backtop


Print 112 comment(s) - last by Etsp.. on Dec 1 at 5:37 PM


Abideen claims to have placed a 45 second video clip on a single sheet of paper, with the possibility of up to 450GB on the horizon - Image courtesy Arab News
Rainbow technology still in the works but holds promise

According to a report from the Arab News, a university technology student named Sainul Abideen has invented a method of storing massive amounts of digital data on a plain piece of paper that he claims could store many times the capacity of the best Blu-ray or HD-DVD discs. In fact, Abideen says that his Rainbow technology can enable him to store from 90 to 450GB on a piece of paper. As far as a real life demonstration of a 450GB paper goes, the technology still needs development.

Abideen, who hails from the Kerala, India, claims that that his Rainbow system is better than a binary storage because instead of using ones and zeros to represent data, Abideen uses geometric shapes such as squares and hexagons to represent data patterns. Color is also used in the system to represent other data elements. According to Abideen, all that's required to read the Rainbow prints is a scanner and specialized software.

The reporter at Arab News claims to have seen 450 pages of fully printed foolscap being stored on a 4-square inch piece of Rainbow paper. The reporter also claimed that he was shown a 45-second video clip that was stored using the Rainbow system on a plain piece of paper. Interestingly, 45-seconds of video isn't a lot, and if the Rainbow system can store up to 450GB, then we need to be watching full length high-definition videos from a piece of paper.

One of the major advantages of the Rainbow system is the fact that it should cost a lot less to produce than typical polycarbonate DVD and CD discs. Abideen claims that huge databanks can be constructed out of Rainbow-based storage mediums. Although the main attraction is cheap paper right now, other media can use the Rainbow system too.

As of right now, Abideen's system is still under research at the Muslim Educational Society Engineering College and although no major companies have expressed interest, Abideen is confident of the system's future. According to the report, Aibdeen is hard at work at developing a Rainbow scanner that would be small enough for integration into notebook computers. If developed, a Rainbow printer will likely be next up.

In other high-capacity storage news, DailyTech previously reported that Hitachi-Maxell is in the progress of producing holographic media for shipment this year. Holographic storage is one of the biggest forward-looking storage technologies and holds a great deal of promise -- as well as data.


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Unpractial
By stevenplatt on 11/24/2006 3:57:42 PM , Rating: 2
This sounds like great technology. But the one thing that i find interesting is the fact that it IS printed on paper. Paper isnt exactly a practical storage medium. If colors are to be used to store information then with paper colors would fade over time and lead to data corruptions. Also paper isnt a sturdy medium at all. I didnt get exactly how important the paper was and if this technology could be applied to plastics like a memory card. Also i do not understand why shapes are more effective than binary. Unless 1 shape equates to more than one byte of data. maybe this is where the current develpement is, and why we are not watching high def movies from paper.




RE: Unpractial
By AlexWade on 11/24/2006 4:05:34 PM , Rating: 2
Supposing we all converted to paper storage. All we would need is a fire, and we are screwed big time. I realize, of course, that fire can damage other storage mediums. But with those, there is a chance of recovery.


RE: Unpractial
By Griswold on 11/25/2006 5:05:24 AM , Rating: 3
You mean like in centuries gone by, where all knowledge was stored on paper? Seems like we were able to preserve some of it until today. ;)

By the way, you can still recover information from a burned piece of paper - as long as no water was poured over it (which usually happen when you try to put a fire out).


RE: Unpractial
By ninjit on 11/24/2006 4:08:59 PM , Rating: 2
The use of shapes and colors means more symbols allowing you to store more data.

Think of it in terms of a binary digit string vs a hexadecimal string:

a 32-bit binary string is 32 characters long, but the same information in hexadecimal is only 8 characters long.

The problem with more symbols is that it could lead to ambiguity - e.g. an octagon could look like a circle or a hexgon, etc. leaving a lot more room for errors in interpretation.
Binary data has the least ambiguity: it's either there or it 's not


RE: Unpractial
By Russell on 11/24/2006 4:13:36 PM , Rating: 2
Exactly.

I think the ambiguity issue isn't a big one so long as the scanners and printers use a high enough resolution to scan or print them without confusion.

What concerns me is what happens to the data if the page is folded. That could screw up everything along the fold.


RE: Unpractial
By ghost101 on 11/24/2006 4:17:35 PM , Rating: 2
I was just wondering about the scanning issue. Wouldnt it be exactly the same as using binary?


RE: Unpractial
By ghost101 on 11/24/2006 4:16:02 PM , Rating: 2
I agree, nothing is as simply and clear cut as binary. As for people complaining that paper can burn and fade. Well, if printed and stored properly im sure it will last as long as for what it has been designated for. As for burning, i dont think CDs will survive a fire.

I dont think this will catch on though as shown by the lack of corporate interest.

Also, whats this errorsafe popup i always get on dailytech but nowhere else?


RE: Unpractial
By KristopherKubicki (blog) on 11/24/2006 4:31:38 PM , Rating: 2
quote:

Also, whats this errorsafe popup i always get on dailytech but nowhere else?

We have a rogue ad we're trying to track down and get rid of.

Kristopher


RE: Unpractial
By saratoga on 11/26/2006 3:36:53 PM , Rating: 2
quote:
The use of shapes and colors means more symbols allowing you to store more data.


Colors yes, shapes no. Shapes are just groups of pixels. A shape that is made of 10 pixels contains no more information then the 10 pixels that compose it. Colors do help though since they allow each pixel to have more information.

quote:
Think of it in terms of a binary digit string vs a hexadecimal string:

a 32-bit binary string is 32 characters long, but the same information in hexadecimal is only 8 characters long.


Yes but the characters are now more complex. Encoding the 32 binary characters can be done using just 1 32 bit color pixel, or 32 black/white pixels. Encoding each of the 8 hex characters requires 4 black white pixels. Either way the maximum information capacity is the same.



RE: Unpractial
By originaldosa on 11/24/2006 6:18:33 PM , Rating: 2
I would be a waste of space but if this technology was some how used for future Blue-ray or HD-DVD rentals or other commercial rentals it would could prevent copyright issues. The paper would eventually become corrupt and unplayable resulting in purchasing Blu-ray or HD-DVD movies.


RE: Unpractial
By leidegre on 11/25/2006 7:46:31 AM , Rating: 3
Paper is just a type of print, this technology is something i read about before and it's bascially about geometry.

The principle of this technique is to store information not in a planer binary map, but in a multi-dimensional space, not just 3D, any kind of dimensional space you can com up with. Shapes and colurs allow such dimensions to exist, but this kind of gemetric storage could reside within platic as well, and even if the print in this case is written on a paper, that is still not representing a depth. The medium could be anything really.

It's always intresting with new stuff, just have to wait and see what happens.


RE: Unpractial
By ElJefe69 on 11/27/2006 2:03:25 PM , Rating: 2
You are the only person who is not a complete and utter ignorant moron of enormous proportions.

clearly, this inventor has an enormous understanding and also design that no one on this forum even has 1% of.

It took someone like 20 posts down to begin to discuss hex and all that. See, when someone INVENTS something, that means it has not been around yet for you consuming morons of miniscule tidbits of knowledge.


RE: Unpractial
By audiophi1e on 11/30/2006 12:15:34 PM , Rating: 2
I agree.

This is a novel application to storing data. Whereas bits or even hex are values stored in linear fashion, using a plane (2 dimensional storage) adds spacial data that has no way of being emulated by simple hex.

You can specify in binary the existence of 10 dots by having 1111111111 as your data value, but how do you specify WHERE in the plane of that media (paper in this case) they are, and HOW they are spacially arranged? They can be in a circle, square, triangle...and then how large is the shape? These are parameters that add nearly incalculable permutations of data.

Conventional data is only 1 dimensional--stored in linear array. This is 2 dimensional data. If he were to perfect his compression algorithm , I'd wager good money that 450GB IS MERELY THE BEGINNING.


blah
By AppaYipYip on 11/24/2006 6:56:41 PM , Rating: 2
my USB flash drive is far greater than his peice of paper.

A.) can be used almost anywhere
B.) waterproof, weatherproof
C.) does not requier a scanner or special software
D.) can store a multitude of mediums or digital information




RE: blah
By Suomynona on 11/24/2006 7:48:02 PM , Rating: 2
Feasibility test:
I have just scanned an A4 size image into lossless 24-bit BMP at 300 dpi. It produced a 26 MB file. Doubling the resolution makes it over 100 MB, still far away from the GB range. This invention is flawed at its root:
- I do not see how you could increase the capacity by using shapes like circles and XYgons. Color dots are your best bet.
- you can NOT practically increase capacity by using greater color depth. To reliably retrieve more info stored in color, one would require ALL printers and scanners with unprecedented calibration, e.g. the shade of orange printed on ALL printers MUST be read on ALL scanners as the exact same 24-bit color.
- you can NOT practically increase resolution. Depending on the placement of paper, the info could be read back in different ways, for example.

- on the other hand, you COULD add redundant recovery information, like in RAR, which decreases capacity but increases readability in case of bad media.


All in all, we will either never hear of this "invention" again (as he will not be able to produce working prototypes), or we will, but in a scandal...

Macro-world (like printed stuff) will never reach the "information stored per square inch" ratio of the micro-world (a bunch of electrons in a memory cell).


RE: blah
By mpb on 11/24/2006 10:22:10 PM , Rating: 4
Assuming you have a REALLY high quality printer that can print at 4800 x 2400 dpi, and you're doing a full bleed print onto 8.5 x 11 paper, there's the potential to have over 1 billion dots on that piece of paper.

Encoding 1 gigabits though, even under ideal situation of a printer that maintains that print resolution, and a scanner that can perfectly read it, you'd need either some sort of laser alignment, or you'd need to have "stop" bits, perhaps only using every other "bit" on the paper so that an encoded "11111" wouldn't just be an indistinguishable line.

Using shapes to represent various bit patterns, like saying that a square would be 1011 and a circle is 1100 and triangle is 0011, etc. would be one way to encode data into unique identifiable patters, but then creating those shapes requires probably at least an 8x8 dot area, so you've just reduced the number of "patterns" per page from 1 billion to 16.8 million (each patters being 64 dots).

If you were able to come up with 16 shapes that could be uniquely identified, without misidentifying problems like mistaking a hexagon for a circle, then you can use each shape to represent a 4 bit number... hooray. However I'm not sure how many unique shapes you can squeeze into an 8x8 matrix before it gets harder to distinguish...

So each pattern can represent 4 bits, putting you at about 64 megabits on the paper.

Using colors, let's say an 8 color mix to start although you wouldn't be able to use white if you're using white paper, you're looking at squeezing another 3 bits into each pattern, for 7 bits total (16 shapes, 8 colors).

That still only gets you about 118 megabits per page of paper.

Now, that's assuming ideal conditions, an unrealistically high printer and scanner accuracy, etc.

The ONLY way you're going to get 450GB per page is if you think you can print in millions of colors accurately enough that a perfectly tuned scanner would read that color back just as accurately, letting you encode each shape and color with far more than the 7 bits I hypothetically mention.

Since 450GB is 3.6 terabits, we're talking about taking those 1 billion dots on the paper and using shapes/colors to squeeze another 3342 times as much info. That's a pretty amazing feat if possible and I don't think this "invention" is anything new...

Someone mentioned a magazine that used to print a bar-code like thing in it's articles that included source code. You could use their special scanner to import that code into your system, saving you all that typing. I think that was PC Magazine or something back in the late 80's or something similar. Cool when it worked, but when the price of CD's/floppies and especially online access came around, the concept wasn't necessary anymore. And you couldn't really encode all that much on a page, but I guess by 1980's standards it may have seemed like a lot.


RE: blah
By bunnyfubbles on 11/25/2006 12:48:06 PM , Rating: 2
Even if it could work, imagine the read/write times...

Then there's durability. With discs you have to worry somewhat about scratches, but with paper, what happens when it is easily distorted/faded by simple sunlight or perhaps is folded or crinkled slightly?


RE: blah
By robbie1687 on 11/25/2006 1:27:34 PM , Rating: 2
quote:
creating those shapes requires probably at least an 8x8 dot area, so you've just reduced the number of "patterns" per page

Exactly. You seem to be the only person here who understands that symbols divide the bit capacity into cunks. They don't increase the capacity. Everybody else seems to think the symbols magically multiply the number of bits. They are making the same mistake as somebody who thinks 32 ounces of gasoline magically turns into 96 ounces when you pour it into the tank, because those ounces can be divided into 4 cups or two pints or one quart, and those units add up to 96 ounces.

quote:
The ONLY way you're going to get 450GB per page is if you think you can print in millions of colors accurately enough

But here you're making that same mistake yourself in a different form (with regard to color instead of shape) because the printer is able to print millions of colors only by combining dozens of spots of ink, and the ink comes in far fewer colors. It's the individual spots of ink and the number of colors of ink that determines the bit content. Combining those things into units (in this case, the combined unit is a pixel) doen't add capacity. It merely divides the existing capacity into a chunk.


RE: blah
By Wwhat on 11/25/2006 8:54:52 PM , Rating: 2
There's one thing not taken into account with your example, you can also use the relative position of symbols to eachother to expand the data, the same way RAR's compress data, as a simple example if you for instance say 'X' represents 'now follows a sequence' and then the next byte says what byte and the next how many times, then you can represent 256 bytes with just 3 bytes of data like 'XA256' equals 256 A's.
Of course that only works for sequences of similar characters in this example, but that might be the joke here, will this work for uncompressible information?
And even when using compressible information the claim of such sizes seems impossible unless you use a new algorithm that could just as well be used to expand the datacapacity of say DVD's.


RE: blah
By oTAL on 11/26/2006 5:43:24 PM , Rating: 2
If that's the case then it incorrectly presented.
You should always give a media true capacity... that would be like seeling a 600GB HD and when you bought it you'd find out it was really 300GBs and you'd have to activate NTFS compression and use only a selected few types of files in order to achieve 600GBs...
I believe this is not the case or this would be a truely disgusting spin on the researcher's part. Storage technology and compression technology are two different (obviously very related) areas and should not be confused or used to confuse.


RE: blah
By ElJefe69 on 11/27/2006 2:07:04 PM , Rating: 1
what the hell is wrong with compression? all data is stored in some sort of compression and read in some sort of compression. nothing is read in nonsensical 1's and 0's and understood as that. HDrive Compression is just a top down after the fact form compressing information. Also, colours and symbols are way beyond that anyways.


RE: blah
By Live on 11/24/2006 7:52:12 PM , Rating: 2
Does it hold 450gb and cost like well a piece of paper?


Possible alternative use...
By Kerdal on 11/24/2006 4:12:26 PM , Rating: 2
If 450gb of data can be stored on a piece of paper and you only need a scanner to read it then a simple picture file could contain all that data too... This could possibly be used as a new compression method. Unless, of course, the resolution needed to read such data is too high resulting in a picture file almost as big as the original data.

I agree stevenplatt, paper isn't a good medium for that much information because the data could be lost so easily : humidity, fire, ink fading with age, even folding the paper could alter the data too.




RE: Possible alternative use...
By Russell on 11/24/2006 4:16:01 PM , Rating: 2
I think the point of it is just to have a cheap storage medium that anybody can carry around or use at any time.

It's also a technology concept demo. They could use anything to store the data. What's critical is this "rainbow code" thing they talk about.

This isn't an original idea, we all know that hexadecimal can store more data than binary with fewer digits, but its an extension of that idea designed to get this dude his graduate degree of some kind.


RE: Possible alternative use...
By Gentleman on 11/24/2006 4:16:01 PM , Rating: 2
This just sounds like a new way of compressing data. ie have these different shape and color combination in a lookup table and use it to encode a few bytes into a shape. Images of sticking cards into a computer comes to mind, oh wait we already past that stage of punch cards..is this the next evolution? I hope not.


RE: Possible alternative use...
By osalcido on 11/24/2006 6:05:31 PM , Rating: 1
carrying around data like this is gonna be a thing of the past... everything is going wireless now and everything will be transferred via internet

beides, this whole thing sounds stupid. so you can store data on a piece of paper... ok... but how do you rewrite it? or erase it? you just have to keep going out and buying more and more paper? retarded


RE: Possible alternative use...
By tungtung on 11/24/2006 6:25:55 PM , Rating: 2
I think the main point of this is just that this method allow the use of cheaper medium to store data. Instead of producing those flash memory chips or CD/DVD (including HD and Blu-Ray) discs, it will theoretically allow the use of regular paper or plastic to store your data.

However I think it is just too impractical for too many reasons. First of all the media is just too frail, and as pointed by several people paper can be easily folded, and easily destroyed, as well unless they can invent inks that will not fade over time or peel off the paper, then data integrity is almost impossible to achieve.

Secondly as pointed by another poster, you will need a very very high resolution scanner, and not to mention printer to actually store the data and then retrieve it again. And not to mention how much processing power will this entail. I mean in the end we're still dealing with a binary system, so this technology will just slow down the data processing.

I think at the end of the day, the impracticality, the lack of data security is just way too much to overcome as opposed to the gain in savings on the medium. This I believe will just become a dead end technology that might never see the light of day in the retail market.


Could that much data be stored on one page?
By GeeWhizBang on 11/25/2006 1:43:05 AM , Rating: 2
10.5 nches x 8.5 inches x 600 dpi x 600 dpi
= 32,130,000 pixels

450 GB x 1,073,741,824 Bytes/GB x 8 bits/Byte
= 3,865,470,566,400 bits

with 24 colors and 4 colors/ symbol:
(24 x 23 x 22 x 21) = 7,650,720 permutations per symbol

7,650,720 x 30 symbols / 36 cell size (pixels)
= 6,828,267,600,000 Encoded bits

(I adjusted the second set of numbers until I got a reasonable redundancy ratio:

1.766477711 Redundancy Ratio

The math looks very good. It even looks like much more than 450 gb could be encoded on a single sheet of paper. The main limitation is how much CPU processing do you want to do to read the data.




By crystal clear on 11/25/2006 3:18:18 AM , Rating: 2
" The main limitation is how much CPU processing do you want to do to read the data."

Response-As quoted in my post-

TYAN-
personal Super computer - PSC concept-release 1Q07.
The T-600 series is the kind of stuff we would using very soon.Massive computing power & speeds.


Unquote-PSC is fast becoming an option for day to day use.
As for your Maths & the post-GOOD JOB.



By Visual on 11/27/2006 6:38:14 AM , Rating: 2
what the beep's wrong with you and your math? none of what you wrote makes sense.


By CascadingDarkness on 11/28/2006 1:55:30 PM , Rating: 2
I'm not sure but I think
quote:
7,650,720 x 30 symbols / 36 cell size (pixels)

would actually be 6,375,600 (assuming you mean 6x6 pixels = cell size for symbol to take up)

Also, it seems unlikely the symbols would be up to/always be 4 colors each. Just guessing from your limited info on this test case, but 450 GB is total bull I would say.


Simple though shows this is BS
By hvdh on 11/25/2006 9:22:46 AM , Rating: 2
Simple thought:
if a piece of paper could store 450GB,
- your print job to print that page must have at least 450GB
- and your scan of that page must have at least 450GB
to contain that information.
This is about storage, so don't throw in compression.

Usually, the print size should be larger because of redundancy for error correction, and you scan should be larger (higher (spatial/color) resolution) to improve correct readout.




RE: Simple though shows this is BS
By Nacho on 11/25/2006 10:11:23 AM , Rating: 2
Just think of this new technology as a bigger DVD-R drive. The drive doesn't have 8+ GB of internal storage, the data is transfered (from DVD to HDD or RAM when reading, from HDD or RAM to DVD when writing).

About the print size: instead of increasing it, you reduce the amount of data available to the end user, just as a DVD holds a lot more than 8+GB of data, but uses some of it for error correction and leaves you with about 8GB.


RE: Simple though shows this is BS
By saratoga on 11/26/2006 3:39:35 PM , Rating: 2
quote:
Simple thought:
if a piece of paper could store 450GB,
- your print job to print that page must have at least 450GB
- and your scan of that page must have at least 450GB
to contain that information.


Congradulations on being smarter then 80% of this thread.

quote:

Just think of this new technology as a bigger DVD-R drive. The drive doesn't have 8+ GB of internal storage, the data is transfered (from DVD to HDD or RAM when reading, from HDD or RAM to DVD when writing).


Better luck next time.


RE: Simple though shows this is BS
By Etsp on 12/1/2006 5:37:52 PM , Rating: 2
Try its compressed into that image BEFORE it gets sent to the print queue...and it is decompressed AFTER it gets scanned... basically like a zip file, or even more accurate, a .7z file. I'm not sure how many people have downloaded a GOOD .7z file, but I've decompressed a <60mB file that ended up being more than 1gB in pictures.... but, I dont know how they did it, I really dont. but its cool to know that it can be done


He Looks Like a Terrorist
By Crazyeyeskillah on 11/25/06, Rating: 0
RE: He Looks Like a Terrorist
By Flunk on 11/25/2006 12:41:23 PM , Rating: 2
Not all Muslims are terrorists nor are all terrorists Muslim. This sort of random stereotyping is highly insulting and will not benefit anyone.

I do however, agree that combineing reglion and science is a terrible idea and that his claims of capacity seem far fetched. Say the standard has 10 shapes and 6 colours (60 combinations). If we are using an 8.5 X 11 sheet of paper we have a total of 93.5 inchs square to work with (assuming no margins at all). 450GB (3865470566400 bits) assuming 60 combinations is still 64424509440 unique shapes that will have to occupy that page. or about 689032187 per square inch. My current inkjet printer has a maximum resolution of 4800x1200 or about 5760000 dots per inch. Now each shape would need to be composed of at least 9 dots (probably more), probably much more. This means you would need a printer that is about 1076 times as accurate as this (I haven't even gotten to ink bleed or scanning this in accurately afterwards).

I don't see any way this could be feasible as a replacement for optical or magnetic storage, the density of symbols required to compete would be too high to accomplish at a lower expense than just using a DVD or hard disk.

This might be useful in developing countries where more complex storage systems are not avaialable or too expensive but these claims of storage capacity are just too overinflated.


RE: He Looks Like a Terrorist
By Wwhat on 11/25/2006 9:04:01 PM , Rating: 2
I think you just exposed yourself, he never mentioned terrorist, that was all in your mind when he mentioned muslims.
It's you that are insulting therefore.


RE: He Looks Like a Terrorist
By Wwhat on 11/25/2006 9:05:51 PM , Rating: 3
Oh, sorry, didn't read the title of his post, nm

I agree though that it sounds like some silly propaganda trying to give the impression they invented something the rest of the world could not and then not having any proof for that claim.


A slightly different turn
By iNGEN on 11/24/2006 4:37:39 PM , Rating: 2
This is one of those "silent thunderclap" technologies people never pay much attention to until it is in full use somewhere. This particular approach may never see widespread use, as there have been numerous attempts to encode data in similar means. So the idea is more evolutionary than revolutionary. If you really want to evaluate this technology simply remove the term paper and replace it with "two dimensional optical media".

The compression analogy is a good one. The bulk of the idea is to separate data density from what it has, throughout the history of modern computing, been inexorably tied to: The physical capacity of the storage medium.




RE: A slightly different turn
By Wwhat on 11/25/2006 9:00:33 PM , Rating: 2
They already used a pattern printed on ID cards that decode to the picture on the ID's, to prevent manipulation or falsification, by scanning the pattern a computer display shows the picture and if it's different from the normal one on the ID you know it's fake.
Same thing as now the RFID does but without using chips.
So indeed the concept as such is not completely new and has been tried before.


RE: A slightly different turn
By Visual on 11/27/2006 6:25:41 AM , Rating: 2
you are so very wrong...
patterns on id cards, or the data in a rfid chip for that matter, usually only represent a database id, and then the actual data like photo or anything is queried from the database.


i have seen something like this before...
By darcmatter on 11/24/2006 5:47:05 PM , Rating: 2
I remember a long time ago where a company had developed a storage method in which an executable as a matter of relevance when an average software installation was a few mbytes in size was encoded into a special barcode that could be printed on paper that when rescanned into another computer with their own software ocr would produce the file in the original size with out any data corruption. I believe the technology was evaluated in a magazine called network world from what I believe 1997-2001. I have been trying to find the article for a while with no avail. if anyone remember who the company was or if they still exist i would like to know thanx.




RE: i have seen something like this before...
By Wwhat on 11/25/2006 9:17:39 PM , Rating: 2
I (and many others) do believe it is theoretically possible to uniquely represent enormous amounts of data with a small value and then using the right algorithm retrieve that data back.
For a while they thought MD5's were always unique (and with enormous effort it is reversible) but then mathematicians proved it wasn't 100% foolproof and theoretically two dissimilar sets of data can have the same MD5, still, there must be an algorithm that makes it possible, problem is finding it and then making it reversible with current computing powers.


By saratoga on 11/26/2006 9:31:36 PM , Rating: 2
quote:
I (and many others) do believe it is theoretically possible to uniquely represent enormous amounts of data with a small value and then using the right algorithm retrieve that data back.


Everyone believes this. What you're talking about was proven by Shannon in the 1940s, before there were even computers. He used books in a library which he opened to random pages. Seriously, what you're talking about has been a huge area of research for 50 years now. Wikipedia him and start reading his papers.

quote:
For a while they thought MD5's were always unique (and with enormous effort it is reversible) but then mathematicians proved it wasn't 100% foolproof and theoretically two dissimilar sets of data can have the same MD5,


No one ever thought MD5 was unique. By definition MD5 was not unique. They thought there was no computationally reasonable way to calculate one of the infinately many possible files that would have a given MD5 sum.

quote:
still, there must be an algorithm that makes it possible, problem is finding it and then making it reversible with current computing powers.


What, reverseing a hash? Theres a universal algorythm for that. Just compute the has of all possible files and see which ones match.


Lack of Vision
By hurricane51 on 11/24/2006 8:16:43 PM , Rating: 2
Wow, I can't believe all the negative comments I'm reading here. Granted, I'm taking all this on the assumption that there's some truth to this technology, but most of the responses have been to compare it to current storage technology and find it's not up to snuff.

This appears to be an entirely new way of looking at storage media and algorithms, and there is simply no way to guess where it might lead. I think it is myopic to immediately categorize it as a replacement for anything. If you just sit down and give a thought to ways in which the current storage technoloigy are deficient, I think you can make some guesses as to how this new tech might me used. Immediately I came up with one: paper vs extremely complex and exotic hard drive platters as the storage medium. Cost? Availability?

Is there anyone else out there who can see beyond their noses?




RE: Lack of Vision
By saratoga on 11/26/2006 9:34:21 PM , Rating: 2
quote:
Wow, I can't believe all the negative comments I'm reading here. Granted, I'm taking all this on the assumption that there's some truth to this technology, but most of the responses have been to compare it to current storage technology and find it's not up to snuff.


Bad assumption.

quote:
This appears to be an entirely new way of looking at storage media and algorithms, and there is simply no way to guess where it might lead. I think it is myopic to immediately categorize it as a replacement for anything.


I'd say it appears to be an entirely new form of poor journalism. Whatever this guy invented, its not what the press release says.


RE: Lack of Vision
By Schrag4 on 11/27/2006 3:08:36 PM , Rating: 2
Ok, whether or not the report is accurate, why is this considered interesting storage news? It's not because of the speed, ability to rewrite, or durability. It must be because claims were made that VERY large amounts of data could be stored on a single sheet of paper.

Ok, so what makes it able to hold 'so much'? I think it's because each piece of info can hold more than just 2 values, unlike current storage, which holds zeros or ones (bits). So, can't we cram something onto a platter that has more than 2 possible states? Why do we have to make a quantum leap backwards to visually representing data on paper?

I believe that's why there is such skepticism going on here.


Beyond BS
By OCedHrt on 11/24/06, Rating: 0
RE: Beyond BS
By AlabamaMan on 11/25/2006 1:31:48 AM , Rating: 2
Whenever someone tells you they found a way to store 450GB in a 100MB bitmap, just slow walk away.


RE: Beyond BS
By JarredWalton on 11/26/06, Rating: 0
RE: Beyond BS
By Visual on 11/27/2006 7:13:10 AM , Rating: 2
what drugs are you on?

17179869184 values = 16GB of information?
since when?
2 pixels, each 4 bytes, or 64 bits of information total, can represent
1.84467441×10^19 possible values. thats way more than this 17179869184 number you come with (from where?), and it's still just 8 bytes. it's not 16GB or anything. get a grip.

try to figure out the difference between number of possible values and number of bits. 256 possible values are still just 8 bits, and 1.84467441×10^19 possible values are still just 64 bits of information, no matter how huge a number it may sound. your 17179869184 possible values are only 34 bits, not 16GB.

-----------
even more bs below the line in your post, but i hope you'll realise where you went wrong.


RE: Beyond BS
By saratoga on 11/26/2006 3:40:45 PM , Rating: 2
quote:
The only true benefit will be in storing certain data types that are limited to smaller subsets of the potential pattern of data. In which case you're really just applying a weak compression before storing.


Why was this rated 0? Its argueably the most relevent post in the entire thread.


Emphasis on 2007 news.
By crystal clear on 11/25/2006 12:40:10 AM , Rating: 2
AS a Venture capitalist-this news could make interesting reading,as you invest in new technologies , concepts & products.
High risk ventures that can go flat on the face or strike oil.It all depends.

DT should concentrate on news items like-

TYAN-
personal Super computer - PSC concept-release 1Q07.
The T-600 series is the kind of stuff we would using very soon.Massive computing power & speeds.

Intel-
In addition to the Core 2 Duo E4300 CPU, which Intel is slated to launch in the first quarter of 2007, the chip giant will add two Core 2 Duo E4000-series CPUs, the E4200 and E4400, in the second quarter of next year, according to sources at Taiwan-based PC makers, who viewed Intel's latest roadmap.

Source-Digi times

I as reader is more interested on product releases,technology,news items that we expect to see & hear in 1&2 Q 07.
Its more realistic & very useful.

So concentrate more on the first,second,& maybe third quater
of 2007.Technology news,products & processes,Hardware & software.





RE: Emphasis on 2007 news.
RE: Emphasis on 2007 news.
By lufoxe on 11/27/2006 11:28:50 AM , Rating: 2
one word comes to mind for this... BURN!


Cool, but I've got a better idea....
By Egglick on 11/24/2006 9:59:18 PM , Rating: 5
It sounds like he just created a new language with many many variables. Each of the many symbols represents a string of binary, so as others have said, it's basically compression technology.

However, I don't quite understand why he chose paper, or why it would even make sense to store the data in that medium. Couldn't you just store several high quality jpegs of these sheets of paper on a memory chip, and effectively have hundreds of terabytes worth of storage??




One Time Use
By Goty on 11/24/2006 6:05:00 PM , Rating: 2
The only problem I can see here is that you'll print something on a sheet of paper, but then that's it, the paper is used up. I know it's extremely cheap and certainly has some valid uses, but it's really not valid replacement for any available technology today.




RE: One Time Use
By Nacho on 11/25/2006 10:40:54 AM , Rating: 1
Just like a DVD-R!


RE: One Time Use
By ElJefe69 on 11/27/2006 2:10:56 PM , Rating: 1
er, do you realize how toxic and terrible the production of any form of plastic is? the factory to make it, the building of that, the crude oil for the plastic, waste products, etc etc, all the things people do not focus on besides the disposal end.

Take a new prius hybrid. Good for the environment? hell no. The amount of waste it produced and energy wasted developing and making it far exceed the environmental benefits. Get a 1990 toyota and fix it, youll be actually helping the environment. lots of ideas are backward and consumer focused. polution and inefficiency are from the manufacturers and developers.


DT's summary needs two words added.
By ninjit on 11/25/2006 12:39:46 PM , Rating: 3
I just read the linked article, and DT fails to mention that this student, and his engineering school, are in India.

Yes, it was reported by Arab news and his college is Muslim specific, but that's no different than having Christian colleges in the US.

Most people reading the DT summary would automatically presume that this news is from the UAE, Saudia Arabia, or some other middle eastern country.

Simply adding "in India" after the college name would help alleviate any confusion.




RE: DT's summary needs two words added.
By Flunk on 11/25/06, Rating: 0
By Suomynona on 11/25/2006 8:15:46 PM , Rating: 1
This coming from a moron writing "a oxymoron" (sic)...


Expanding the idea
By 7DrFunk7 on 11/27/2006 10:01:33 AM , Rating: 2
Sure, this is the start of a great idea. Think of it, companies being able to trasfer data in mass storage on a piece of paper. But the problem is that paper is not very reliable for storage considering how fragile it is.

I think it would be more reasonable if a program was created that allowed data to be converted to his type code and then put on a CD or DVD. Then once data is transfered it can be convereted back to its original type of data.

This is the same idea of compressing files but in a more effective manner.

Also the code he is using is the same idea of quantaum computers except a quantaum proccessor use numbers like 1,2,3,4,5,6 to convert data instead of colors and shapes.




RE: Expanding the idea
By jwilson02 on 11/27/2006 8:54:22 PM , Rating: 2
ok lets say tht a circle repesents 0101, a square represents 0110, a square in a circle represents 0011, a red square is 1001, a red circle is 1100, a red circle in a blue square is 0011.. ok so wht is going on here? a circle with 1 color represents a half byte, a circle with 6 colors gives 6 bytes. and samewith a square. Now, circle in a square with six colors give 6 new bytes for one color if both same color, so 2^6 combination of colors and 6x that. so 384 new bytes right there. now that is just two shapes...if three shapes involved then 3^6 just for colors not mentioning orientation of objects... for instance circle in squre in triangle or triange in circe in square...get my drift....i think ai software algorithms would fit nicely in this type of software compression


Punch cards, here we come!
By darkfoon on 11/27/2006 8:59:07 PM , Rating: 2
My Advanced Operating System textbook says "phylogeny recapitulates ontogeny" which basically means that as computers evolve, the end up going through the same stages again, just at a more advanced level.

I can see it now, in a few years from now we'll use business cards to store huge amounts of data, like the days of punch cards.
Or maybe not.
Either way, this could be really cool. Instead of burning CDs, or DVDs, we could print off a sheet of paper with our data on it, go to a friend's house and read the data there.
The technology, however, needs some serious error-checking/correcting power. Printers aren't perfect, neither are scanners.
But this could be really cool. Like the poor man's CD-R.




RE: Punch cards, here we come!
By Spivonious on 12/1/2006 1:18:42 PM , Rating: 2
The poor man's CD-R....Let's see... $15 for a CD Burner, $0.01 for a CD-R...


They are just confused . . .
By SurreDeth on 11/28/2006 10:16:24 PM , Rating: 2
. . . they've never used toilet paper before.




RE: They are just confused . . .
By lacay on 11/29/2006 4:58:59 AM , Rating: 2
costs:
sheet of paper : 2c
reader/writer (could be bulky) : $2000
total costs: $2000.02

heh, not so practical :)


Trees being cut down fast enough already
By tigz1218 on 11/27/2006 1:24:58 AM , Rating: 1
Im not a stupid tree hugger so dont flame me please. But seriously with all these environmental issues today I really dont think using paper would be a good idea. Think about all the CDs and DVDs now if all of these were paper I dont think think this would be good....just my thought but Im prolly wrong and going to get flamed by all of you.




By Live on 11/27/2006 4:27:26 AM , Rating: 2
Well trees being cut down are not really the problem. The problem is that they are not replanted at the same rate as they are being used. Trees are a renewable resources as long as we get proper laws in place that requires that you plant enough trees to replace the ones you use.

Oil on the other hand is not renewable in the way we use it today. Oil is what is used to make CD/DVD so they are much more harmful to the environment then paper.

If you read the article you would also learn that this tech doesn’t involve paper per se. Its printing that is required. But you can print on many materials like plastic, metal etc if you think its better.


one concern assuming it works
By lufoxe on 11/27/2006 11:24:00 AM , Rating: 2
This technology would be awesome if it works, the only question I have is, just like all paper, what happens when it fades, or turns yellow? Your data won't be the same, if legible at all. How about the issue with he actual scanners themselves. I know I've scanned pictures that have come out more blue or red then they are supposed to be. While I may be able to fix it, I bet the attorney's or doctor's office that is using it won't. (although it could be a new form of data recovery? haha) As I said before, an interesting technology, assuming it works.




RE: one concern assuming it works
By ElJefe69 on 11/27/2006 2:17:43 PM , Rating: 1
that clear block thing that they use on star trek is this thing right here. all one needs is to print in layers on top of layers with some sort of xray or laser images and it is read in those layers.

all of you are stupid. I am just ignorant. that is more certain than this guy being a hoax.


Is paper really worthless?
By Don Tonino on 11/24/2006 6:44:17 PM , Rating: 2
Concerning the worthy worries about paper durability, I think it's well to remember that parchment followed by paper have been capable to survive hundreds of years while still being fully capable to convey the information 'encrypted' on them.

True, here we are talking about a totally different order of magnitude of storage required, but still I doubt any modern storage medium has a useful and reliable lifespan that would allow it to legally enjoy a beer.




Hmmm..
By mintbiskit on 11/24/2006 10:14:32 PM , Rating: 2
I thought computers were supposed to help us stop using paper as a storage medium.




By holoman on 11/24/2006 11:56:33 PM , Rating: 2
here's something to sink your teeth into.

http://colossalstorage.net





Export to PDF?
By gmyx on 11/25/2006 10:25:05 AM , Rating: 2
Hum, I wonder what RIAA would think of this. You could just output it to a PNG or a PDF and "scan" it in and decode it. The image would be much smaller than the 450GB on the paper. Even at high resolution, this seems very interesting.




Cauzin Softstrip
By peternelson on 11/25/2006 11:32:14 AM , Rating: 2
In the 1980s, a system called Cauzin softstrip worked like this ie like 2D barcodes.

Computer mags used it for listings and binaries.

Adding colour will increase capacity.

Where this could take off is in mass produced commercially printed material, not home printing.

It's likely capacity estimates exceed those realisable in a practical system.




And you stil haven't seen it all cause...
By Kyanzes on 11/26/2006 5:01:56 AM , Rating: 1
...from what I heard it also supports RAID. :D




By ElJefe69 on 11/27/2006 2:15:03 PM , Rating: 1
hawt


ummm...
By L1NUXownz1fUR1337 on 11/27/2006 9:58:10 AM , Rating: 2
I have a bridge for sale - in new york. Real cheap.





Seems flimsy...
By Rock Hydra on 11/28/2006 3:50:05 PM , Rating: 2
I do not think this is a very suitable solution. A slight rip, that data is unusable or corrupted. Drop it or someoen steps on it, data ruined, gets water spilled on it, the ink bleeds and is destroyed. A crease or fold will probably cause slight shadows altering the color, causing incorrect data to be read. Also, paper is organic and has a shelf life and will yellow over time, ruing the data. I'm also assuming this would be read-only instead of rewritable. Also, with the cost of ink these days...Yeesh...Don't get me wrong, I think it's cool....just not practical.




I too think this is bogus
By s12033722 on 11/28/2006 8:13:30 PM , Rating: 2
Let's assume they were talking about an 11x17 sheet of paper. Let's further assume that they can deposit pixels at 2000 dpi. That gives 748 million pixels assuming 100% coverage. Now, lets assume that they can deposit color with 12-bit accuracy and read at the same accuracy - that's 12 bits per color, for 48-bit color, which is about the limit of possibility with real equipment (that is to say, each independant color can be measured to a signal to noise ratio of 4096 to 1, or 72 dB). That gives 36 bits of information per pixel, 12 each from R, G, and B. 748 million pixels times 36 bits = ~ 27 gigabits, or 3.3ish GB.




No really new technology...
By Insurgence on 11/29/2006 8:52:29 PM , Rating: 2
Just a new way to do it. For example:
http://www.thereviewzone.com/gameboy/gb-ereader.ht...





Sure
By DigitalFreak on 11/24/06, Rating: -1
RE: Sure
By Gunter on 11/24/06, Rating: -1
RE: Sure
By ponytrack on 11/25/2006 11:07:17 AM , Rating: 5
Think about it like this, in binary there are only two characters 0 and 1, but if you convert that using software to shapes, and those shapes in different colors, if you use three shapes and 6 colors, already you have 18 characters plus a blank character. Then compare 2^10 (binary) and 19^10 (rainbow) and thats how many bits of information each would have with equivalent space.


RE: Sure
By PrinceGaz on 11/25/06, Rating: 0
RE: Sure
By Shining Arcanine on 11/25/2006 12:18:20 PM , Rating: 2
PrinceGaz, your calculations are incorrect. The amount of data any amount of bits can hold is 2^n. For example, with two bits, you can have 4 different combinations of data (i.e. 00, 01, 10, 11) and 2^2 = 4. Your 2 x 10^n formula predicts 200 different pieces of information for two bits, which is wrong as there are only four different ways that you can utilize two bits.

Can you find any other combinations that two bits can form other than 00, 01, 10 and 11? If you cannot, then it stands to reason that the formula for a base-19 encoding scheme would be 19^n.


RE: Sure
By robbie1687 on 11/25/2006 12:25:38 PM , Rating: 2
quote:
In fact I fail to see how storing the information as coloured shapes could possibly have a higher data-density than simply using coloured dots...

You're right, it doesn't. The data capacity is the number of bits. The fact that the bits can be viewed in different ways (as composing symbols) is irrelevant.

Think of it this way. Eight bits can be permuted in 256 ways. That's why a byte (which is made of 8 bits) has 256 possible values.

Now we can think of a byte as a hex number. Or we can think of a byte as a color. Or we can think of a byte as a code for "circle", "square", "triangle", etc. But no matter how we think of that byte, it still contains only 8 bits, and it still can encode only 256 possibilities.


RE: Sure
By aGreenAgent on 11/29/06, Rating: 0
RE: Sure
By player x on 11/25/06, Rating: 0
RE: Sure
By 9nails on 11/27/2006 11:35:44 PM , Rating: 2
This logic has one flaw - mechanically, you can't print on the edge of the paper where you hold the most data. Also, printers don't print single shades of colors, they print in single color dots that "look" like a shade of colors if they are gapped on white paper. This means that you have to calculate the color for a larger region of paper, lowering your data density. Perhaps paper or laser printing will never be the solution to data storage.


RE: Sure
By Alpha4 on 12/1/2006 2:51:48 AM , Rating: 2
I do not think the first two posts deserve to be shunned so quickly.

I'm sure Digital Freak & Gunter both understand that on a theoretical level that one could conceivably fabricate a storage medium like this which can store 450GB of data, but based on how little the technology has progressed according to this article, 450GB seems to be an extremely unrealistic projection.

I choose to think of the inventor's over-eager prediction as tasteless hype.


RE: Sure
By Russell on 11/24/2006 4:12:09 PM , Rating: 2
If you actually read how it works you'd have seen that it makes perfect sense.

I'm not saying it's practical, but it sounds pretty possible to me.


RE: Sure
By TomZ on 11/24/06, Rating: 0
RE: Sure
By Russell on 11/24/06, Rating: -1
RE: Sure
By Shining Arcanine on 11/25/2006 12:24:06 PM , Rating: 2
If you were to have 1 cm^2 blocks on a piece of paper, with each one either being there or not, each one would represent a bit. Now, imagine the same encoding scheme, except with different colors and shapes either being there or not being there.

With this scheme, you can theoretically store an infinitely greater amount of information in the same area that you would have a single bit, but as the data density rises, reading it becomes more and more complex, and eventually, it becomes absurdly difficult to discern between one symbol and another, especially as the wavelengths of each color rises and the number of sides on the polygons increase. However, it is possible. The Chinese have hundreds of thousands of different characters and they are able to read them without much a problem, although learning to use them is extremely difficult.

If this scheme is adopted for data storage, it will only be used for short term read-only data storage, as the colors will fade with time and it is impossible to rewrite to such a medium.


RE: Sure
By Jaylllo on 11/25/2006 2:42:43 PM , Rating: 2
It's kinda funny. This idea seems like a beefed up cellular automata.

I really wonder if it is even feasible in terms of speed. CA is considered a "junk"/"turing tarpit"
Any experts on Theory of Computation here?


Also, Chinese doens't have hundreds of thousands of characters. That's nonsense.

Of meaningful words ~20,000.
Actually used ~12,000
Academic level is supposedly 7-9,000
Literacy is about 3,000.

Learning any language takes a lot of time. If you think Chinese is "hard" try Hindi... Anyway, Linguists say all languages are equal in terms of information exchange. Chinese doesn't have tenses, English doesn't incorporate mood into conjugation unlike Spanish.

Different paradigms, different trade offs. All the same.


RE: Sure
By Daigain on 11/25/2006 6:15:32 PM , Rating: 2
Actully there was some guy who wasted his whole life writing down all the Chinese symbols that he could find, If i remember correctly he got up to about 250.000(twohundredandfiftytousand) symbols.


RE: Sure
By AnnihilatorX on 11/26/2006 8:18:24 AM , Rating: 2
Jaylllo I stronly believe the figure you quoted is the number of Chinese characters used by Japanese (as Kanji known by them)

7-9000 is number of kanji requried to be learnt in the Japanese syllabus.
Of course the Japanese uses a mix of kanji and Hiragana (and Katakana) in thier writing system. Chinese however consists entirely of characters. We would therefore use much more in everyday usage


RE: Sure
By Daigain on 11/26/2006 6:33:41 PM , Rating: 2
Japanese people learn about 2000 Kanji's in School. And Chinese people about 5000.

But you cant just look at the numbers japanese Kanji's often have more then one way to be read.


RE: Sure
By rushfan2006 on 12/1/2006 12:39:27 PM , Rating: 2
Couldn't agree more on this one.

I think its bunk....not to mention I'm still trying to think of the value of it. Like even for argument's sake pretending you really CAN store 450 gigs on a plain piece of paper...the ONLY advantage I can see off the top of my head would be cost savings (because paper is pretty cheap).

Also I'm keeping in mind other technologies in the works...damn I wish I remembered the specifics of the one that really interests me -- there is one group (I think it may be MIT even) that is aiming to store something like 1 TERAbyte in a space no bigger than a postage stamp.

I'm much more interested in that than this paper stuff....

Paper after all is very very easily damaged, torn, ripped, ruined, folded, etc. etc.

Anyway it doesn't much matter becuase I think its BS.


RE: Sure
By feelingshorter on 11/24/2006 5:57:39 PM , Rating: 2
I think this is BS also. Now that i think about it, you can only compress an imagine so far. To be able to put 450gigs or 95 full 4.7 gig dvds on a piece of paper is laughable. Lets say you are storing 95 full 4.7 gig movie DVDs. I can only imagine that it is impossible to find a pattern to which you can create a logarithm to compress 450 gigs on a single piece of paper. Sure this is taking into account that the shapes will also be colored, but DVD movies are also in color. To say that you can store 450 gigs worth of...say movies... on a piece of paper...

I'm just saying 450gigs on a regular sheet of paper is BS. Maby somewhere along the lines of 10-40 gigs more like it. For 450 gigs to happen, it will be a special factory made "paper size" material that wont be made of paper. In the future of course. Even then, you can only compress an image file so much.

What if it was 450 gigs worth of JPEG files? To say that you are going to use shapes/colors to replicate the image in a digital format on a piece of paper....hell you mind as well just print the image out. Maby someone has a 450 gig picture that they print out on a piece of paper. Hows that for 450 gigs worth of information?


RE: Sure
By psychobriggsy on 11/25/2006 7:51:54 AM , Rating: 2
Whilst I agree that the 450GB aspect is nonsense right now, and even 1GB seems preposterous (and why use large 'shapes' instead of small shapes for binary representation?), I take issue with your claim that printing the image would be better, especially in the case of video.

JPEG and digital video (e.g., H264) are heavily compressed to fractions of their decompressed size.

How much? Lets take a 720P 10mbps video stream. A single frame (1280x720, 24-bit colour) takes up 2700KB of data. A second (60fps) is therefore 158MB of data, or 1266mbps. Yet the stream is 10mbps! The compressed file is under 1% the size of the decompressed file.

Quite clearly storing the compressed file is better than printing out the images from each frame of the video, even if the encoding used is bulkier.


RE: Sure
By rmaharaj on 11/24/06, Rating: 0
RE: Sure
By Lazarus Dark on 11/24/2006 7:57:39 PM , Rating: 4
Though I don't see paper as the way to go here due to durability issues, I long ago thought that holographic storage would benefit from using a type of visual language like the shapes described here to store larger amounts of data in a smaller space. dammit, I knew I shouldof patented that.


RE: Sure
By robbie1687 on 11/25/2006 11:27:02 AM , Rating: 2
Two mistakes keep cropping up in this thread. First, the fact that dots are combined into symbols has nothing to do with the amount of information that can be encoded. Second, this "invention" has nothing to do with compression. It has to do with storage capacity. (One way of thinking about the second issue is to assume the data was previously compressed as much as possible.)

As with all digital data, the amount that can be stored by this technology can be measured in bits. This is independent of whether the bits are arranged in symbols or hex numbers or anything else. In this case the maximum number of bits is the number of dots that can be both printed AND read accurately by the scanner, multiplied by the number of possible colors for each dot, minus some factor that has to be reserved for error correction.

To estimate the maximum amount of data that could be encoded, let's assume (very liberally) that the maximum number of dots per inch that can be printed AND read accurately by the scanner is 4000 dots per inch. (This is an extremely generous estimate because even though scanners can measure 4000 points per inch, the points won't usually line up with the dots). And let's assume that the printer can print a dot in 6 possible colors (including white as a color). And let's assume that the printed area is 8 " x 10.5". Then the maximum number of bits on the page (aside from error correction) is:

16 mllion bits per square inch x 84 square inches x 6 colors = 8.065 billion bits = 1 billion bytes

So the maximum data capacity, even before we subtract a chunk for error correction, is less than one gigabyte.

But in fact the real capacity would be less because the scanner won't be in alignment with the dots. Think of a the read/write head of a hard disk. It prints exactly like a printer, except that the dots have only two colors (magnetized and de-magnetized). And it reads dots just like the scanner. Except the hard disk has a single read/write head that is designed to stay in alignment with the dots on the disk. The scanner will not be in alignment with the dots on the page. Therefore the actual density of dots that can be successfuly read by the scanner will be lower than what I just assumed in the calculation.


RE: Sure
By vorgusa on 11/27/2006 10:32:40 AM , Rating: 2
Yeah they would have to make a bigger dot to guarantee that the scanner would be able to read every bit of information correctly, and this is assuming that once you get to the size that is required to get 450Gb you would probably start having problems with distortion of the paper, since it obviously will start getting bumpy or porous zoomed in that much. and I would assume that touching the paper in any way would probably ruin the information with the oils on your skin. I would assume that there would be better ways of storing information then on paper


RE: Sure
By Motley on 11/28/2006 5:10:56 AM , Rating: 3
The math is fine.

First, notice he said you needed a scanner. He also says he'd need to develop a rainbow printer as well, you can't just use off the shelf $80 ink jets. The article, if you read it... He stored a 45-second video clip... And in the FUTURE, using the tech you may be able to store between 90 and 450GB of data.

Given a standard piece of paper (Of which the article doesn't even say it's that small), but let's just start with that. Today's scanners can do 48+ bit color scanning, and over 4000dpi.

8.5*11*48*4000*4000=71,808,000,000 bits of information. Or almost 9GB. Toss in some compression (ZIP-type lossless) which CAN compress some data to 1% of it's original size, and you have a theoretical data storage capacity of 900GB on a page. Add some error correction, and some fancy way of allowing the system to detect minor skews, and it's possible to store 450GB of data on a piece of paper TODAY.

Of course that 1:100 compression ration with ZIP-type compression isn't common, but that's not what the article said. In fact the guy didn't even claim he could do 450GB today. He said he did a 45-second video clip. I can store a 45-second video clip in my phone, and it'll take all of about 100k. It looks like total crap, but it is 45-second of video.

Not really BS, just not very useful.


RE: Sure
By s12033722 on 11/28/2006 8:19:50 PM , Rating: 2
What you are missing is that those "48+ bit" scanners are DIGITIZING to 48 bits, but do not in fact have 48 bits of meaningful information. The signal to noise ratio of those scanners is typically around 512 to 1 (or 8 bits per color for true 32 bit color) for a good unit.


RE: Sure
By poohbear on 12/1/2006 1:21:39 PM , Rating: 2
"the person saying it cant be done shouldnt be interupting the person doing it."


"Young lady, in this house we obey the laws of thermodynamics!" -- Homer Simpson

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki