one of you asked why I would post the the GPS coordinates publicly, because that meant someone might go and destroy my little science experiment.

That's because I wanted to know if anyone actually would! ( I know there's a couple other lainchan users from this area.) So I planted a decoy. The decoy jar contained only a cute drawing and a single .357 magnum FMJ cartridge, a small gift to whoever decided to go out there and dig it up. On /k/, users will often leave stashes of ammo or other prizes in places that users will challenge each other to camp out at, but no one came for this, not that it was really the important part. I noticed that on the inside of the cap there was water. And the note had gotten wet as well. And it looked recent. This gave me a bad feeling about the actual time capsule, which was actually buried a few feet to the left of the decoy.

Now I started to dig it up for real.

First thing I noticed about the chinese food container when I opened it was that a spider laid eggs in it. First, fuarrrk spiders. Second, if spiders can get in, so can moisture. Yet the thing seemed sealed pretty tight when I pulled it open.

>>6000 GET

Next is the can. It seemed like mosture did get past the lid since there was rust on the inside.

I pulled the scrap paper packing I had stuffed in it and all the flash memory devices seemed free of any rust.

Now I head home, package in hand.

So before I post what I found I want to talk about what I thought would happen. What I was assuming last year was that the floppy disk will 100% die. Those things die under normal use in ideal conditions. No way can it live through this. I basically just put it in so I could say I tested some form of film type memory. I was pretty sure the SD card would die. I have always found them to be kind of brittle, not strong in general. And it was already several years old. But it's still electronic flash memory, and being cold and not plugged into anything for a year could easily be what kills it. I knew in my heart the CF card would live. I just felt that it would. Out of all the different types or memory cards I've always found CF cards to be physically the toughest. But it's still electronic flash memory, and being cold and not plugged into anything for a year could easily be what kills it. I thought that the thumbdrive might be fine. The USB connector was something I was worried about getting rusted beyond workable conditions...but it's still electronic flash memory, and being cold and not plugged into anything for a year could easily be what kills it. The CD-ROM was something I was worried about in terms of warping. If it's sitting underground it's not going to get scratched, but the change in temperature between winter, to summer, and to winter again is something that can certainly cause it to warp. I've had disks warp because of being in a hot car and they only semi-worked after that. If the CD ends up being readable at all, any one file being corrupted or not is a 50-50 coin toss. That's what I had been predicting, now lets see if I was right.

:0

Allow me to crack open this bottle of finely aged memes.

>>6008 Quite literally dank.

Put me in the screencap and put it in next year's capsule.

Let's try the SD card first.

>>6011 The SD card was....fine. It was perfect, actually. The is probably the oldest SD card I've owned and after being in the cold dark ground for a year it's just fine? I played all the video files, audo files, opened every document and viewed every image file. Not a single corruption. I was surprised.

Thumbdrive next.

>>6014 The thumbdrive was....also fine. I was most worried about it because of the metal USB connection itself, but not a single video file, audio file, document file, image file or whatever showed any signs of corruption. I will say that it is noticeably slow to read/write. I don't see why or how this would happen but I tried writing new data to it and even plugging it into a different computer and copying a big video to it and it really is just...slow. How could it take damage from the cold in a way that no data shows any signs of corruption but now it just reads+writes much slower than it used to? No idea. But there was zero data loss.

Last in the flash memory category is the CF card.

I'd wager the correct way to check whether there was actually no damage would have been to save an image of the drive before you stash it, pull an image after you retrieve it and see if they are identical.

>>6016 And the CF card is completely fine! I knew it would be! CD cards are tough little dudes. There was zero file corruption and it didn't have any mysterious slowdown like the thumbdrive did. Not much to say here, I was 99% sure it would be fine and I was right.

Now it's time to crack this one open. There is definite moisture contamination getting in there, as I can see spots of mold on the sticker label of the SD card.

>>6019 Let's try out the CD first.

>>6020 I have been proven wrong about the CD. All files on the CD are completely readable. Audio, video, text, image. All good. My PC seems to take a long ass time to read it, though. Reading data from CDs is never a fast process in the first place, but my computer seems to want spin the CD way more than I would think is needed to read any one file. Maybe something is physically off with it just a teeny tiny bit, but the computer can compensate for any unannounced read errors by just making a lot of repeated passes over the same spots with the laser. CD works, just slowly.

Now awaiting for your verdict on floppy disk

And now we hook up Ye Olde Floppie Drive and slide in this poor, unfortunate, sacrificial offering to science. Imagine being a brand-new floppy disk purchased direct from the manufacturer, stuffed in box with your many colorful siblings, sent across the world and ending up at a hospital. You're stored in a office. You await your task. "Oh boy! I hope I get an important job like holding passwords to a medial database!", you think. Soon a worker comes and takes one of your red siblings. The someone comes for a yellow sibling. For years you wait in the storage closet with your remaining siblings, but for many years no one comes. One day you are all handed over to some other nerd who takes you and your siblings home. Your new owner chooses you! Finally your time has come! You are taken from your siblings and instered into a PC. You are immediately filled with memes, stuffed into a used food container, taken to a forest, and dumped into a shallow grave. Time ticks by slowly in the darkness of this cold holl, as mold forms on your outside and baby spiders craw all over you. That's basically that this poor little disks life was like LMAO anyway let's see if it even reads.

Instead of jars and tupperware, what would actually be the best and cheapest ways to completely seal the contents from moisture? I'm also curious whether an internal HDD or portable HDD would survive. >>6017 Good idea.

>>6023 I am astonished! It actually reads! Good lord this thing is LOUD AS fuarrrk, even by normal floppy disk noise standards! And it makes a pretty bad crunching sound, which I assume is bad. But it frigging reads! I can actually see files and file names and holy soykaf was I wrong!

>>6024 just off the top of my head, you could take an altoids tin and try to seal the edge with a silicone glue to waterproof it, then just melt the glue when retrieving it it would be too small for the floppy or CD, but should fit several SD cards, flash drives, CF cards, etc.

>>6024 I think jars would still be your best bet, assuming moisture doesnt get in hdds should survive. might help to leave a bit of drying agent in the jar if you would put it away for years. and try to get the ones with rubber seals used for food storage. >>6025 neat! I had a feeling the floppy would pull through since it looked pretty new(from my experience floppies only die from severe old age), the crunching probably means it'll kill itself soon though.

But, thumbnails don't seem to loading, mostly. My file manager displays a thumbnail for a single pepe and nothing else. But it DOES show that those files ARE there, just corrupt? I try to open a .PDF file. The crunching noise is in full force at the .PDF loads for like 20 minutes before this error saying it "contains only empty pages" appears. The crunching stops. The file manager says the .PDF does still take up about 400KB of space, so something is still there, but it is corrupt.

>>6028 Can you open the smug-pepe image?

>>6028 Have you tried just moving it all over to your main drive? It could just be that the read speeds are so slow that it's failing to load. Otherwise do it anyway then send us some samples and we check for errors.

>>5996 I've been awaiting this for so long, OP you just made my day. This is probably the best thread I've seen on here in a long time! After seeing this, I would be interested to see what lainons would think of putting a hard drive in a food saver vacuum bag with a pack or two of desiccant. You would not not have to focus on an outside container being water proofed, and you would just need a structural container to keep the cold/dirt from physically destroying the inner packing.

I tried to open a .TXT file in mousepad and it displayed an empty file. It did open, it was just blank. I was surprised as I would have thought a file format like .TXT was pretty barebones and simple compared to images. If anything would survive it'd be the .TXT files, or so I thought. I clicked on one of the images to get Mirage going. While the files on the floppy disk is mostly image files, when Mirage opens it shows that it only even recognizes a mere four of the itotal mage files on the disk as actual image files. Two of the image files start to have thumbnails generated for them but only the first few rows of pixels get made and the rest is just grey. Clicking on them yields nothing. They only get a tiny bit of thumbnail and they can't be opened. A third file that Mirage actually recognizes as a real image file gets stuck while generating its thumbnail and never finishes. It also cannot be opened. Out of all the images on the floppy disk only a single pepe is allowed a fully generated thumbnail, but even he cannot be opened. He sits in the side pane of Mirage, staring at me with a smug look. He seems to say: "The Costanzas, the Swole cat, the pokemon, the Gondola, even the anime girls. So many in this floppy disk have perished. Now I alone remain, and I won't let you open me." The green creature is trapped in a state of purgatory the green floppy disk. Too lazy to die, too corrupt to open. Even in his tiny thumbnail, I rue the look in his eyes. The crunching sound is louder, and I figure something might be not mechanically lining up just right. I don't want anything to break or get worse. I eject the disk and put it back in.

>>6032 >Out of all the images on the floppy disk only a single pepe is allowed a fully generated thumbnail, but even he cannot be opened. He sits in the side pane of Mirage, staring at me with a smug look. He seems to say: "The Costanzas, the Swole cat, the pokemon, the Gondola, even the anime girls. So many in this floppy disk have perished. Now I alone remain, and I won't let you open me." The green creature is trapped in a state of purgatory the green floppy disk. Too lazy to die, too corrupt to open. I love how your write, I have been following this thread with a smile on my face. I'm also getting some good ideas on how to improve data retention and protecting it from the elements.

>>6032 >>6030 I am sorry to say that it seems like this is the end of this little green disks story. All that noise it was making must have been it's death throes. This disk is entirely unmountable now. I don't know what to say. At first I didn't care. I expected the green floppy disk to die in that grave. And after a year I expected to bring it back dead. Since I had dug up its coffin and found it impossibly clinging to life I became emotionally invested in it. I am, or was, excited by it's ability to still be kicking after what it went though and I wanted to try what >>6030 said to check out its other files but now...I think that's all, folks. This is goodbye.

Good thread. Time for another time capsule with improved water resistance?

>>6034 Oh no. ( ;_;)

Press F to pay respects.

would be interesting if you used a hdd and checked how many bad sectors accumulated over a year, and checked the contents against an image. also consider a sealed plastic wrap around your container. vacuum, or just a gallon ziplock would go a long way toward keeping moisture out. if it was data i actually cared about, id include dessicant and use an ammo box with 2 or 3 layers of sealed plastic.

For the next tests, vacuum-sealed bags and mason jars. Try to see if adding desiccant makes a difference or not (so 4 containers, really).

>>6017 I think it would be interesting to use something like ssdeep or some other fuzzy hash mechanism to see how it gets corrupted, rather than just a binary "yes or no". Another test which could be interesting is seeing if packing say... 10% recovery par2 files would be enough to deal with any incurred corruption.

Very cool OP!

>>6001 maybe the spider was inside when you locked it? >>6006 temperature underground is usually way more constant than above, so the deeper you go the less temperature variations you get (for the CD). Even 20cm underground, temperature change (esp. in a forest under humus) is veeeery slow so there's little risk of endomaging warping i think.

>>6038 F.

>>6038 F.

>>6038 F.

>>6041 I think the mason jars will work better. They won't keep the moisture out unless you actually canned your time capsule (this would obviously be too hot for digital storage), but I can't imagine any insects getting in. I look forward to seeing how the bags perform. >>6038 o7

>>6090 What if you closed the container inside a vaccume container? or filled it with a heavy gas like nitrogen in a bucket to purge it. This is a cheap trick for purging and assembling nitrogen filled electronics and some optical devices. Step one is to put the gas into a bucket, step two is to put the object in there and shake it around so the air surfaces on the gas, and finally you reassemble in the gas filled container.

>>6038 F Thanks for the follow up OP, this is some fun soykaf

>>6038 And since now, floppies are to be concidered unreliable.

>>6038 One of the best thread I've seen, gratz

>>6006 warping tends to occur with very sudden temperature changes.

>>6038 F.

Great experiment with valuable lessons. Good job, lain.

>>6038 F

>>6038 F.

nice thread OP >>6038 see you space cowboy

There are various methods of minimizing humidity inside container with data storage media. Most basic is just layering containers. You can throw silica gel granules inside. You can also use a heat sealer for bags. etc. etc.

>>6038 this was great man, thank you for the effort

Probably wouldn't survive a lightning strike or broken power line sending current through wet soil with these methods.

>>6319 Neither would you.

>>6319 Well how fuarrrking likely is that burried in the middle of a forrest.

Comfy thread my Lainon. Thanks. Thank to the ones who bumped it. All hail the Great Floppy.

>>6323 Quite likely considering once lightning hits trees it would go into the ground and dissipate through the soil. It wont be full strength but enough to cause damage unless it's electrically insulated. This can be tested using a Tesla coil or perhaps just a high power battery with wires going into wet soil with added electrolyte.

OP here. I am happy you all seem to be enjoying my findings. Some of you talked about doing the experiment again, but with better materials and containers. I believe this defeats the purpose. The point of the the whole experiment was to see if we could possibly hide away derezzed information with soykafty materials during an emergency. (And it seems we can!) If we redo the experiment with containers actually made for this type of thing, then I don't really think there's much of an experiment to be had. A container like that would probably keep the data safe for longer than my entire lifespan. I've been working on other projects lately, but I had a lot of fun with this. It seems that the truth is that CDs and flash memory is a lot tougher than I thought!

>>6339 ok fren, see you next year in the winter time capsule under the sea :)

>>6350 Like bioshock? >>6038 F. Body too short? What is this soykaf, Lainchan? Please let me post. Pretty, please?

I'd like to see how well LTO and DLT tapes hold up in a time capsule, since they're designed for long-term storage.

>>6468 Don't those cost a lot?

>>6023 Genuinely laughed out loud

>>6017 >>6024 >>6028 >>6032 The correct way would have been using a sha512 checksum file. For example on some Linux: cd "/media/grey/disk/timecapsule" # make checksum find . -type f -exec sha256sum {} + | tee timecapsule.sha512 # check sha256sum -c timecapsule.sha512

>>6038 A big hearty F to my little green guy. Rest in data fragments.

>>6556 I never knew about this.

Nice experiment anon, pic unrelated

>>6556 you're right, but this way a single bit flip will register as a difference and if he relies on this only he'll never find the extent of the damage

>>6038 Green-kun really did his best. F 7

I'm a bit late, but would there be different results if tested in a tropical climate with big amounts of precipitation and heat? Also, are there chances of the capsule being moved to another place if the soil is too loose? Some biomes, such as the Amazon rainforest (as the name implies) become flooded during seaons. My Hypothesis is that too much material would be deposited on top of the capsule, making it difficult to dig and find it.

>>7416 There's little organic matter in the soil of tropical rain forests and the soil is held together by vegetation. The capsule should be fine if heavy and buried deep enough, also don't bury it near a river.

>>5996 Is there a handbook of best practices for making time capsules? At least for ones stuffed into buildings brickwork, those were quire popular at the beginning of 20th century. I've seen an instruction on burying firearms and ammo underground once. It included lots of machine oil and plastic bags, but we're dealing with small digital media here, so it's better to worry about data integrity: so duplicate copies, parity checks and the device capable of reading it should be included in a serious package of future-proof anti-totalitarian geostashe. Actually thinking in this notion I've devised a simple idea. NFC/Mifare memory cards can hold up to 8 KiB of data on them, can be write-protected or have 50k read-write cycles, they also can be read with average smartphone's antenna assembly behind a thin (<4mm) layer of dielectric material: plastic, wood, ceramic tile and I assume any sort of concrete-like or gypsum plaster. In a manner wireless chargers are embedded into furniture. Before tiling your bathroom you glue NFC chips to the wall, now you have a whole wall of data, or just a tiny spot for emergency stuff. Maybe stuff a few inside your bed's legs, or since furniture is more accessible - it's possible to hide whole microSD in a tiny slit.

>>6015 >steal this book based

>>8756 Anything that's good for dry food is more than good for solid-state digital media (CDs and MicroSDs). I was thinking more in a sense of manuals for guerillas and whatever stashes were discovered by authorities historically.

>>5996 very nice thread OP >>9830 >I think I might have fun running a similar test, except with burying data within properly sealed containers for extended periods. Godspeed lainon

>>5996 Unrelated question: why the op pic can't be saved in original resolution via "save image as.."(5.83MiB), but saves just fine via "copy image"(29.28MiB)? As far as i know, browsers usually don't stealth compress images when downloading.

>>9931 "Copy image" is a decompressed bitmap of this image put into PNG container. But the image saves fine as JPG on my machine. Right click - "Save link as".

>>9829 Also saved for my own personal archive. This was a great read :) >>6038 F This picture shall become my wallpaper now. Kinda makes me want to try it myself.

>>9933 Oh, my bad. Just opened in wrong app.

Very interesting thread OP >>9830 Bury them below the frost line. A PVC pipe with the caps glued on would probably work well but it's hard to fit a CD inside, and I think CDs will end up being the long term winner. Use a gold archival disk and a basic cd-rw.

>>6859 For longer time spans, tape isn't good. The magnetic material will delaminate from the structural tape. Speaking of. Any flash storage will leak charge in time and have bit flips. CD-, DVD- and BD- RWs are made from organic materials that decay. (Not sure about -Rs, I think so too) The mass produced CDs and DVDs are stamped so they can live way longer. There are MDs - Millennium Discs, those are DVDs and BDs made out of inorganic material that claim to survive for 1000 years, hence the name. Sadly I think only 1 corpo has the right to produce them now so supply may dwindle as most people don't use discs anymore. You can read them in any DVD or BD drive, but u need a special writer since it uses a different, more powerful laser to etch the disc. Then there are HDDs, with proper storage they might be the best density vs. lifespan? As long as moisture doesn't ruin thr mechanism I think the platters can keep their data for a long time. Please do correct me if this isn't true.

This a really cool idea. A long time ago (like 2008) when I was still in school, we had a class project where we went outdoors to do geocaching and we were given some little handheld GPS devices to find stashes. Your post sparked an interest for me to try it out again. But I feel like someone, whose uninterested in such things, would just steal it if they found it. I could bury it and try to leave a clue behind that would help someone who's aware of the cache but in a way that a normal park-goer won't notice. I'd throw in a small flash drive and could put in a small notebook and maybe write down a link to a secure form of communication. I'd have to think about what all I'd want to put in it.

this thread gives me an idea, i live on the west coast. thinking about trying this but adding it with a message in a bottle. doubling container plus silica packets, might work. would have to test it with a 5g bucket of water outside for a couple of weeks to test before littering the sea with more trash.

>>11271 mind that tap water aint sea water so maybe test with some of the real stuff and not the soycafe you bathe in

>>6038 F. Might want to try sealing the jar with wax for the next year, it's an old and easy trick for this sort-of stuff.

If I to store data for a long time in uncontrolled conditions I would use EEPROMs. They are supposed to store data for over 100 years and work in a wide range of temperatures, especially automotive ones can tolerate -65 to +150 C. Usualy they can store a couple of megabits. Enough to store a cryptographic key, a password vault or a small book. Despite manufacturers promising storage time over 100 years (or even 200), no chip actually existed for such long time to be tested. Likely these figures come from extrapolating results of accelerated aging tests, like exposing chips to high temperatures, UV, vibration, etc. Even if manufacturers put their money on it, there is always a possibility of some degradation through a yet undiscovered process. I would store three copies of the same data on three different chips. One of them would be some fancy brand name chip, the best you can afford. $10 Microchip is good enough for me. If you have money to burn you can buy an aerospace-grade chip for hundreds of dollars, but it's likely to come from the same conveyor as the cheap one, as it is likely that the price difference is only in manufacturer's responsibility for higher promises about chip performance. The second one is the cheapiest garbage you could find. $0.1 Noname from Aliexpress. The third one is a good quality ferroelectric chip. In principle it can be a different combination of manufacturers, quality levels and technologies. The most imortant here is to have as much variation as possible in as many variables as possible to minimize the probability of a correlated failure. If the modern silicon process used to maufacture the fancy chip turns out to be susceptible to doping migration after decades of storage, it is possible that the cheap chip made with older process will survive. If it turns out that flash memory can't reliably store data over long durations of time without regular supply of power, then maybe ferroelectric memory will survive. (The problem of loosing data without power cerrainly exists for modern high-density flash used in SSDs. I am saying that as a fan of SSDs who has an SSD RAID.) As stated above temperature is not a problem for EEPROMs. What might be a problem is humidity. I was unable to find humidity rating in datasheets I looked up. In any case, humidity can be solved by potting chips in parafin. Depending on it's type it can melt at ~50 C, which is in far in the tolerable temperature range. If you really paranoid about some EMI pulse from nuclear explosion, alien invasion or someting wiping out all electronics, you can make a box out of copper-plated textolite, soldered shut along all 12 seams. To make it actually solderable, each edge has to be filed at 45 degree angle from the unplated side. And solder pins on every chip together for a good measure. I primarly consider SPI/I2C chips, since they are the most common and easily programmable type. They come in various packages, but I prefer DIP-8 or SOIC-8. Both can be connected without soldering. DIP-8 can be connected with female jumper wires or a ZIF socket on a programmer. SOIC-8 can be connected with a SOIC-8 clip. A programmer can be bought for $5 on aliexpress, clip and delivery included. The programmer should work out of the box on Linux with flashrom. It is literaly as easy as buying a programmer, connecting it between a PC and a chip and then flashing any file to the chip with flashrom. (Assuming you checked for correct pinout and voltage levels.) Alternatively Arduino can be used as a programmer. As a bonus, simple protocols like SPI or I2C can be used on low grade hardware and can be bit banged on anything remotely reminding of a computer. In post-apacalyptic scenario if there will be any machine capable of processing any iformation it is likely to be able to speak SPI/I2C, unlike SATA, NVME or magnetic/optical disk/tape.

>>9916 thankyou so much for this lainon

>>5996 >>6008 >>6013 >>6038 >>9916 Lmao based physical archiver thread! We need to preserve the dankest memes for future generations and confused historians! Pls put me in the screencap