Slashdot Mirror
Data Preservation and How Ancient Egypt Got It Right
storagedude writes to tell us that a storage geek has an interesting article on why ancient Egyptians were better than us at data preservation — and what we need to do to get caught up. "After rocks, the human race moved on to writing on animal skins and papyrus, which were faster at recording but didn't last nearly as long. Paper and printing presses were even faster, but also deteriorated more quickly. Starting to see a pattern? And now we have digital records, which might last a decade before becoming obsolete. Recording and handing down history thus becomes an increasingly daunting task, as each generation of media must be migrated to the next at a faster and faster rate, or we risk losing vital records."
As recording things became easier, more things were recorded. At some point we began recording things that no-one will ever care about, and now keep things recorded that we didn't even know were recorded (care to see my router logs?). The less significant something is, the less we need to worry about preserving it. Of course, there are things worth preserving, but most of it just isn't.
A lot of data retention is because of legal requirements. At the bank I work at, we're required to keep *everything* for at least seven years - all our emails are archived, instant messenger communications, etc.
There's no place like localhost
I don't know of any other way to preserve our pr0n on rocks.
Etch barcodes into rocks.
I am very small, utmostly microscopic.
AS technology changes, so will the format of the data.
Sure, in days of yore, before everyone had hard drives, and data cluster, certain technologies would become obsolete and reader would go away. Those days are gone.
As everything moves to the 'cloud' the data will be stored forever.
well, OK, until something happens that destroys all the 'nodes' that are housing data.
The Kruger Dunning explains most post on
I'm not sure if they "got it right". After a few thousand years we have yet to agree on what they were even writing.
Interesting, TFA goes on about strategies for making sure stuff lasts. But he even touches on the more interesting facet of this briefly - no one can read the damn Hieroglyphs any more, so what does it matter that it lasted 4000 years?
What is more interesting to me is a way to cheaply, efficiently, include a sort of Rosetta Stone along with archival data meant for long-term storage. Hell, even the devices themselves... he talks at the end a bit about format issues, frex. Some kind of key to the interface or logic needed to reconstruct the method of reading the medium..? Anyone got a wax cylander lying around? If you ran across one, how long would it take you to be able to hear what was on it - and what're the odds of you damaging it in the process, especially if you had to dig up schematics and build a player yourself..?
That which does not kill us makes us... st
When we are born the only thing we know how to do is suck on a tit. We must continually pass down the information from one generation to the next or it will be lost forever. We had better find a way to permanently store our history or we will lose everything our for-fathers have worked for.
Why would a future generation want to read about me? Why ruin a perfectly good rock with a biography? ;)
Actually it's an interesting topic. On the other hand, we have a lot of backups now. We are much more efficient at producing a backup now. It's a tradeoff of producing many copies quickly or few copies that last for a long time (i.e., a chunk of rock).
Seems that the more copies you have, the easier it is to retain them through history... proliferation as opposed to preservation.
In other words: offsite [reliable] backups.
Seriously what a piece of complete and utter rubbish. From Ancient Egypt we have an extremely limited set of information because stone tablets crack and they aren't exactly the most portable things in the world. Go through to the Romans and paper, and the Chinese and you are seeing massively more information become available down the centuries. Zoom forwards into the 14th Century and we have a massively detailed view of what life was like which becomes more and more detailed as time goes by. The key here is detail, the amount of information in Ancient Egypt was huge, probably comparable to today, but the amount that was etched onto pyramids was tiny and quite a lot of that didn't survive anyway.
The key things that future historians need are prime sources and one thing that the internet is massively impressive at is the duplication of information and the avoidance of redundancy. Stone is rubbish for this, no-one bothers making copies so you lose the original and you lose everything.
Printing introduced simpler copies which meant that the information was more likely to survive down the years. With modern digital technology this increases still further. It is ridiculous to claim that digitally we won't have more information about the major events and people of today which is available in 400 years. We will have more CRAP available in 400 years (blogs, twitter, Slashdot) than any generation of historians have had to wade through.
Digital technology makes accurate duplication simple and that is the most powerful way to make sure information survives. Wikileaks is the embodiment of that view. The issue is that there is now SO MUCH CRAP that the issue for future historians will be in wading through all of the blog posts of "Obama is a Muslim" to find out that in fact he wasn't.
A rubbish supposition which is massively undermined by every time there is a censorship case the plea to "mirror the information".
Some information will be lost but the amount that will survive is miles higher than the amount of information that survived from Ancient Egypt. For instance its amazing to Bible Literalists that NOT ONCE in their SIX THOUSAND YEARS OF RECORDED HISTORY did the ancient Egyptians ever mention all getting drowned in a global flood... and you'd have thought they'd have noticed that.
An Eye for an Eye will make the whole world blind - Gandhi
Etch barcodes into rocks.
We Lenny them into rocks.
If it wasn't for the lucky find and preservation of the Rosetta stone, how long would it have taken us to decipher Egyptian hierogylphics? Not exactly an open standard...
I know nothing about the field of data preservation, but is there a Darwinian pruning of data that occurs? Do we really need to keep copies of ALL of our data for thousands of years, or do the truly "vital" emails/books/stone tablets have a much greater lifespan because they have actual value?
Yea, rocks don't need backups, but very few people could read them, and even less could 'etch' them.
I think the unprecedented decentralization and free flow of information of our time is far superior, even if the media we use is much less durable.
On the issue of formats he makes a very valid point tho. All we can do is support open formats and hope others follow our example so they gain momentum and become widespread and long lived.
Send your spendthrift head of state this
This is another case of only seeing part of the problem. Data preservation is easy. The problem is, we generate massive amounts of data. Data doesn't have an expiration date. It doesn't automatically categorize itself, know its own relevance, or volunteer itself for tasks. See, the vast majority of "data" floating around can be safely discarded. Do you really need an ethernet sniff log of everything you've done on the internet over the past ten years? The government might want a copy, but chances are pretty good its just as useless to them as you. How about those four (broken) copies of that mp3 you downloaded from Shareaza? Or outdated installers of software? Is there a reason to keep around those Netware 3.12 floppies (besides impressing other old farts)?
The problem isn't preserving data, it's knowing when to let it go. We have many, many, many methods of data preservation. We are drowning in information. The internet is generating petabytes worth of data every day, and only the smallest fraction of that really has any reuse value. And most of that, in six months, or a few years, probably not. What we need is better methods of sorting data, and ways to expire data safely.
Also, we also need control over our data. Corporations have been trying to take that away now for years. You don't need a copy of our software that can run on any computer, we're going to mung it up so it only runs on one computer, and if you have to reinstall the operating system or change the video card or anything else, that copy will cease to work. An irony, really -- because I know plenty of people that love playing old video games whose manufacturers long ago gave up on, but won't release the copyright for. Fifty years from now, I doubt a single copy of the game will still exist -- the concept, maybe. But it will have died and yet someone will still own the copyright and think money could be made off it. When we buy a chunk of data, we need to be able to control it, not just use it in some narrowly-defined way. Because otherwise, what's the point of data preservation in the first place? To stockpile more useless data that -- even worse, holding onto could be a liability to you?
#fuckbeta #iamslashdot #dicemustdie
I'd show you some examples, but they kinda fell to pieces sometime around 200BC. What we have left is the stuff that preserves well.
Red to red, black to black. Switch it on, but stand well back.
The finer the print, the more susceptible to wear, I would expect.
If your only tool is a hammer, every problem becomes a nail.
You can't put DRM on a rock.
Anybody want my mod points?
You mean like "all the nodes" that stored WebHostingTalk's data? "All the nodes" that were hosting Ma.gnolia bookmarks? "All the nodes" running Journalspace?
In the cloud, you can't tell who's a dog. I was stunned at every one of these events, which was totally preventable. But I read everything I could about each one, in order to be sure to avoid their mistakes.
You've got to ask the hard questions about how your data is being handled when you entrust it to somebody in the cloud.
The entire piece consists of:
1. Saw an Egyptian obelisk which had lasted for a long time.
2. Our modern data preservation methods aren't built for longevity.
3. Rocks have better data integrity than digital archives.
Thanks for the heads up. I'll be sure to keep that in mind when I'm deciding whether to save my memoirs on rock or .doc. Really helpful stuff.
Sorry, papyrus just doesn't scale that well.
on our hard drives. Porn. That will keep them scratching their heads for years.
"This primitive race seemed to be preoccupied with sex. So how did they fail to reproduce and let their race die out?"
Way back in the ancient times, only important stuff was carved into stone. Now everyone on our planet is squirreling away all kinds of useless crap on digital media.
Future alien archeologists will have a hell of a job sorting out the crap from the, well, stuff that is just a little less than crap.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
You can only write on so many walls before there aren't any walls left to write on. In fact, just because they wrote on walls didn't mean all data was preserved either. The fact that the we don't have an index of what was created means we can't tell how wide-spread a data loss would have been.
To extend on the thought, data preservation from ancient Egypt in modern times means you need an entire profession (archaeologists) just to dig up and translate these texts into a modern form. Its like a data recovery analyst, but much more specialized. You also need expensive structural engineers to make sure the buildings aren't going to fall apart. Then you need security guards to protect the walls from theft, desecration, accidental damage, etc...
I guess the while take-away is getting easier instead of harder. The poster may have cited some good examples of how works deteriorate over time, but not necessarily about the ability of restoring/reproducing/distributing such works.
After all, all that 'data' that was so useless for hundreds and hundreds of years was because we didn't have the ability to decode it.
Hmm. Perhaps we need to have a 'new rosetta stone' project that all programs and decoders have to submit to (for hardware and software.)
No! It's a *SIG*. Keep the Special Interest Groups away! (Con joke!)
I really want to be sure that I can find my shell scripts, so I've hired a team of down and out Egyptian peasants to carve them in to stone for me. My first script should be ready on 6 months.
Totally agree. Those supermarket plastic bags with a half-life of 50,000 years bare important messages for distant generations. And you know they're going to misinterpret the purpose of our junk and come up with weird theories to explain why we were so polite.
I imagine a data archival system of 100,000 monks scratching 1's and 0's into slate tablets.
Look at the opposite trend that went from stone to paper to electronic:
1) Data became more portable
2) Data became more easy to edit and change
3) You can store more data in the same size container
4) You can store more types of data, like sounds and moving pictures!
5) You can store more than data, you can store programs which do things that can create and transform their own data!
What can a 50 lb stone slab do?
Also, what is the quality of the content? How much quality information does it tell us? There's some good history there, but I don't see someone using a stone slab to store a human DNA sequence any time soon.
Nice try, but this is the most extreme hyperbole I've ever seen. I'm glad for the mental exercise but it was way too lightweight and way too easy to shoot holes in this.
"All great wisdom is contained in .signature files"
Digital information can be easily duplicated and transferred to other media. You can save the entire library of congress on hard disk, convert it to DVD, or print it out on paper. And all of it can be almost fully automated with near zero chance of error. Try doing a backup of your stone tablet library in a reasonable amount of time, labor, and accuracy. There is just no comparison.
nuclear war clearly states all this. The people that preceded the Egyptians got in a war with the middle east and they nuked each other - hense the deserts
Since all their technology was gone and they couldn't even make a pencil because they had grown so dependant on technology they had to go back to writing on rocks. And so will we!!!
Being creative is fun.
So what if it doesn't last as long. The vast majority isn't worth remembering anyway.
But seriously, lets say your historical rock recording is damaged - well it's not likely that there was a backup. But what about my photography collection? Oh, that's right, I have seven copies in four different locations, one of which is over 2000 miles from the others. And if I want another copy, safely tucked away on the other side of the world, it takes about a day to get it there. Let's see you do that with your rock carving.
No sig for you. YOU GET NO SIG!
Since when does plastic have a half-life?
Making fun of dumb people since 2009
Apparently, the least biodegradable substance is silicone. So our most enduring mark on the sands of eternity will be "breast enhancements".
Red to red, black to black. Switch it on, but stand well back.
I really thought there was going to be something special here, that the ancient Egyptians found some way to preserve data better we do now in modern society.
Does the author not realize that he's only looking at a rock that survived, and not one of the millions of rocks that turned to dust over the years?
If someone in 5,000 years finds a USB flash drive exhibit in some park with the data still readable off the device, that will not be proof that USB flash storage is the ultimate in storage technology, it'll only prove that that one USB flash drive lasted for 5,000 years.
I believe that we (us netizens right now) will be remembered as the root users of the internet, the place where the buck stops, the people alive at 0 A.D. if you will. Sort of like us surfing to web.archive.org to check out sites from 1995, only people in 2500 will be checking us out and tracking us down. Perhaps even celebrities may come of it, unknown in our time. In 20 years people will have all of today's internet on an USB stick (or obviously some other far more advanced replacement) and this whole data retention thing won't be an issue. We're here to stay, from now on we're switched on. Cheers.
There are a number of problems, some of which are obvious like longevity of media. A bigger problem is the hardware to access the media. And even bigger, is the format of the data.
OK, under proper conditions a properly made CD or DVD will last nearly forever. But it requires some really fancy optics and a solid state laser diode to read it. And a DVD requires a microprocessor - can't get away with some simple logic chips like you can with a CD. And the encoding of the data itself is complicated - probably too complicated to consider it for long term storage.
But the biggest problem is data format. People keep crowing about open formats, but that is nearly irrelevent. If I handed you a 10.5 inch tape reel with data from 1955 on it you couldn't read it without having (a) a tape reader mechanism, (b) a knowledge of BCD (the predecessor to EBCDIC) and (c) a knowledge of the format of the data being used. Having a 60-bit binary number spread across 10 characters is pretty useless unless you have some idea what that number represents. And there is the problem.
Open formats are fine, but they are too complicated for archival purposes. Things change, and the changes often make the very definitions of the data obscure. Today in Europe would a set of construction plans from 1920 make any sense at all? Probably not, because all the units have changed. We face a similar problem - only more so. In 100 years it is likely that a PDF document will be utterly unreadable because it uses ASCII to reference glyphs to be rendered on the screen with fonts. None of ASCII, ASCII fonts or anything else will exist any longer.
So it will not matter one tiny little bit that an "open format" was used. The material will still be unusable and unreadable without special conversions. Can open formats be more readily converted as format change? Possibly. I suppose if you have a lot of word processor documents from the Atari 800 today that you might find them difficult to convert, if not impossible. I would offer that even if they were done in an "open format" (like plain text with control words, like WordStar) you would stll find them unreadable and unusable - Atari didn't use ASCII.
And in 100 years the likelyhood that either ASCII or Unicode will survive is very remote.
The whole article is ridiculous. The first sentence is
My wife and I were in New York's Central Park last fall when we saw a nearly 4,000-year-old Egyptian obelisk that has been remarkably well preserved, with hieroglyphs that were clearly legible
What is remarkable about that? If you want to put a ancient Egypt rock in the Central Park, do you use a shattered obelisk where you can't read anything or do you take the nice one?
And how ignorant is the author to ignore all the broken, lost and otherwise destroyed rocks that didn't survive?
If you want to write an article about the lack of metadata standards and your perceived lack of long-term storage options, fine, but don't built it around your wifes spontaneous epiphanies.
In that order. Still working on how to write data on scissors..... Magnets!
What?
This is why I've been chiseling reddit headlines into the concrete in my driveway. And the neighbors call me crazy!!!
I hold very few opinions. I hold information based on observation and fact. If you wish to disagree, please use facts.
Wait, someone ACTUALLY had to write an article to tell me that scratching some letters into a rock will last longer than paper? Uh... You dont say?
THIS IS THE FUTURE!
Seriously though, this would be better than all hard drives, all optical discs and solid-state in the world.
Just make thick platters of plastic / rock / something else, then physically etch a shape into it.
At least, this is for backups, not rewritables.
The center of each disc could show some sort of diagram that shows data getting smaller and smaller, then have some sort of magnifying glass zooming into the smaller data.
And if something finds this in the future and fails to decipher it, then screw them, they don't deserve such knowledge at their current evolutionary level.
Brb, creating a business.
I can see it now. Graveyards with nothing left but pairs of implants buried in the ground.
no one can read the damn Hieroglyphs any more, so what does it matter that it lasted 4000 years?
Actually, I can read some Hieroglyphics. For example, the ones in the article's picture refer to something about "DVMCAIIXV takethdown notyce for CovpyriGt Infrryngemynt" or something like that.
At the bottom it is signed by the "RIVV".
How many rocks do I need to chisel to keep a copy of the current Wikipedia? Are there even enough mountains in Egypt for the top 10,000 articles?
I think we'll manage to keep enough of the important data by migrating to newer media over time. Besides, it's not like we have any better options.
Everything has a half life. It is merely the time it takes for half of a given sample of items to decay, or be destroyed. Granted, the term is mainly used when talking about radioactive isotopes, however it can be used in other realms.
When our name is on the back of your car, we're behind you all the way!
Since when does plastic have a half-life?
Man if it does, i wonder what product it will turn into?
The useful bit about this article is the recognition that the format is the key to any hope of understanding digital data over any period of time. Specifically
We can save bits forever, but if we don't know what they mean, then its all a waste. Standardized, open formats provide the only real hope of a Rosetta Stone" for data.
He and his wife should do more research. There are standards for Digital data management concepts, technologies and standards, such as the OAIS Reference Model for an Open Archival Information System (ISO 14721:2003). Maybe he's spending all his time looking at rocks.
Is a 4GB Bigfoot (5-1/4") Hard Disk the same as a stone obelisk?
Or should i mod my CD writer to write on stone? May need a little bigger PSU for that but upgrades are fun.....
Seriously tho, the older the better for long-term storage. Can apply this large scale or small...stone tablets or old disks.
Can anyone tell me what the conversion factor is from Libraries of Congress to Libraries of Alexandria?
We should take cue from battlestar and pass on the good parts of ourselves but leave out all the bad stuff. We should be able to fit that on a rock.
Vellum (animal skin) lasts a lot longer than Papyrus ever did (1,000 years vs a few hundred tops). Papyrus only subsisted in very dry climates. Most stuff that's found in archeology are fired clay tablets and pottery. Things seem to survive in caves pretty well, too.
Do you realize how many frames I'll have to carve in stone for just one DVD, much less my whole collection? After that, there are thousands of pics!!!
Can you say: "Holy Bleeding Blisters, Batman!"
Time to move near a stone quarry. I guess...offsite backups go to the moon, Alice...to the moon!
Down With Slashdot BETA!!! I've been around the corner and seen the oliphant; you can only abuse me from your perspecti
Back in 1999, the New York Times conducted a competition to design a time capsule that would last 1,000 years. The winning design, by Santiago Calatrava, used a technology called HD-Rosetta. Links to relevant articles are here: http://www.amnh.org/exhibitions/timescapsule/ and here: http://www.norsam.com/hdrosetta.htm
why do we need to have records stored on media that lasts a long time when we can so easily just copy them to new media and have as many redundant copies as we want since data storage is so much cheaper now?
Even after all these years I could still read "My name is Ozymandius, King of Kings", but I looked at Big Blue, and a Wang, and I despaired.
DRM'd data will be useless to future civilizations. Our "legacy" will die with it. JUST SAY NO for the good of mankind's future preservation!
I hope someone involved with DRM reads this and actually cares...
Fact: Everything I say is fiction.
The way our world is going, we may all find ourselves living in caves sooner than later. Then we can just write our stories on the cave walls and presto, long term data preservation!
That wasn't so hard now, was it?
"Suppose you were an idiot...and suppose you were a member of Congress...but I repeat myself." Mark Twain
Not anonymous; it was said by the psychologist Abraham Maslow.
And the Egyptians didn't have the signal-to-noise ratio, with images of cats saying "I CAN HAZ BASTET?" Or Horus' Tweeting "OMG My brother Set is such a dick!"
Luckily scientology has enough money to inscribe their beliefs on stainless steel records and the foresight to place these in titanium boxes in underground vaults, Find me on google earth
look what they came up with. "When the crocodiles feed on the hearts rejected upon the scales of Horus, blah de blah de blah...."
Don't confuse taking the time to chisel something with profundity. Even older than the pyramids are animals scraped on rocks.
Agreed. A small group of us are just beginning a project called the Digital Poetry Archive of Canada, dedicated to saving analog recordings of poetry that are fading fast as well as obsolescing digital files, and our first order of business is to figure out an enduring archival format.
Rest assured it won't be stone! But it looks to me like we're going to have to take the approach that whatever format we choose is temporary, and instead establish protocol for regular translation into newer formats.
Still, we have to aim for longevity. Standards are a boon here, but are pitted against popularity. It's easy to assume for text: the .txt files will be a good gamble, for instance, though in the case of poetry the visual formatting can be crucial, so hopefully a comprehensive presentation format like pdf (or an alternative! suggestions?) will have some endurance.
Recordings are another matter. Video codecs are in intense flux, and H264 looks like it might hit problems in 10 years. DV is perhaps worse, though it is less lossy. Audio codecs too (spare me the OGGvangelism for the moment), even though AIFF has both popularity and licence-free going for it.
Any suggestions would be welcome!
Damn those pesky terrorists
Titanium plates!
Your information on the Internet will never go away unless the server dies. In which case there is probably a backup somewhere if you said anything incriminating.
God spoke to me.
Not everything has a half-life. Take entropy as an example. :-)
Most posts miss the point for failing to appreciate the incredible difficulty not only in finding a medium that survives long-term, but in finding one that will be readable beyond even one generation. Like it or not, the best "high density" medium discovered to date which has at least proven itself to last a few millenium is ink-on-paper (or parchment, etc.) I'm involved in the memorabilia market, where there are constant challenges regarding validation of authenticity. Last week a customer told me that someone had been in to try to sell him on the idea of embedding a computer chip (akin to an RFID tag, as I understood it) in with encased memorabilia to provide the ultimate in verification without intrusive efforts. I immediately laughed and said, "That will be completely worthless".....I proceeded to explain that the longest surviving electronic storage of ANY FORMAT TO DATE is probably about a tie between the cassette tape, or perhaps the "78" vinyl record; although the vinyl still survives, finding a player is getting pretty much impossible. Most tape memory, disk drives, magnetic media, etc. have inherent failure modes which almost ensure only a few decades of lifetime. How many of you "kids" of even 35 years old are still able to read the 8" floppies you were using in the 80's? Absolutely nobody seems to appreciate this dilemma. I read of these big arguments about universal formats and all the anti-M-Soft rhetoric and just laugh -- in probably 50 years, and most certainly 100 years, the word Microsoft will sound just like the word EDSEL sounds today....it won't matter whether today's storage format is ASCII or proprietary, there won't be a machine in the universe to read any of it. The only records which will survive into the next millenium are those which are important enough to undergo the ongoing expense of translating through the ever-changing media types, such as property records at the courthouse. The average individual, using e-mail and texting, will preserve essentially NOTHING for follow-on generations. Family histories are basically doomed, as the last generation to use formats with any chance of surviving even a few centuries are your grandparents - they actually wrote letters and printed pictures, and stored them in cedar chests. Assuming the bugs, light, and moisture keep away, your grandparents history might just last a few centuries. It's a damn shame really, and yet even though I'm completely aware of the situation, I have no real idea how to combat the problem effectively.
Mankind != important in the grand scheme of things, thus nothing we produce is all that important either.
How do you know that? As far as we know we might be the only intelligent life in the entire universe (the probabilistic arguments suggesting this is unlikely have huge unknowns) i.e. the only way for the Universe to understand itself. Until we know otherwise perhaps we should not write ourselves off. Besides, even if there is other intelligent life out there, who knows how far we'll have spread a few billion years time.
The biggest problem I foresee to long term data preservation is not the media so much as the users.
Many cultures have had their history destroyed by missionaries, immigrants, and other groups that found their written language to be politicly, religiously, or culturally incorrect at the time.
Even cave drawings have to be protected from those who want to feel powerful or significant by destroying them or covering them with their own graffiti tags.
Perhaps the best answer would be in the storage location, like the moon or some other hard to reach place. The assumption being "If you're tall enough to reach this book, you're less likely to eat the cover!"
The one thing these stories have in common is that they vastly underestimate past civilizations and the difficulty archeologists and historians have in piecing it together.
As Dawkins pointed out with his meme idea, ideas evolve, first they were handed down through speech and gesture, then theatre and drawings, then writing and math, then print, then film, radio, TV... That's where it stood when I was a kid in the 60's. Since then there is so much of the stuff and it's so easy to generate bits to make more stuff that we now simply call it "the net".
Just as past civilizations and tribes have left an excellent dating record in the fashion and technology of their broken pots, we will leave a rich record in our massive landfills and floating islands of plastic. Wether there is anyone around to dig it up in a century or two is anyones guess, but the data isn't looking good right now.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
Put together some kind of computer-drive laser etching device that can imprint onto rock or hardened steel, at a fairly high bit-rate, and etch the important stuff onto something that's not going to degrade for a thousand years. Plus, anybody with a microscope and a pen & paper can read back the data and transcode it into any future format that might evolve. The Rosetta Stone of the future.
Poor means hoping the toothache goes away.
Well I think this AC on said it better and a whole day before this article.. :-) And that comment was marked funny, which is what this is as well.
When I started with computers I had storage measured in kilobytes, now I have storage measured in terabytes. That 10^9 difference means that even if I lose 99% I preserve millions of times so much information. IT's really quite simple, if we acn't easily keep this much information it's not worth keeping this much information.
Imagine that we decided WWIII is a good thing, nuked off the CPU factories, HDD factories and so on. You still have your computer, an UPS generator and some fuel, printers and enough ink so that you can use maybe 5-20 years grabbing what you can but there will be no replacement parts and the factories won't be rebuilt. Would we lose information? Sure. Would it be important compared to the cost and effort of preserving every crap tune someone made? No.
A society thrives on our ability to record, compile, store, find and disseminate information. Stone tablets solves at most one of those problems. Recording them is hell, compiling them into useful information with crossreferences is hell, finding the right tablet is hell and sending stone tablets to whoever needs them is hell. A great hall of stone tablets would be infinitely less useful to me than internet, google and wikipedia.
Preservation is the kind of disaster recovery where we might as well ask "IF this disaster occurs, how much could we preserve by dumping everything we have online into durable media", not try some absurd scenario where we'd just let our data collapse and disappear. If not even the computers in our deepest vaults survive, we got much bigger issues anyway.
Live today, because you never know what tomorrow brings
"Stone also leaves a lot to be desired as a useful format. Transmission can be problematic."
Ha! - Didn't you ever hear of a catapult?
The problem I see with this analysis is that it is flawed in the sense that we only see the one obelisk that did survive. There are probably 100,000 obelisks that didn't survive. So what kind of retention rate would that be?
Pretty soon, we'll be archiving so much information that it'll be easier just to create a multidimensional backup of the entire universe.
Course, that'll take a spitload of DVD's and a lot of RAM. Hmmm. Wonder what God uses for backups?
"My country, right or wrong; if right, to be kept right; and if wrong, to be set right." --Senator Carl Schurz (1872)
Don't forget that the data has to be encoded/decoded to retrieve its information. Until the decoding of the rosetta stone, the rocks were just gibberish to us. And, yes, there is a difference between data and information. Obviously most data we see contains very little information, including this article.
...is a collection of oral histories from Israel and the areas surrounding it, modified to promote the idea of one true God. The oral histories are a collection of both fictional morality tales and actual facts, and is therefore one of the best guides to what exactly was going on during the transition from pre-history.
That said, I am an atheist. I just don't discount the information that has been passed down for thousands of years as a bunch of hogwash, especially when much of it has in fact been proven true by following the text and discovering ancient ruined cities.
The half life of a concept is the amount of time it takes for half the people taught about it in school to forget about it.
So the half life of entropy is about 45 minutes.
You keep using that word; I do not think it means what you think it means.
If you can remember about ten years ago, CDs appeared (especially from Kodak) with an expected storage life of several hundred years. The longevity (or lack thereof) was a function of the dyes used in the layer actually burned by laser. In theory, gold dye meant long life, green dye meant two or three years tops. This bit of folklore was followed up by greenish-gold dyes and a shift to DVD-R discs with over 4 gigabytes of storage and an unknown (or undocumented) shelf life. The holy grail of long term storage, the way it should be done using technology, was a throwaway moment in Forbidden Planet, i.e., Krell theremin tunes embedded in small crystals which had survived over 700,000 years.
However, there is another longterm information storage model, biological in origin: DNA embedded in diverse ecosystems which brook no deviation from competitive norms, but which in fact do drift over millions of years. So, what you need is a self-replicating, self-correcting mechanism for information transfer which is amenable to criticism by peers which are only indirectly related to the data in temporal transit toward unknown futures.
Oral tradition, anyone?
``Tension, apprehension & dissension have begun!'' - Duffy Wyg&, in Alfred Bester's _The Demolished Man_
Who cares whether future generations will be able to read our data. We generate and store data in order to advance our lives NOW, not the career of some archaeologist 3000 years in the future. It's THEIR problem if they can't read the data, nor ours.
Since the nuclear payload causing monstrous mutations was unleashed upon the jungle by a rocket gone wild!!!!!!!!!!!!!!!ONE
-josh
Rocks DO need backing up. Scribes in Sumer maintained traditions over thousands of years of recopying clay tablets to preserve them. Even the ancient Persian conquerors of Babylon constructed museums of the already very ancient objects they found there. The same was true of later scribal traditions on leather and parchment which preserved classical documents for us, and the ways of reading them. In fact, if it weren't for the far superior concern for posterity the middle ages showed, we would not have the smattering of knowledge about the classical world we have managed to hold on to.
I am a linguist who studies clay tablets and ancient writing systems, and let me tell you, I lose sleep over this problem every day. What will happen (and note that I don't say would, because it is inconceivable that the "cloud" will last a thousand years, let alone five thousand) when they don't know what kind of electricity we used? Where will the remains of our civilization be? There is a basic point here which the "wayback machine" doesn't go far enough to answer. Where will they find our information stored, and how will they ever, ever, devise a way to read it? Bear in mind that we have trouble deciphering the earliest and most primitive writing systems ever devised even now. There are still dozens of these we can't read, and many more we haven't even rediscovered yet.
And, it turns out, a lot of what has happened to survive for us to read from all that time ago really is about as exciting as server logs - receipts for tithes, buying and selling grain, etc. And those tell us so many surprising and extraordinarily valuable things about the way the people who produced them lived, which the documents they intentionally preserved (such as king lists, prayers, mythologies) would never have thought to mention. So don't underestimate the value of the information you think is worthless! A thousand years from now they will regard you as a deluded primitive, but they will be interested in your internet traffic and your credit card records. But of course, don't forget to preserve the art too.
The significance of a half-life for plastic is mostly negligible. Most plastics will biodegrade if left in an uncontrolled environment in (I'd guess) well under 2,000 years.
I realize you were being somewhat glib, but...
Now, ceramics are a different matter entirely.
Probably the best post in the thread, IMO.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
And yet not one mention of iron mountain and their vast collection of data, mounds and mounds of boxes of paper data to be sanitized, then approved for digitization and distribution.
http://web.archive.org/web/19961220154856/http://www.aol.com/
http://www.rosettaproject.org/disk/concept/
Why, yes I have been touched by His noodly appendage. And I plan to sue.
We know quite well how paper and ink age through the years.
We know quite well how to preserve paper for a long time.
We know how to efficiently encode digital information onto paper, using bidimensional barcodes.
We know that paper tolerates a big amount of damage without losing the information encoded on it.
We have ample experience recovering information from damaged -even charred- paper.
We know paper requires no energy to maintain the information stored on it.
If we combine these factors, it's not hard to conceive a long-term storage facility for digital data (could be encrypted - they're just dots on paper).
For data retrieval scan the pages, decode the dots, decrypt the bytes.
If we grow plants to make paper for backup, the plants wil sequester CO2 while growing to a profitable size.