Slashdot Mirror
Nanotech Memory Could Hold Data For 1 Billion Years
Hugh Pickens writes "Digital storage devices have become ubiquitous in our lives but the move to digital storage has raised concerns about the lifetime of the storage media. Now Alex Zettl and his group at the University of California, Berkeley report that they have developed an experimental memory device consisting of a crystalline iron nanoparticle enclosed in a multiwalled carbon nanotube that could have a storage capacity as high as 1 terabyte per square inch and temperature-stability in excess of one billion years. The nanoparticle can be moved through the nanotube by applying a low voltage, writing the device to a binary state represented by the position of the nanoparticle. The state of the device can then be subsequently read by a simple resistance measurement while reversing the nanoparticle's motion allows a memory 'bit' to be rewritten. This creates a programmable memory system that, like a silicon chip, can record digital information and play it back using conventional computer hardware storing data at a high density with a very long lifetime. Details of the process are available at the American Chemical Society for $30."
If you don't misplace it..
That's great. Will the readers and systems able to display such information be around for even a hundred? Will they even accept the same power?
The main problem isn't the length of time that data can be stored. Hard drives and tape drives still carry data from the 1970s, but no one can use them. Why? Because of format changes. We recently transitioned to Blu-Ray, and there are countless codecs for video at this point in time. I don't think the problem is with the length of time for storage, as useful as that is, but rather with the format in which we store them.
An excellent anecdote was mentioned on slashdot recently: http://science.slashdot.org/article.pl?sid=09/04/13/005224
Wow, what a claim. And by the time someone figures out it's bullshit, the guy who made it will be dust long ago.
BRILLIANT!
WTF? Over?
The problem with CD-Rs, DVD-Rs, tapes, and so on is that they have extremely short lifetimes (6 to 3 years for most optical media, 10-20 years for most magnetic media).
This is a solution that would finally allow our civilization's information to last beyond the apocalypse occurring in 2012.
Or think think how long Atlantis was lost to intelligent life...
Ok, while I find the tech cool and this is certainly News for Nerds, Stuff that Matters but, seriously? "Details of the process are available at the American Chemical Society for $30." Seriously? We're just abandoning any pretense that these are news summaries now and just outright turning them into ads for products? We're now outright trying to sell things? Weak. Very weak indeed.
nobody will give a damn about our data anyway.
I knew this had the ring of truth about it
http://www.scribd.com/doc/13855395/Weaseljumper-Read-Me-First/
If Slashdot were chemistry it would look like this:Cadaverine
Zettl is a pretty well known figure in this field. He's not throwing around the term because it's a buzzword.
Learn the difference.
Sheesh.
"Details of the process are available at the American Chemical Society for $30."
Does anyone else find the trend of pay-per-view science disturbing?
All too often, if you search the internet for a topic with ongoing research, you may likely find links to papers with restricted access and not generally accessible.
Any you should assume that several patents are pending based on this ongoing research, even if the idea is a seemingly obvious application of the research.
In software, it is worse. Papers are rarely written, as there are rarely any new ideas. Most all software companies reinvent the same wheels, then attempt to patent cosmetic qualities of the wheeels. Then other companies apply effort to avoid use of such cosmetic patents. and create their own similar cosmetic features (and patents).
Nanotech - 1 Billion years
Elephant - Forever
Technology simply cannot compete with mother nature.
Summation 2
This just in, people use buzzwords to sound smart, get funding. I mean seriously, how else are we going to syngerize our companies to their maximum efficiency? It isn't all about the low hanging fruit, you know.
"A billion years ought to be enough for anybody." - Me
"Study your math, kids. Key to the universe." -The Archangel Gabriel
I guess the question is, is the data of today's living really that important? I mean, sure, you might wish you had every bit of minute info from the builders of the pyramids, but, does it really undermine our life to not have it? Indeed, can the imagination and argument required to envision how the past was actually make the past more relevant to us today?
I almost wonder if, instead of having data that lasts forever, if we should have data that deletes itself when you die.
This is my sig.
Do I get a refund if the memory fails before a billion years?
It reminds me of the word "ubiquitous". Prior to 1997 or so no one had ever heard of this word, much less used it in a computer/business setting. Now I see even my boss, someone who does not come from an IT background, using it.
No offense, but is English not your first language? Because that word has been in use for nearly 200 years, and therefore was not originally IT-specific.
Get off my launchpad!
No, Mork from Ork was using it in the mid 70's ("Nano Nano").
Table-ized A.I.
Lots of work? Normal weathering will destroy stone carvings, and many ancient carvings are either lost completely or so faded as to be unreadable simply because they were left out in the weather for a few thousand years. The well-preserved ones are the ones that were kept in big vaults like the pyramids and protected from the weather. Also, lots and lots of stone carvings have been deliberately destroyed throughout history for various reasons, including times when invading armies tried to destroy the relics of cultures they were attempting to subdue.
What kind of skill is required to see a billion years into the future?
Umm... how about the skill of science?
Okay, to be fair, the summary exaggerates the claim from the scientific paper quite a bit. The summary implies that they are claiming to have built a device that will last for a billion years. Not so. They are claiming that the individual bits should be stable to random thermal flipping over that timescale. Whether or not a device can be built around those bits that also last a billion years is another question. In the words of the authors:
Again, they are not claiming that they have built a device that will last a billion years. But they are saying that they have at least achieved the first step for archival storage. If you want a device that will last for, say, a thousand years, then having bits that persist over at least that long is required. Of course, there are gotchas:
-A real device may have other weak points that degrade first.
-The analysis only considers some dangers of long-term storage. E.g. electric or magnetic fields could cause the bits to flip. Elevated temperatures would reduce the stability time.
-Many memory devices would in principle be stable over very long timescales if analyzed similarly. E.g. for a normal hard drive, at room temperature without any electric or magnetic fields, the actual magnetic domain orientation is also stable over very long times.
Point being, the authors of the paper are correct in what they wrote (it's not hard to calculate the kinds of things they were considering, even over timescales of billions of years), but as they point out that's not the whole story for a real device.
You know I am sick of people saying the Egyptians had us beat. Sure the data was there but they didn't leave any way to read the data. A lucky find hundreds of miles away called the rosetta stone is what allowed us to crack their encryptian.
Format matters little if you don't leave a method of retrivial. I have tons of programs written back in the early 80's. However since they are all for a TI 99/4a on 5 1/2" floppies I can't use them anymore.
i thought once I was found, but it was only a dream.
Actually Bell Telephone popularized the term "ubiquitous" in a series of nationwide print advertisements in the mid-1960's.
> In another 50-100 years we ourselves may not be able to read
> those audio messages we sent out to space on those golden records.
Why not? You should be able to play them back with any sharp metal wire poked through a sheet of paper or plastic. Put the disk on a turntable, poke a wire through a thin piece of plastic or paper, lay the point of the wire in the groove while holding the sheet. When you turn the disk, the needle will vibrate the sheet and you'll hear the sounds.
Of course, you'll probably scratch the shit out of it.
Kids: Don't try it with your Dad's mint condition LPs (the black disks sealed in the plastic sleeves that he only handles while wearing gloves). Ask your grandmother if your aunt's 45s are still in her attic ;-)
As long as we can still figure out that the squiggly grooves are sounds, we should be able to play them fine. Hell, if there's no wire you might be able to get it working with an oak splinter.
I think there is prior art on this one:
-- Ed Avis ed@membled.com
Just the single bit isn't stable for a billion years. It's merely theorhetically stable from a single influence for a calculated billion years. That's pure bullshit.
If you're saying that there is always room for us to discover new effects and revise our calculations, then I agree. But if you're saying that we cannot make any kind of predictions, with useful error bars, about events over long timescale, then I have to disagree.
Rocks used to be stable -- until general weathering was observed. And it wasn't observed on the first day.
That's a good example. Apparently you accept the general theories of erosion and weathering, even though we have not measured them over the timescales we think they operate. It wasn't observed on the first day, but we also have not watched a mountain for 100,000 years... and yet we accept explanations and predictions that invoke those timescales. Similarly for plate tectonics, star formation, radioactivity, chemical stability, and so on. The long timescales certainly have effects on our predictions (e.g. error bars, predicting details, etc.), but we can still make statistically-significant predictions.
Tell me, what _does_ affect nano-scale devices? The answer is that no ones been looking for very long.
Okay, the last decades or century of science is nowhere near the billions of years timescale. But we can still make sensible predictions. We know what forces are operative on nano-scales (quantum mechanics is quite well-established). If our theories didn't account for some really-longterm effect, then we would expect to see measurable deviations from our predictions in the composition of the universe, decay rates, chemical stability, or something else. Are you suggesting that an N2 molecule isn't stable over billions of years? Are you saying that there is some as-yet-undiscovered process that causes it to break-down over a timescale of billions of years? If so, where's the evidence? On the other hand, if you accept our theories can make sensible predictions for some nano-objects, why can they not make sensible predictions for other nano-objects?
I promiss you that within a billion years, some effect, some dynamic, some event will break the device.
I am certainly willing to accept that some future scientific discovery will modify our current best theories. But absent such evidence, why should we not trust a spectacularly successful physical model? Just because the timescales are long? (The timescales of plate tectonics, star formation, cosmology, etc. are long, and yet we trust our theories because they work...) Your declaration that "something" will happen, without any particular evidence, isn't scientific and isn't convincing.
Sure, you could play it like that. But will you be able to match the right speed? How about the sound volume? And like you said - scratching problem.
We MAY not be able to read those messages.
Most people WILL not be able to read them pretty soon due to obscurity.
As you've implied - many kids today don't know they can play a record without electricity.
Heck, a dedicated tinkerer could relatively easily make a magnetic tape player from scratch.
Not so likely with CDs. Nearly impossible with DVDs.
The point of the post was that the recording mediums often become unreadable through becoming obsolete BUT that the data recorded may well be readable for a much longer time.
Attaching instructions how to read it to the device (as they did with Voyager disks) that should be readable in the distant future is a matter of adding 2 and 2.
Mit der Dummheit kämpfen Götter selbst vergebens
Build nano-elephants.
That way we will be combining nano-technology and nature and we will have a device that stores data for billion forevers.
Mit der Dummheit kämpfen Götter selbst vergebens
A little while ago there was an article on Slashdot by someone who wrote some software that played LPs using a flatbed scanner. The resolution on a cheap consumer-grade scanner is high enough that the sound is recognisable. You wouldn't want to use it for music, but to get a rough idea it's fine, and this is using hardware that a lot of people have sitting around at home. Specialist firms will use a laser to read the disks and will copy them for you - for a much larger fee.
I am TheRaven on Soylent News
Would someone explain to a simpleton why there isn't a multiple redundent opensource backup system for using recordable media?
My understanding of the failure mode of CD's and DVD's is that they go bad a bit or a sector at a time, not all at once everywhere.
Given that, it should be possible to design a multiply redundent data scheme to make recovery possible.
My prof told a story of one of the early vacuum tube computers. A 'bit' was stored in a shoe box sized module that had 7 tubes in it. ANY 5 of the tubes could die and the device would still work. (Everything was drawer mounted Grad students with shopping carts full of vacuum tubes would run through the halls replacing tubes on the fly)
E.g. Reed Solomon codes can make it possible to correct an X bit error, and detect a Y bit error in a block. I don't know if these are the best encoding scheme for redundency.
Record onto two disks, both with RS error detection. If you know which block is bad, you have good odds to recover the corresponding block form the other disk.
Use a format so that losing one sector only loses you one sector of a file, not the entire file/directory/disk. This requires additional redundency for meta data, and it will mean don't use compression or encryption schemes that require the previous block to be read correctly to read the current block.
If someone does this I suggest a tier of standards, based on the desired probability of full recovery, and the probability of a read error on the media.
So for example Level 1 is based on making 2 copies + RS codes + redundant metadata on a single disk.
Level 2 is based on making at least 2 copies, and enough recovery code so that ANY one scratch across the face of the disk can't delete all the data.
Level 3 is based on making two copies of the disk, and labeling appropriately.
Level 4 is based on making 3 copies of the disk, with one labeled for cold storage.
It may be to do this, you will need to modify the media writer to access arbitrary locations on the media -- e.g. if the chunk of the disk that says what the disk is is bad.
So why isn't this done?
Or is it, and I've just not heard of it.
Third Career: Tree Farmer Second Career: Computer Geek First Career: Teacher, Outdoor Instructor, Photographer.