Slashdot Mirror


The Disposable Computer

sp00 writes "A disposable paperboard computer has been developed and is already in use in Sweden. Developed by Cypak AB, the paperboard computer can collect, process, and exchange several pages of encrypted data, the company says." Pretty impressive, given that they say it has a mere 32K of memory.

15 of 358 comments (clear)

  1. Article Text by Anonymous Coward · · Score: 5, Informative
    It Had To Happen: The Disposable Computer
    March 4, 2004 (11:40 a.m. EST)
    By W. David Gardner, TechWeb News

    A disposable paperboard computer has been developed and is already in use in Sweden. Developed by Cypak AB, the paperboard computer can collect, process, and exchange several pages of encrypted data, the company says.

    "Initially, it will be used in industrial-specific applications as an enhanced and secure RFID device," said Cypak marketing director Strina Ehrensvard in an email. "Today, in pharmaceutical and courier packaging as a data-collection device; tomorrow maybe for interactive books, lotteries, passports, and voting cards."

    With just 32 Kbytes of memory, the paperboard computer's functionality is somewhat limited at present, but the firm believes its future will be broad. Cypak has entered into an agreement in the U.S. with MeadWestvaco Healthcare Packaging, which has marketing rights to the product and technology in the Americas.

    Ehrensvard said the device is currently in use in a trial sponsored by a Swedish university involving compliance monitoring of pharmaceutical packaging. The trial tracks when a medicine tablet has been taken out of a package; it is then placed on a Cypak scanner connected to a PC on which the information can be viewed and stored. Ehrensvard said the paperboard computer is being considered in another healthcare application, as well: doctors would use it to help authenticate the administration of pharmaceuticals.

    The Cypak product utilizes RFID technology that is based on printable sensors and electronic modules. The components are integrated on a variety of products, ranging from packaging and plastic cards to adhesives. In healthcare applications, Cypak says the paperboard computer time-stamps medicine dosages, which can be integrated with a patient's electronic diary. It can deliver sound reminders, too.

    Cypak has also developed a companion device--a smart card with an integrated numerical keypad. The firm expects this to be used initially in applications demanding high security. By entering a unique PIN on a card, a user can connect to the Internet and exchange data. Cypak says the card's encryption can't be copied or broken, enabling it to deliver "military-class security."

    "The paperboard computer concept and the PIN-on-Card are the same core technology--components integrated in different products," Ehrensvard said. "They exchange information to a PC with the same reader."

    Cypak offers the components on an OEM basis for about $1 each. The firm added that OEM components for its readers are available for approximately the same price in large volumes.

    The firm has developed a tamper-proof package technology with the Swedish Postal Service. Called SecurePak, the packaging technology stores sender and receiver relevant data and alerts receivers of any possible package tampering before the package is opened.

    Cypak will demo the products at the CeBIT 2004 exposition, in Hannover, Germany, later this month.

  2. Chip specs by nmoog · · Score: 5, Informative
    Theres some intersting specs about its S2C chip on the Cypak website:
    8-bit microprocessor
    32kBytes non-volatile FLASH memory with >10 years of data retention.
    2kBytes Static RAM
    It also mentions using 128bit AES encryption, with RSA under development.
    1. Re:Chip specs by Kris_J · · Score: 4, Informative
      That's quite similar to the Atari 7800, if anyone wants any idea of how complicated the software can be for it.

      I wonder how many you can talk to simultaneously if they're all in a pile. What's the RAM footprint for the D.Net RSA core?

    2. Re:Chip specs by Anonymous Coward · · Score: 1, Informative

      I've written both AES and RC5 in Z80 assembly, AES needing about 1K of ROMable space for both encrypt/decrypt halves (significally less then 2K if both are needed), RC5 being about 30% larger (but symmetric) and magnitudes slower. RAM needs are 35/60 bytes respectively.
      Those footprints are near the lower edge of size and performance, allowing more code and data makes both algorithms a lot faster. RC5 is still too slow for 5MHz Z80. Rijndael is the way to go with low-performace processors.

  3. Re:Those swedes, whats the deal? by twaltari · · Score: 2, Informative

    Ikea is a Swedish company but I guess most of their stuff was manufactured in China instead of Sweden. Ikea has pretty much the same business model as H&M:
    1. Design good looking stuff
    2. Market it (e.g. printed catalog with lots of great photos)
    3. Find a Chinese subcontractor to manufacture it as cheap as possible.
    4. Profit!

    So perhaps that explains the quality issues. But on the other hand, they never promised you high quality and it was propably quite affordable.

  4. Re:32k is more than I dreamed of a few (25) years by Anonymous Coward · · Score: 1, Informative

    "32K is a lot of RAM. It's enough to do a fairly useful voice recognition system, or word processor, or even spreadsheet."

    WordStar takes 8k...

    Elite, one of the best Sci-Fi computer games ever, was developed on a 32K computer.

  5. Re:It was Steve Jobs by seanadams.com · · Score: 3, Informative

    It was computer design by temper tantrum.

    That may well be, but it's still perhaps one of the smartest moves apple ever made. While the rest of the computing world focused on standardized components, interchangeable io cards and memory, replaceable drives etc... Apple made an all-in-one system whose only interesting upgrade path was a new machine from Apple.

    And so, those who valued a complete, tested, supported system went with Apple, and that was the basis for (one of?) the most successful hardware companies ever.

    You can't argue with success, even if Jobs is a bit of a whack.

  6. Youngsters have it too easy by Anonymous Coward · · Score: 2, Informative

    The first computer I used had an architectural limit of 8192 39bit words (none of this power of two rubbish), and the fastest instruction took 576ms (2.4kips). Somebody heroically modified it to act as a timesharing system for 3 terminals.

    The first computer I built had a massive 128 bytes; never did manage to fill it up with anything useful.

  7. I looked over cypaks site by Anonymous Coward · · Score: 2, Informative

    No new technology seems to be involved, some comments would lead one to believe that there was some inkjet printed circuitry involved and that it is a paper computer. Not so, I could see no evidence of this.

    The pin on card product looks like it has some applications.
    To order the development kit which includes a usb reader and two cards was 500 pounds(the money sort) the license to get this kit did not allow any reverse engineering or disassembly so it would not seem well suited to determining just how secure it is, they also reserved the right to take the equipment back. Postage and handling was another 100 pounds.

    I use an RSA secureID it has a keypad and numeric lcd, this does everything except the contactless is achieved by reading the correct number off the lcd.

    One of the goals of having a keypad on the card is to prevent the pin going through untrusted chanels and for achieving this I say bravo! every access card should have this feature, however combine to a wirless power and comms means you have to put it down on a reader surface.

    Ignoring cameras I still wonder how easy it would be to steal the pin by sensing the pressure through the card as it is used. Of course the card would still need to be stolen. Mind you if the reader is untrusted then how can you even be sure what your authorising?

    The package dispensing is a fancy birthday card as someone else mentioned, the pin on card is only good to tell a remote and trusted web site that yes it really is you and I'd still trust my RSA SecureID over this.

    What is truely needed is a card or cheap pda (cost similar to card) that has a keypad and a display big enough to display an incoming request.
    This can then be aurthorised (signed) and the response sent. Contactless is important for security but enough distance so that emissions and key pressure do not factor in. IR beaming is probably the best, keeping it low power and rechargable should not be insurmountable problems.

    Once something like that is built then untraceable digital cash and communications would be instantly realizable. Once you have secure comms and finance then all other freedoms are possible.

    If you wish to help build such a device or would be a potential user of it then by all means contact me.

    -----BEGIN PGP PUBLIC KEY BLOCK-----
    Version: PGPfreeware 6.5.8 for non-commercial use

    mQCNA0BH2CQAAAEEAOPWa1ggC5SxV+CSOJUF0u47ZjRr7RfY XN DuXjmUV8L4csZf
    N5ks+ALXtga4GyWuEyUWFiGvik/jCdQTQu f23F4H6t48GNwxex SEvRGke8favAC0
    BMv+crgfmYF0//yAZLZnaemyTWttBvqfuV JJP5dGsm3o86FXxB kGobiHr/3FAAUR
    tCNZb3VzZWYgQWhtZWQgPHlvc2VmYWhtZW RAeWFob28uY29tPo kAlQMFEEBH2CQZ
    BqG4h6/9xQEBnJcD/1dFH+HIQouQR9VY+e MxN16fSQGNSFZ/hA eedRDP9rHMqpGX
    7siNbJ66Mecc1XQCtNnZpKiKUsGZ4Psosh Bvu4wgIk2OQzubGM LVIjHcV0kRQiqj
    xg3ZIIu04RMaROpvSYc88dygnjjxMciZtL xod4kZHCDxFizUkh bs6mudzzmN
    =iTJe
    -----END PGP PUBLIC KEY BLOCK-----

  8. Re:32K?! by Urkki · · Score: 4, Informative
    • i was quite pissed off when he said theres no reason to use byte, short or int anymore since all machines have memory to spare we should store every variable as a long.

    Well... For most current computers, long takes as much space as int, 32 bits (or 64 bits for 64 bit CPUs). And for any individual variable, it'll take at least that amount of memory even if it's 1 bit bitfield (not sure if common x86 compilers by default pack variables to take less space if they can, though).

    So using anything else doesn't really gain you anything in space, but it may well hurt you in peformance (and code re-use).

    Bloat comes from copy-paste coding, and slowness comes from incorrectly chosen data structures and algorithms. This all arises from poor design, so for example the programmer is often faced with decision to change entire existing code to make some piece of it more generic, or copy-pasting that piece and modifying it a bit to fill the new need.
  9. Re:32K?! by zero_offset · · Score: 4, Informative

    Speaking as a programmer whose first programming experience was Z80 assembly with 4K of RAM, I had to seriously revamp my thinking in recent years, but basically 32-bit processors work significantly better with 32-bit values. Also, except when memory is statically allocated and you've forced the compiler to pack memory allocated to your variables, that carefully considered allocation of a single byte is going to take up 32 bits of actual memory anyway, because once again, the CPU can access it that way significantly faster.

    In a nutshell, if you use a byte, short, or int, you're slowing down the machine. The only good reason to optimize to that level is when you're expecting memory problems, and as your professor pointed out, that is rarely a consideration these days (especially when you add modern paging techniques to the vast amounts of available physical memory).

    The more interesting question is the chicken-and-egg angle on runtime bloat -- did RAM increase because users needed it to run steadily ballooning applications, or did applications get bloated because cheap-as-dirt RAM meant end user machines had memory to spare? (I suppose both are probably true... which would be a bizarre conclusion if we were actually talking about chickens and eggs...)

    --

    Slashdot quality declines as the number of hot grits posts decreases. - Provolt's Law, Apr-09-2005

  10. Re:It was Steve Jobs by jsebrech · · Score: 2, Informative

    You might want to read up on your history yourself. Jobs took over the mac project when he was kicked out of the lisa project, and Raskin quit over what Jobs did to it.

    The problem with Jobs is that he always wanted perfection, and perfection is too expensive. The lisa, the mac, next, all suffered from overdesign and the resulting price tag.

  11. Re:I need this by zero_offset · · Score: 2, Informative

    If America wasn't ass-backwards, I'd just SMS the stuff to my cell phone.

    Give up your ass-backwards ways. I'm in Florida and I send text to my cell phone all day long. Some of my servers send activity reports every few hours. I also threw together a simple web page to help friends and family members send messages to my phone. Easy.

    --

    Slashdot quality declines as the number of hot grits posts decreases. - Provolt's Law, Apr-09-2005

  12. Zenith tried this sort of thing once.... Bad idea by Warlock7 · · Score: 2, Informative

    Zenith tried using paper/cardboard circuit boards in their televisions once and it was a complete failure. It seems that many of the televisions arrived at stores and simply would not work out of the box. Turns out that the vibrations created during shipping was enough to break the circuit boards and forced them to recall the sets and replace the boards with ones that weren't as flimsy.

  13. Old news... by MrScience · · Score: 2, Informative

    I wonder if this guy will be remembered? He tried to do it in the US ten years ago.

    --

    You quitting proves that the karma kap worked. The most annoying of the whores shut up. --CmdrTaco